Developing a Theory of Planned Behavior Questionnaire for Reproductive Health: A Comprehensive Guide for Researchers and Clinicians

Abstract

This article provides a comprehensive framework for developing, validating, and applying Theory of Planned Behavior (TPB) questionnaires in reproductive health research. Tailored for researchers, scientists, and drug development professionals, it covers the foundational principles of TPB, detailed methodological steps for instrument design, strategies to overcome common measurement challenges, and rigorous validation techniques. By synthesizing evidence from recent global studies, this guide aims to empower the creation of robust, theory-driven tools that can accurately predict health intentions and behaviors, ultimately strengthening the evidence base for reproductive health interventions and clinical trials.

Theoretical Foundations: Applying the Theory of Planned Behavior to Reproductive Health

The Theory of Planned Behavior (TPB) is a psychological framework that links beliefs to behavior, positing that three core components—attitude, subjective norms, and perceived behavioral control—collectively shape an individual's behavioral intentions, which are the most immediate antecedents to actual behavior [1]. Originally developed by Icek Ajzen in 1985 as an extension of the Theory of Reasoned Action (TRA), the TPB incorporates perceived behavioral control to improve the prediction of behaviors where individuals lack complete volitional control [1] [2]. This framework has been extensively applied across various domains, including public health, environmental science, and notably, reproductive health research, where understanding and influencing health-related behaviors is crucial [2].

In reproductive health, the TPB provides a systematic framework for designing evidence-based interventions and measurement tools. It helps researchers and practitioners understand the cognitive determinants behind health behaviors, enabling the development of targeted strategies to improve outcomes such as preconception care utilization, endometriosis management, and sexual health practices among adolescents [3] [4] [5]. The theory's structured approach to quantifying behavioral antecedents makes it particularly valuable for creating reliable questionnaires and evaluating intervention effectiveness in this sensitive domain.

The Core Constructs of TPB

Attitude Toward the Behavior

Attitude refers to an individual's positive or negative evaluation of performing a particular behavior [1]. It is determined by behavioral beliefs—the subjective probability that performing the behavior will lead to specific outcomes—and the evaluations of those outcomes [1]. In reproductive health contexts, attitudes often encompass beliefs about the benefits and drawbacks of health-seeking behaviors, such as the perceived effectiveness of preconception care in improving birth outcomes or the anticipated emotional and physical consequences of discussing sexual health with a partner or provider [4] [5].

• Theoretical Foundation: According to the TPB, attitudes are formed through an expectancy-value model, where the strength of each behavioral belief (bi) is weighted by the evaluation (ei) of the associated outcome [1]. The aggregate of these products determines the overall attitude: A ∝ Σbi ei.
• Measurement in Reproductive Health: Questionnaires typically assess attitudes using Likert-scale items that tap into both cognitive (e.g., "I believe preconception care is effective for preventing pregnancy complications") and affective (e.g., "I feel embarrassed to discuss sexual health with my doctor") components [4] [5]. For instance, in a study on endometriosis and reproductive health, attitude was measured through items assessing women's feelings about sexual relationships, fertility problems, and sexual satisfaction [4].

Subjective Norms

Subjective norms represent an individual's perception of social pressure to perform or not perform a behavior, influenced by the judgment of significant others such as family, friends, healthcare providers, or society at large [1] [2]. This construct is grounded in normative beliefs—beliefs about whether important others think the individual should engage in the behavior—and the motivation to comply with those expectations [1].

• Dimensions: Subjective norms can encompass both injunctive norms (perceptions of what others approve/disapprove) and descriptive norms (perceptions of what others are actually doing) [2]. In reproductive health, these may include perceptions of family expectations regarding family planning, peer behaviors related to sexual activity, or societal norms about discussing reproductive concerns.
• Contextual Considerations: In many cultural contexts, reproductive health behaviors are strongly influenced by familial and community norms. For example, in a study among Ethiopian women, subjective norms significantly predicted intentions to use preconception care, reflecting the importance of social influence in health decision-making [5]. Similarly, research on adolescent sexual health in Iran incorporated parental perceptions as a distinct normative factor [3].

Perceived Behavioral Control

Perceived behavioral control (PBC) refers to an individual's perception of their ability to perform a given behavior, encompassing assessments of ease or difficulty, and the perceived presence of factors that may facilitate or impede behavioral performance [1] [2]. This construct is conceptually related to Bandura's concept of self-efficacy and is determined by control beliefs about internal and external barriers and facilitators [1].

• Dual Function: PBC is theorized to influence behavior both indirectly through intention and directly when perceptions of control accurately reflect actual behavioral control [1] [2].
• Application in Reproductive Health: In this domain, PBC might include perceptions of one's ability to access healthcare services, overcome logistical barriers (cost, transportation), navigate interpersonal dynamics in sexual relationships, or adhere to treatment regimens [4] [5]. For women with endometriosis, perceived control over managing symptoms and maintaining reproductive health significantly influenced their health behaviors and intentions [4].

Table 1: Core Constructs of the Theory of Planned Behavior

Construct	Definition	Composition	Measurement Examples in Reproductive Health
Attitude	Individual's positive or negative evaluation of performing the behavior	Behavioral beliefs + Outcome evaluations	"How beneficial do you believe regular gynecological check-ups are?" "How positive or negative do you feel about using contraception?"
Subjective Norms	Perception of social pressure from significant others to perform the behavior	Normative beliefs + Motivation to comply	"Most people who are important to me think I should use preconception care." "My partner supports my reproductive health decisions."
Perceived Behavioral Control	Perception of ability to perform the behavior, considering barriers and facilitators	Control beliefs + Perceived power	"How confident are you that you can access reproductive healthcare when needed?" "How much control do you have over preventing unintended pregnancy?"

TPB in Reproductive Health Research

Quantitative Evidence from Intervention Studies

The application of TPB in reproductive health research has demonstrated significant effects across diverse populations and health behaviors. Randomized controlled trials and cross-sectional studies have consistently shown that interventions targeting TPB constructs can effectively modify reproductive health intentions and behaviors.

Table 2: Summary of Quantitative Findings from TPB-Based Reproductive Health Interventions

Study Population	Intervention Type	Attitude Change	Subjective Norms Change	PBC Change	Behavioral Intention Change	Behavior Change	Citation
Iranian high school girls (n=578)	School-based sexual education	+16.8 points (15.3, 18.3)*	+16.4 points (14.83, 18.11)*	+18.0 points (16.6, 19.4)*	+18.4 points (14.8, 18.3)*	+18.5 points (16.8, 20.2)*	[3]
Women with endometriosis (n=71)	Online reproductive health education	Significant improvement (p<0.05)	Significant improvement (p<0.05)	Significant improvement (p<0.05)	Significant improvement (p<0.05)	Significant improvement in overall reproductive health (p<0.05)	[4]
Ethiopian women of reproductive age (n=427)	Community-based cross-sectional study	β=0.320 (p=0.0418)†	β=0.344 (p<0.001)†	β=0.512 (p<0.001)†	N/A (dependent variable)	19.8% had used preconception care	[5]

*Values represent score differences with 95% confidence intervals; †Standardized β-coefficients from multiple linear regression analysis

Modifications and Extensions in Reproductive Health Contexts

Research has demonstrated that extending the core TPB model with additional constructs can enhance its explanatory power in reproductive health contexts:

• Incorporating Parental Control: The study with Iranian adolescent girls added perceived parental control as an additional construct, recognizing the crucial role of parents in shaping adolescent sexual health behaviors in certain cultural contexts [3]. This extension acknowledges that for adolescents, parental monitoring and rules may constitute a distinct form of social influence beyond general subjective norms.

• Integrating Habit and Self-Identity: Some TPB extensions in health behavior research have incorporated habit (automaticity of behavior) and social identity (group membership and identification), which have been shown to improve prediction of behaviors that may become routine or are strongly tied to group membership [2]. While not always specific to reproductive health, these extensions offer promising directions for future research.

• Moderating Effects: Recent investigations have explored how PBC moderates the relationships between other TPB constructs and intention. Evidence suggests that strong perceived behavioral control tends to strengthen the effect of attitude on intention while weakening the effect of subjective norms [2]. This has implications for understanding how different constructs might be weighted in interventions targeting populations with varying levels of personal agency.

Research Methodology and Questionnaire Development

Experimental Protocols in TPB-Based Interventions

TPB-based interventions in reproductive health research typically follow structured protocols:

• Elicitation Studies: Prior to main data collection, qualitative methods (interviews, focus groups) identify salient behavioral, normative, and control beliefs specific to the target population and behavior [5]. For example, in the Ethiopian preconception care study, researchers conducted 18 in-depth interviews and 20 face-to-face elicitation interviews to identify locally relevant beliefs [5].

• Questionnaire Development: Based on elicitation results, structured questionnaires are developed with items measuring:

  • Direct measures of attitude, subjective norms, perceived behavioral control, and intention
  • Indirect measures assessing underlying behavioral, normative, and control beliefs
  • Demographic and background variables [4] [5]

• Intervention Implementation: Multi-session educational programs typically address all TPB constructs through various techniques:

  • Knowledge and attitude components: Presentations, lectures, slideshows
  • Subjective norms: Parental workshops, discussions about social influences
  • Perceived behavioral control: Skill-building exercises, problem-solving scenarios, resource identification [3] [4]

• Evaluation: Pre-test, post-test, and follow-up assessments measure changes in TPB constructs and behaviors, using statistical analyses (e.g., ANCOVA, multiple regression) to determine intervention effects [3] [4].

The Scientist's Toolkit: Essential Research Reagents

Table 3: Essential Methodological Components for TPB Research in Reproductive Health

Research Component	Function	Implementation Examples
TPB Questionnaire	Measures core constructs and underlying beliefs	Structured instrument with Likert-scale items for direct and indirect measures of TPB constructs [4] [5]
Elicitation Study Guide	Identifies population-specific beliefs	Semi-structured interviews or focus group protocols to uncover salient behavioral, normative, and control beliefs [5]
Educational Materials	Delivers intervention content	Booklets, presentations, and session plans tailored to the target population's literacy level and cultural context [3] [4]
Validation Instruments	Assesses criterion variables	Behavioral measures, clinical indicators, or validated scales measuring the target health outcomes (e.g., Endometriosis Reproductive Health Questionnaire) [4]
Data Collection Platforms	Facilitates data gathering	Online survey tools (e.g., for remote populations), face-to-face interview protocols, or paper-based questionnaires for low-resource settings [4] [5]

Conceptual Framework and Visual Representation

The following diagram illustrates the structural relationships between TPB constructs in the context of reproductive health behavior:

TPB Framework in Reproductive Health Context

This conceptual model illustrates how behavioral, normative, and control beliefs about reproductive health behaviors give rise to the core TPB constructs, which collectively influence behavioral intention. Intention, in turn, predicts the performance of reproductive health behaviors, with perceived behavioral control potentially having both direct and indirect effects, particularly when it accurately reflects actual control conditions.

The Theory of Planned Behavior provides a robust conceptual framework for understanding and predicting reproductive health behaviors across diverse populations and contexts. Its systematic approach to measuring and targeting the cognitive antecedents of behavior—attitude, subjective norms, and perceived behavioral control—makes it particularly valuable for developing evidence-based interventions and reliable assessment tools in this critical health domain.

The quantitative evidence from reproductive health studies demonstrates that TPB-based interventions can effectively modify key constructs and ultimately improve health behaviors and outcomes. However, researchers should consider contextual factors and potential model extensions—such as incorporating parental control for adolescent populations or accounting for cultural specificities—to optimize the theory's applicability and predictive power. As reproductive health challenges continue to evolve, the TPB remains a versatile framework for designing targeted behavioral interventions that address the complex interplay of individual, social, and environmental factors influencing health decision-making.

Why TPB? The Framework's Predictive Power for Health Intentions and Behaviors

The Theory of Planned Behavior (TPB) has emerged as a predominant cognitive framework for predicting and understanding health intentions and behaviors. Within sexual and reproductive health (SRH) research, its utility is particularly pronounced for designing evidence-based interventions. This whitepaper examines the core constructs of TPB and its predictive power, supported by quantitative data from experimental studies. It further provides a detailed methodological protocol for implementing TPB within SRH questionnaire research, serving as a guide for researchers and drug development professionals working in behaviorally-focused health outcomes.

Theoretical Foundations of the Theory of Planned Behavior

The Theory of Planned Behavior (TPB), developed by Icek Ajzen, proposes that an individual's decision to engage in a specific behavior is predicated by their intention to perform that behavior [6]. Intention is considered the immediate antecedent to behavior, capturing the motivational factors that influence it; it signifies how hard people are willing to try and how much effort they are planning to exert to perform the behavior [6]. The theory posits that the stronger the intention to engage in a behavior, the more likely its performance will be.

As an extension of the Theory of Reasoned Action, TPB incorporates perceived behavioral control as a critical third determinant, alongside attitude and subjective norms [6] [7]. These three core constructs collectively shape an individual's behavioral intentions [6]:

• Attitude: The individual's positive or negative evaluation of performing the specific behavior, encompassing their beliefs about the outcomes of the behavior.
• Subjective Norms: The individual's perception of social pressure from significant others (e.g., family, friends, peers) to perform or not perform the behavior.
• Perceived Behavioral Control: The extent to which an individual believes they can control the performance of the behavior, factoring in internal constraints (e.g., skills, willpower) and external barriers (e.g., resources, opportunities). This construct is conceptually aligned with Bandura's concept of self-efficacy [6].

The following diagram illustrates the logical relationships between these core constructs and their subsequent impact on behavior and behavioral achievement.

Quantitative Evidence of Predictive Power

Empirical evidence robustly supports TPB's utility in predicting health intentions and behaviors. A systematic review of TPB-based interventions in low- and middle-income countries (LMICs) found them to be effective in changing health behavior and the underlying TPB constructs across various chronic diseases, establishing their feasibility and fidelity in diverse settings [8]. The review highlighted that structured education based on TPB constructs led to measurable improvements in health outcomes.

A pivotal randomized controlled trial (RCT) conducted among 578 high school girls in Tehran, Iran, demonstrates the significant impact of a TPB-based educational intervention on SRH outcomes [3]. The study measured changes in TPB constructs before and after a three-month intervention, with results showing remarkable improvement in the experimental group compared to the control group.

Table 1: Quantitative Outcomes of a TPB-Based SRH Intervention in Adolescent Girls

TPB Construct	Pre-/Post-Intervention Score Difference	95% Confidence Interval	P-value
Attitude	16.8	15.3, 18.3	<0.001
Subjective Norms	16.4	14.83, 18.11	<0.001
Perceived Behavioral Control	18.0	16.6, 19.4	<0.001
Perceived Parental Control	17.0	15.1, 18.9	<0.001
Behavioral Intention	18.4	14.8, 18.3	<0.001
Behavior	18.5	16.8, 20.2	<0.001

Source: Adapted from [3]

These findings are corroborated by earlier research, which found that the three components of the model correlated with alcohol addicts' intentions to limit or stop drinking, and these intentions subsequently predicted the approximate number of alcohol units consumed at one- and three-month follow-ups [6]. However, it is critical to acknowledge the intention-behavior gap. A meta-analysis of 47 studies indicated that while a link between intention and actual behavior exists, the effect size can be small, suggesting that other moderating factors may influence the final behavioral translation [6].

Experimental Protocol: A TPB-Based SRH Intervention

The following detailed methodology is adapted from the RCT conducted in Iran, which serves as a model for implementing TPB in SRH research [3].

Study Design and Setting

• Design: Randomized Controlled Trial (RCT).
• Participants: First-year high school girls, aged 12-16 years.
• Sample Size: 578 participants (289 in experimental group, 289 in control group), calculated with 90% power to detect a 10-score difference in knowledge.
• Randomization: Multistage random cluster sampling. Three districts were selected via proportional stratified random sampling, with schools and classes randomly selected within districts. Schools were then randomly assigned to experimental or control groups using permuted block randomization.

Intervention Program

The intervention was a multi-component, school-based program delivered over three months with a six-month follow-up, designed to address the core TPB constructs [3].

• Content: Focused on precise information regarding puberty health, adolescent health, STD diseases, HIV transmission routes, HIV prevention, and skills to make healthy choices.
• Techniques:
  • For Adolescents: Training sessions using presentation classes, lectures, slideshows, role-playing, training animation, story-writing, brainstorming, and small group discussions. A specially designed 17-page pictorial booklet entitled "Training 12-16-yr-old Adolescents at a Turning Point in Life" was provided.
  • For Parents: A two-hour workshop on high-risk behaviors and SRH issues to influence subjective norms and perceived parental control.

Table 2: Mapping of TPB Constructs to Educational Strategies

TPB Construct	Targeted Educational Strategy
Knowledge/Attitude	Lectures, slideshows, role-playing, animations
Subjective Norms	Parental training sessions, booklets
Perceived Behavioral Control	Story-writing, small group discussions, skills-based booklet
Perceived Parental Control	Dedicated training sessions for parents

Data Collection and Measures

A structured, anonymous, self-administered TPB-based questionnaire with 132 items was used for pre- and post-intervention assessment [3]. The questionnaire was developed based on WHO instruments, literature review, and prior qualitative work (eight focus group discussions).

Questionnaire Sections:

• Part A: Demographic information.
• Part B: 28 items on general knowledge about SRH (e.g., "HIV is transmitted through unprotected vaginal and anal sex").
• Part C: TPB constructs measured by:
  • 21 items on Attitude (e.g., "In my opinion, sexual and reproductive health is a serious problem for the health of all people.").
  • 19 items on Subjective Norms (e.g., "People around me think that I should avoid risky behaviors...").
  • 25 items on Perceived Behavioral Control (e.g., "I'm sure that I can avoid risky behaviors such as unprotected sex and tattoos...").
  • 18 items on Behavioral Intention (e.g., "I'm going to talk about sexual and reproductive health problems with my parents...").
  • 8 items on Perceived Parental Control.

The workflow for developing and implementing this TPB-based intervention is summarized in the following diagram.

The Researcher's Toolkit: Essential Reagents for TPB Research

For scientists designing TPB-based SRH questionnaire research, the following "research reagents" are essential for ensuring valid and reliable data collection and analysis.

Table 3: Essential Reagents for TPB-Based SRH Questionnaire Research

Research Reagent	Function/Description	Exemplar from Literature
TPB-Based Questionnaire	A validated, multi-item instrument measuring demographic data, knowledge, and all TPB constructs (Attitude, Subjective Norms, PBC, Intention).	132-item tool with sections from WHO questionnaire and locally adapted qualitative work [3].
Intervention Booklet	A theory-based educational booklet using simple language and graphics to address knowledge gaps and target TPB constructs.	17-page full-color booklet, "Training 12-16-yr-old Adolescents at a Turning Point in Life" [3].
Structured Educational Sessions	Manualized sessions (e.g., presentations, role-playing) for participants to deliver standardized intervention content.	Four sessions held over two 45-min parts with a 15-min break [3].
Parental Workshop Materials	Educational content for parents/guardians designed to influence participants' subjective norms and perceived parental control.	A dedicated two-hour workshop for parents [3].
Data Analysis Software	Statistical software capable of handling covariance-based analysis, regression, and other advanced tests to evaluate intervention effects.	SPSS version 16 used for analysis of covariance (ANCOVA) [3]. Other tools like JASP and IBM SPSS are also applicable [9].

The Theory of Planned Behavior provides a robust, empirically-supported framework for predicting health intentions and behaviors, with particular utility in the sensitive domain of sexual and reproductive health. Its predictive power stems from a structured model that accounts for personal, social, and control-related factors. The quantitative evidence and detailed experimental protocol outlined in this whitepaper demonstrate that TPB-based interventions, when properly designed and implemented with validated tools, can produce statistically significant and clinically meaningful improvements in SRH outcomes. For researchers and drug development professionals, TPB offers a reliable roadmap for developing and evaluating behavior change interventions, from initial questionnaire design to final outcome assessment.

The Theory of Planned Behavior (TPB) provides a robust framework for predicting and understanding health behaviors by examining the psychological determinants of intentions and actions [10]. This social cognition theory posits that behavioral intentions are shaped by attitude (personal evaluation of the behavior), subjective norms (perceived social pressure), and perceived behavioral control (beliefs about capabilities to perform the behavior) [10]. In reproductive health contexts, where behaviors are often influenced by unique cultural, social, and relational factors, researchers have recognized the need to extend the standard TPB model by incorporating context-specific constructs such as parental control [11].

This adaptation enhances the theory's predictive validity and practical utility for designing targeted interventions. The incorporation of parental control is particularly relevant in adolescent reproductive health, where parents often serve as key influencers of behaviors, norms, and access to resources [12] [11]. This technical guide provides a comprehensive framework for systematically adapting the TPB through the integration of parental control and other context-specific constructs, with specific application to reproductive health questionnaire development and intervention design.

Theoretical Foundation: Extending the TPB Framework

The Standard TPB Model

The standard TPB model conceptualizes behavior as directly determined by behavioral intention, which in turn is influenced by three core constructs [10]:

• Attitude: An individual's positive or negative evaluation of performing the particular behavior
• Subjective Norm: An individual's perception about the particular behavior, influenced by significant others
• Perceived Behavioral Control: An individual's perception of their ability to perform the behavior

Meta-analyses of TPB applications in health contexts have consistently supported these hypothesized relationships, demonstrating the theory's capacity to account for unique variance in multiple health behaviors across diverse populations [10].

Rationale for Incorporating Parental Control

In many cultural contexts, particularly those with collectivist orientations or where adolescents remain financially and emotionally dependent on parents, parental influence extends beyond general subjective norms to include specific control mechanisms [11]. These may include:

• Direct monitoring of activities and relationships
• Rule-setting regarding social interactions
• Control over resources necessary for accessing health services
• Explicit communication of expectations and consequences

The addition of parental control as a distinct construct accounts for these unique influence mechanisms that are not fully captured by traditional subjective norms measures [11]. Empirical evidence demonstrates that this adapted model significantly improves the prediction of intentions and behaviors in sensitive health domains like sexual and reproductive health [11].

Table 1: Theoretical Constructs in Standard and Adapted TPB Models

Construct	Definition in Standard TPB	Expansion in Adapted TPB
Attitude	Individual's positive/negative evaluation of performing the behavior	May include specific beliefs about parental reactions to behaviors
Subjective Norm	Perception of social pressure from significant others	Distinguishes between peer norms and parental norms
Perceived Behavioral Control	Beliefs about capabilities to perform the behavior	May include perceptions of control under parental supervision
Parental Control	Not included in standard model	Parental monitoring, rule-setting, and resource control that directly influences behavioral opportunities

Methodological Protocol for Model Adaptation

Systematic Approach to Construct Integration

The process of adapting the TPB framework requires careful methodological consideration to maintain theoretical coherence while enhancing contextual relevance. The following systematic approach ensures rigorous model development:

Phase 1: Qualitative Exploration

• Conduct in-depth interviews and focus group discussions with the target population to identify salient parental influence factors [13] [11]
• Explore specific mechanisms of parental control relevant to the behavioral domain
• Identify language and terminology used by the population to describe parental influence

Phase 2: Theoretical Mapping

• Analyze how identified parental influence factors relate to existing TPB constructs
• Determine whether new constructs require development or existing constructs need refinement
• Develop conceptual definitions for new constructs that distinguish them from core TPB variables

Phase 3: Instrument Development

• Generate items that operationalize the new parental control construct
• Ensure items cover multiple dimensions of parental control (monitoring, rules, resource control)
• Maintain parallel structure with existing TPB measures for comparability

Phase 4: Psychometric Validation

• Assess content validity through expert review and target population feedback
• Conduct exploratory and confirmatory factor analysis to establish discriminant validity
• Evaluate internal consistency and test-retest reliability of new scales

Experimental Implementation Protocol

Based on successful applications in reproductive health research, the following protocol outlines the specific steps for implementing an adapted TPB intervention:

Table 2: Experimental Protocol for TPB-Based Intervention with Parental Control Component

Stage	Activities	Duration	Outcome Measures
Baseline Assessment	Administer TPB questionnaire with parental control scale; collect demographic data	1 session	Pre-intervention scores on all TPB constructs and parental control
Intervention Design	Develop educational materials addressing each TPB construct + parental control	2-3 weeks	Theory-based educational curriculum
Intervention Delivery	Conduct multiple sessions using varied teaching methods (role-playing, discussions, booklets)	3-4 sessions over 1 month	Session completion rates; engagement metrics
Parental Component	Separate workshop for parents addressing communication and monitoring practices	1-2 sessions	Parent attendance; pre-post knowledge assessment
Post-Intervention Assessment	Readminister TPB questionnaire with parental control scale	4 weeks post-intervention	Immediate intervention effects on constructs
Follow-Up Assessment	Readminister key measures to assess sustained effects	6-8 months post-intervention	Long-term behavior change and construct stability

A recent randomized controlled trial implemented this protocol with 578 high school girls in Iran, demonstrating significant improvements in attitude, subjective norms, perceived behavioral control, parental control, and behavioral intentions related to sexual and reproductive health following the intervention [11]. The experimental group showed significant improvement in parental control scores (17% increase, 95% CI: 15.1, 18.9) compared to the control group [11].

Measurement Approaches and Questionnaire Design

Scale Development for Parental Control Construct

The development of a psychometrically sound parental control scale requires attention to several key factors:

Item Generation Based on successful scale development protocols, initial item pools should be generated through [13] [11]:

• Comprehensive literature reviews of existing parenting measures
• Qualitative interviews with target population (34 participants recommended)
• Consultation with content experts (4+ specialists recommended)
• Adaptation of items from validated scales when appropriate

Response Format

• Use Likert scales (5-point) for attitude and perception items
• Employ dichotomous scales (yes/no) for specific behavior inquiries
• Include "don't know" options to reduce guessing
• Maintain consistent directionality throughout scales

Psychometric Validation Establish validity and reliability through [14] [13]:

• Content Validity Ratio (CVR): Minimum 0.62 for 10 experts
• Content Validity Index (CVI): >0.79 for relevancy, simplicity, and clarity
• Face Validity: Assessment by target population for comprehension
• Internal Consistency: Cronbach's alpha >0.70 (preferably >0.80)
• Test-Retest Reliability: Intraclass correlation coefficient >0.80

Addressing Measurement Challenges

Social Desirability Bias Parental control measures are particularly susceptible to social desirability bias. Mitigation strategies include [12]:

• Question formatting: Use normalizing preamble statements
• Indirect questioning: Assess perceptions rather than direct behaviors
• Implicit measures: Complement self-report with reaction-time tasks
• Corroborative data: Collect parallel reports from parents or spouses

Contextual Specificity Reproductive health topics require careful contextualization [13]:

• Use culturally appropriate terminology for sensitive topics
• Consider developmental stage of target population
• Account for gender-specific concerns and norms
• Adapt to legal and policy frameworks affecting parental rights

Data Analysis and Model Validation Techniques

Analytical Approaches for Adapted Models

The expanded TPB model with parental control requires specific analytical approaches to test theoretical relationships and intervention effects:

Structural Equation Modeling (SEM)

• Test simultaneous relationships between all TPB constructs and parental control
• Account for measurement error in latent constructs
• Compare fit indices between standard and expanded models
• Assess direct, indirect, and moderating effects of parental control

Moderation Analysis Examine whether parental control moderates traditional TPB pathways:

• Parental control × attitude → intention
• Parental control × subjective norms → intention
• Parental control × intention → behavior

Validation Metrics Employ quantitative validation techniques to establish model accuracy [15]:

• Bayesian hypothesis testing for model comparison
• Reliability-based metrics for prediction accuracy
• Area metric-based methods for comprehensive evaluation

Intervention Effect Assessment

Randomized controlled trials should assess intervention effects on:

• Mean changes in parental control scores pre- and post-intervention
• Between-group differences (intervention vs. control)
• Mediation effects through TPB constructs
• Behavioral outcomes at follow-up assessments

The diagram below illustrates the analytical workflow for validating the adapted TPB model:

Case Application: Reproductive Health Questionnaire

Exemplar Implementation

A recent study successfully developed and implemented a reproductive health questionnaire for married adolescent women using an exploratory sequential mixed methods approach [13]. The process included:

Phase 1: Qualitative Development

• 34 in-depth interviews with married adolescent women
• 4 key informant interviews with health professionals
• Comprehensive literature review of existing instruments
• Generation of 137 initial items

Phase 2: Psychometric Validation

• Face and content validity assessment → reduction to 85 items
• Exploratory factor analysis → removal of 11 items, retention of 74 items
• Categorization into 9 factors including family support and husband involvement
• Reliability testing: Cronbach's alpha = 0.878, ICC = 0.99

The final instrument assessed multiple dimensions of reproductive health needs, including self-care, self-efficacy, knowledge, and social support systems [13].

Intervention Efficacy Evidence

A randomized controlled trial demonstrated the efficacy of a TPB-based educational intervention incorporating parental control for improving sexual and reproductive health in Iranian adolescent girls [11]. The study implemented:

Sample Design

• 578 high school girls (289 intervention, 289 control)
• Multistage random cluster sampling
• 12-16 years age range

Intervention Components

• Education sessions for adolescents (knowledge, attitude, behavioral control)
• Separate workshops for parents (subjective norms, parental control)
• Educational booklet tailored to adolescent developmental level
• Multiple delivery methods: role-playing, animations, group discussions

Outcomes The intervention resulted in statistically significant improvements (p<0.001) in all TPB constructs including parental control, with effect sizes ranging from 16.4 to 18.5 points on construct measures [11].

Research Reagent Solutions: Essential Methodological Tools

Table 3: Essential Research Tools for TPB Adaptations in Reproductive Health

Tool Category	Specific Instrument	Function	Psychometric Properties
Qualitative Assessment	Interview guides; Focus group protocols	Elicit salient beliefs; Identify context-specific factors	Establish content representativeness; Saturation metrics
TPB Core Measures	Direct measures of attitude, subjective norms, perceived behavioral control	Assess core theoretical constructs	Established reliability (α >0.70); Predictive validity
Parental Control Scale	Custom-developed parental monitoring and control items	Assess specific parental influence mechanisms	CVI >0.79; CVR >0.62; α >0.80 [14]
Reproductive Health Knowledge	Domain-specific knowledge tests	Evaluate factual understanding	Difficulty indices; Discrimination coefficients
Behavioral Measures	Self-report behavioral frequency; Clinical indicators	Assess behavioral outcomes	Test-retest reliability; Social desirability checks
Social Desirability Scale	Marlowe-Crowne Social Desirability Scale	Assess response bias	Established norms; Cross-cultural adaptations

The systematic adaptation of the Theory of Planned Behavior through the incorporation of context-specific constructs like parental control significantly enhances the model's utility for reproductive health research and intervention. This approach maintains theoretical integrity while improving ecological validity and predictive accuracy. The methodological framework presented in this guide provides researchers with evidence-based strategies for model expansion, measurement development, and intervention design. As reproductive health challenges continue to evolve, particularly among vulnerable populations like adolescents, such theoretically grounded yet contextually responsive approaches will be essential for developing effective health promotion strategies.

The Theory of Planned Behavior (TPB) provides a robust conceptual framework for understanding and predicting health behaviors, making it particularly valuable for designing reproductive health questionnaires. According to this theory, behavioral intention—the most immediate predictor of behavior—is influenced by three key constructs: attitude (personal evaluation of the behavior), subjective norms (perceived social pressure), and perceived behavioral control (beliefs about capabilities to perform the behavior) [16] [17] [3]. In reproductive health research, where sensitive topics and private behaviors are common, TPB offers a structured approach to measuring determinants that directly influence health outcomes such as contraceptive use, preconception care, and service-seeking behaviors.

The application of TPB in reproductive health has demonstrated significant predictive power across diverse populations. A 2023 study conducted in Indonesia with 341 women of childbearing age found that preconception behavior was directly and positively influenced by intention (b=0.33), perceived behavioral control (b=0.23), and attitude (b=0.22) [16]. Similarly, research in Inner Mongolia, China, with 1,399 couples revealed that fertility decisions were significantly influenced by perceived behavioral control (including the importance of stable income and costs of raising children), subjective norms (particularly sex preference of the newborn), and attitudes about parental health requirements [17]. These findings underscore the utility of TPB for developing targeted reproductive health interventions across different cultural contexts.

Core TPB Constructs and Their Operationalization in Reproductive Health

Defining and Measuring Key Constructs

Attitude in reproductive health contexts encompasses an individual's positive or negative evaluation of performing specific health behaviors. This construct is typically measured through semantic differential scales that capture both instrumental (e.g., harmful-beneficial) and experiential (e.g., unpleasant-pleasant) components. For example, in a study of high school girls in Iran, attitudes were measured using items such as "In my opinion, sexual and reproductive health is a serious problem for the health of all people" [3]. Research has demonstrated that attitude consistently emerges as a significant predictor of reproductive health intentions, with the Indonesian study reporting a direct positive influence on preconception behavior (b=0.22, CI=0.11-0.36, P=0.001) [16].

Subjective norms refer to the perceived social pressure from significant others (partners, family, peers) to engage or not engage in reproductive health behaviors. This construct includes both injunctive norms (what important others think one should do) and descriptive norms (what others are actually doing). In fertility decision-making research in China, subjective norms were operationalized through items measuring perceived social pressure about "sex preference of the newborn by themselves and their partner" [17]. The influence of subjective norms may be indirect; in the Indonesian preconception care study, subjective norms influenced behavior indirectly through other TPB constructs (b=0.11, CI=0.01-0.21, P=0.037) [16].

Perceived behavioral control (PBC) reflects an individual's confidence in their ability to perform a behavior, accounting for facilitators and barriers. In reproductive health contexts, PBC often encompasses access to services, financial resources, and self-efficacy. The study in Inner Mongolia identified that perceived importance of having a stable income and cost of raising a child were significant PBC factors in fertility decisions [17]. Similarly, the Indonesian study found PBC had both direct (b=0.23, CI=0.12-0.35, P=0.001) and indirect (b=0.31, CI=0.22-0.40, P=0.001) effects on preconception behavior [16].

Quantitative Evidence for TPB Construct Effectiveness

Table 1: Direct and Indirect Effects of TPB Constructs on Preconception Behavior (n=341) [16]

TPB Construct	Direct Effect (b)	95% CI	P-value	Indirect Effect (b)	95% CI	P-value
Intention	0.33	0.22 to 0.45	0.001	-	-	-
Perceived Behavioral Control	0.23	0.12 to 0.35	0.001	0.31	0.22 to 0.40	0.001
Attitude	0.22	0.11 to 0.36	0.001	0.31	0.22 to 0.40	0.001
Subjective Norms	-	-	-	0.11	0.01 to 0.21	0.037

Table 2: Intervention Impact on TPB Constructs in Adolescent Girls (n=578) [3]

TPB Construct	Pre-Intervention Score (Experimental)	Post-Intervention Score (Experimental)	Difference (95% CI)	P-value
Attitude	62.4	79.2	16.8 (15.3, 18.3)	<0.001
Subjective Norms	58.7	75.1	16.4 (14.8, 18.1)	<0.001
Perceived Behavioral Control	61.3	79.3	18.0 (16.6, 19.4)	<0.001
Behavioral Intention	59.8	78.2	18.4 (16.8, 20.0)	<0.001

Questionnaire Development Methodologies

Systematic Approach to Instrument Development

The development of a psychometrically sound TPB-based reproductive health questionnaire requires a methodical multi-phase approach. A recommended framework involves three distinct phases: (1) conceptualization and item generation, (2) content validation, and (3) psychometric testing [18] [19]. This systematic process ensures that the resulting instrument accurately captures TPB constructs while being culturally appropriate for the target population.

The item generation phase typically employs deductive methods (logical partitioning) based on existing theories and frameworks. Researchers first define the constructs to be measured, then generate items that align with these predefined concepts [19]. This approach ensures theoretical consistency and validity, as items are directly linked to established literature and models. For reproductive health questionnaires, this phase often involves adapting items from previously validated tools while ensuring cultural and contextual relevance. The Iranian study on adolescent sexual and reproductive health, for instance, developed a 132-item questionnaire measuring TPB constructs, with some sections founded on WHO questionnaires [3].

Content validation is essential for establishing the relevance and comprehensiveness of questionnaire items. This typically involves expert reviews and focus groups with target population representatives. In the development of the Total Teen Assessment, researchers conducted focus groups with youth (n=8) to review and critique questions, assessing content coverage relative to lived experiences and appropriateness regarding format, verbiage, clarity, and length [18]. Similarly, the Sexual and Reproductive Health Service Seeking Scale (SRHSSS) development included expert evaluation from psychiatric and gynecology nursing specialists [20]. This collaborative process between researchers and stakeholders ensures the instrument's face validity and cultural appropriateness.

Psychometric Validation Procedures

Psychometric testing establishes the reliability and validity of the developed instrument. This phase typically employs both exploratory and confirmatory factor analysis to examine the underlying factor structure, alongside reliability testing using measures such as Cronbach's alpha [18] [20]. The SRHSSS development, for instance, involved exploratory factor analysis with a sample of 458 young adults, revealing a four-factor structure that explained 89.45% of the total variance, with factor loadings ranging from 0.78-0.97 and Cronbach's alpha of 0.90, indicating strong internal consistency [20].

Test-retest reliability is also crucial for establishing instrument stability over time. In the SRHSSS validation, test-retest reliability was performed with 220 participants one month after the initial measurement [20]. For TPB-specific questionnaires, structural equation modeling (SEM) is particularly valuable for testing the theoretical relationships between constructs. The Indonesian preconception care study used SEM to measure both construction relationships (measurable) and measurement relationships (not measurable), determining goodness of fit and significance of variable influences using indices such as CMIN/DF, RMSEA, CFI, TLI, and SRMR [16].

Experimental Protocols for TPB-Based Interventions

Implementing and Evaluating TPB-Based Educational Programs

TPB-based educational interventions in reproductive health follow structured protocols to effectively modify behavioral determinants. A randomized controlled trial conducted in Tehran, Iran, provides a exemplary methodology for implementing and evaluating such interventions [3]. The study employed a school-based approach with first-year high school girls (12-16 years), randomly assigning 578 participants to experimental (n=289) and control (n=289) groups using multistage random cluster sampling. The intervention incorporated multiple educational strategies tailored to specific TPB constructs: knowledge components used presentation classes (40-45 minutes), lectures, and slideshows; attitude components employed role-playing, training animation, and story-writing; subjective norms components included training sessions for parents and booklets; and perceived behavioral control components utilized story-writing, small group discussions, and booklets.

The comprehensive intervention program spanned three months of active education followed by six months of follow-up assessment. Educational materials included a specially developed 17-page full-color booklet titled "Training 12-16-year-old Adolescents at a Turning Point in Life," addressing puberty, menstrual health, and HIV/STD prevention using simple language aligned with TPB constructs [3]. The program also included a two-hour workshop for parents addressing high-risk behaviors and sexual/reproductive health issues, recognizing the importance of parental influence on adolescents' subjective norms and perceived behavioral control. This comprehensive approach resulted in significant improvements in all TPB constructs in the experimental group compared to controls, including attitude (difference=16.8), subjective norms (16.4), perceived behavioral control (18.0), and behavioral intention (18.4) [3].

Methodological Considerations for Behavioral Validation

Behavioral test validation is a critical but often overlooked component of TPB intervention research. As noted in reproducibility studies, even well-established behavioral tests require laboratory-specific validation to ensure they adequately capture expected phenotypes [21]. This process involves implementing three types of controls: baseline control groups (to determine if a test functions under laboratory-specific conditions), positive control groups (to determine if a test detects expected phenotypic changes), and internal controls (integrated into test designs to determine test validity) [21].

Research has demonstrated that behavioral phenotypes can shift due to differences in laboratory environments, animal care personnel, housing conditions, or genetic drift, necessitating periodic re-validation even within the same research group [21]. This principle applies equally to human behavioral research in reproductive health, where contextual factors such as data collection settings, interviewer characteristics, and cultural norms may influence responses. The methodology used in the Inner Mongolia fertility study addressed these concerns through careful instrument development, including a literature review, in-depth interviews, expert panel assessment, and pilot testing before survey implementation [17].

Data Visualization: TPB Pathways and Questionnaire Development

Theory of Planned Behavior Pathway in Reproductive Health

Questionnaire Development and Validation Workflow

Table 3: Research Reagent Solutions for TPB Reproductive Health Research

Tool/Resource	Function/Purpose	Example Applications	Psychometric Properties
TPB Questionnaire [16] [17]	Measures core TPB constructs (attitude, subjective norms, perceived behavioral control, intention)	Preconception care behavior [16], Fertility decision-making [17]	Validated in multiple cultures; Strong construct validity
Total Teen (TT) Assessment [18]	Electronic screening for adolescent health needs across sexual/reproductive health, mental health, and substance use	Primary care settings for comprehensive adolescent health assessment	Three-factor structure; Clinical and statistical validity
Sexual and Reproductive Health Service Seeking Scale (SRHSSS) [20]	Assesses thoughts, attitudes, and perceived barriers to accessing SRH services	Young adults' SRH service seeking behaviors	4-factor structure (89.45% variance); α=0.90
Reproductive Health Literacy Scale [22]	Measures effectiveness of health literacy training in refugee populations	Refugee women's reproductive health knowledge	Combines HLS-EU-Q6, eHEALS, and reproductive health items; α>0.7
Integrated OMSRH Tool [19]	Comprehensive assessment of oral, mental, and sexual/reproductive health interconnections	Holistic adolescent health assessment in Nigeria	81 items across five sections; Content validity established

The Theory of Planned Behavior provides a robust theoretical foundation for developing reproductive health questionnaires that effectively predict and explain health behaviors. The quantitative evidence demonstrates significant direct and indirect effects of TPB constructs on reproductive health behaviors across diverse populations and contexts. By following systematic development methodologies—including conceptualization, content validation, and psychometric testing—researchers can create culturally appropriate instruments that reliably measure behavioral determinants.

The successful application of TPB in reproductive health research requires careful attention to both theoretical fidelity and methodological rigor. As shown in the experimental protocols and validation workflows, this includes implementing appropriate controls, establishing reliability and validity across different populations, and utilizing structured intervention approaches that target specific TPB constructs. The tools and resources summarized in this technical guide provide researchers with essential starting points for developing context-specific instruments that can advance our understanding of reproductive health behaviors and enhance the effectiveness of interventions aimed at improving reproductive health outcomes worldwide.

Questionnaire Development in Practice: A Step-by-Step Methodological Guide

Item generation is a critical foundational step in the development of a psychometrically sound research instrument. Within the context of the Theory of Planned Behavior (TPB), this phase involves the systematic creation of questionnaire items that adequately capture the core constructs of the theory: behavioral beliefs (attitude), normative beliefs (subjective norm), control beliefs (perceived behavioral control), and behavioral intention [23]. The quality of items generated during this phase directly influences the validity and reliability of the entire research instrument. This technical guide provides researchers with comprehensive methodologies for item generation, integrating both qualitative approaches and systematic literature review processes, with specific application to reproductive health research.

The TPB provides a robust theoretical framework for understanding and predicting health behaviors by postulating that behavioral intention—the most proximal determinant of behavior—is influenced by attitude toward the behavior, subjective norms, and perceived behavioral control [23]. In reproductive health research, applying the TPB requires careful contextualization of these constructs to specific behaviors (e.g., contraceptive use, prenatal care attendance, HIV testing). Proper item generation ensures that the resulting questionnaire validly measures these theoretical constructs within the specific cultural and contextual framework of the target population.

Theoretical Framework and Construct Operationalization

Core TPB Constructs in Reproductive Health Context

The first step in item generation involves clearly defining and operationalizing each TPB construct within the specific reproductive health context. The table below outlines the core TPB constructs and their application in reproductive health research:

Table 1: Operationalization of TPB Constructs for Reproductive Health Questionnaires

TPB Construct	Theoretical Definition	Reproductive Health Application	Measurement Focus
Behavioral Beliefs (Attitude)	Beliefs about the likely outcomes of a behavior and evaluations of these outcomes	Beliefs about consequences of using contraception, engaging in safe sex practices, or seeking prenatal care	Perceived advantages/disadvantages of the reproductive health behavior
Normative Beliefs (Subjective Norm)	Beliefs about the normative expectations of important referents and motivation to comply with these expectations	Perceptions of what partners, family, peers, or healthcare providers think about the reproductive health behavior	Perceived social pressure regarding behavioral performance
Control Beliefs (Perceived Behavioral Control)	Beliefs about the presence of factors that may facilitate or impede behavioral performance and the perceived power of these factors	Perceptions of barriers and facilitators to accessing reproductive healthcare, obtaining contraceptives, or negotiating safe sex	Perceived capability and autonomy to perform the behavior
Behavioral Intention	The individual's readiness to perform a given behavior	Intention to use specific contraceptive methods, get tested for STIs, or attend scheduled prenatal visits	Self-reported likelihood of engaging in the target behavior

Defining the Target Behavior

A precise definition of the target reproductive health behavior is essential before item generation. The behavior should be defined using action, context, time, and target elements. For example, rather than studying "contraceptive use," a well-defined behavior would be "consistent condom use with primary partner during vaginal intercourse over the next month." This specificity ensures that generated items accurately reflect the theoretical constructs in relation to the specific behavior of interest.

Qualitative Research Methodologies for Item Generation

Study Population and Sampling

Qualitative research for item generation requires careful consideration of sampling strategies to ensure the inclusion of diverse perspectives relevant to the reproductive health behavior under investigation. The sample should include individuals from the target population who have varying experiences with the behavior (both those who engage and do not engage in the behavior), as well as key informants where appropriate.

Table 2: Sampling Approaches for Qualitative Item Generation

Sampling Method	Application in TPB Research	Sample Size Considerations	Advantages for Item Generation
Purposive Sampling	Selecting participants with specific experience with the reproductive health behavior	Typically 15-30 participants until thematic saturation achieved	Ensures inclusion of information-rich cases relevant to the research questions
Theoretical Sampling	Selecting participants based on emerging conceptual needs from ongoing analysis	Determined by theoretical saturation of constructs	Allows refinement of questions based on emerging themes and theoretical constructs
Maximum Variation Sampling	Selecting participants with diverse demographic and behavioral characteristics	20-40 participants to capture wide range of perspectives	Enhances comprehensiveness of beliefs identified across different subgroups
Snowball Sampling	Accessing hard-to-reach populations in reproductive health research	Varies based on population accessibility	Useful for sensitive topics and hidden populations in reproductive health

For example, a study on DRG payment制度中医生策略性行为 utilized purposive sampling, selecting 308 physicians from various hospital types to ensure diverse perspectives [23]. Similarly, reproductive health research might purposively sample women of different ages, socioeconomic statuses, and reproductive histories to capture the full spectrum of beliefs.

Data Collection Protocols

Semi-Structured Interviews

Semi-structured interviews provide depth and context for understanding beliefs about reproductive health behaviors. The interview guide should include open-ended questions tailored to each TPB construct:

Behavioral Beliefs Questions:

• What do you see as the advantages of [reproductive health behavior]?
• What do you see as the disadvantages of [reproductive health behavior]?
• What other good or bad effects might result from [reproductive health behavior]?

Normative Beliefs Questions:

• Are there any individuals or groups who would approve of you [performing the behavior]?
• Are there any individuals or groups who would disapprove of you [performing the behavior]?
• Whose opinions would matter most when you decide whether to [perform the behavior]?

Control Beliefs Questions:

• What factors might make it easier for you to [perform the behavior]?
• What factors might make it harder for you to [perform the behavior]?
• What would help you overcome these barriers?

Each interview should be audio-recorded, transcribed verbatim, and conducted in a private setting to ensure confidentiality, particularly important for sensitive reproductive health topics.

Focus Group Discussions

Focus group discussions (FGDs) leverage group dynamics to elicit a range of beliefs and normative influences. The protocol should include:

• 6-10 participants with similar characteristics (to encourage open discussion)
• A skilled moderator and note-taker
• A semi-structured discussion guide similar to the interview questions above
• 90-120 minute sessions conducted in a neutral, comfortable location

For reproductive health topics, it may be appropriate to conduct single-gender focus groups to facilitate more open discussion of sensitive issues.

Ethical Considerations

Reproductive health research often involves sensitive topics requiring special ethical considerations:

• Obtain informed consent with explicit confidentiality assurances
• Provide resources for psychological support if discussions trigger distress
• Ensure privacy during data collection and secure storage of recordings and transcripts
• Obtain approval from relevant institutional review boards or ethics committees

Figure 1: Item Generation Workflow for TPB Questionnaires

Systematic Literature Review Methodology

Search Strategy and Protocol

A systematic literature review complements primary qualitative research by identifying established measures, theoretical frameworks, and previously documented beliefs related to the reproductive health behavior. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines provide a robust framework for conducting and reporting systematic reviews.

Search Protocol:

• Database Selection: Identify relevant electronic databases (e.g., PubMed, PsycINFO, CINAHL, Scopus)
• Search Terms Development: Create comprehensive search strings using Boolean operators that incorporate TPB constructs and reproductive health terminology
• Inclusion/Exclusion Criteria: Establish clear criteria based on population, study design, publication date, language, and relevance to TPB constructs
• Study Selection: Implement a transparent process for screening titles, abstracts, and full-text articles
• Quality Assessment: Evaluate methodological quality of included studies using appropriate tools
• Data Extraction: Systematically extract relevant information on TPB constructs, measures, and belief structures

Data Extraction and Synthesis

Create a standardized data extraction form to systematically capture information from included studies. The extraction should focus on:

• Operational definitions of TPB constructs
• Previously identified behavioral, normative, and control beliefs
• Psychometric properties of existing measures
• Cultural adaptations of TPB questionnaires
• Gaps in existing measurement approaches

The synthesis should integrate findings across studies to identify consistent themes, variations across populations, and methodological limitations in existing research.

Item Development and Formulation

Translating Qualitative Data into Questionnaire Items

The transformation of qualitative findings into preliminary questionnaire items requires systematic content analysis and theme extraction. The process involves:

• Transcription and Familiarization: Transcribe audio recordings and immerse in the data
• Initial Coding: Identify meaningful segments of text related to TPB constructs
• Theme Development: Group codes into themes representing salient beliefs
• Belief Elicitation: Identify the most frequently mentioned beliefs for each TPB construct
• Item Formulation: Create questionnaire items that reflect the elicited beliefs

For example, a study on physicians' strategic behavior under DRG payment systems identified salient beliefs through qualitative methods, then developed items measuring behavior attitudes, subjective norms, and perceived behavioral control [23]. Similarly, in reproductive health research, if qualitative data reveals that "fear of side effects" is a salient behavioral belief regarding contraceptive use, this would be translated into one or more items measuring this belief.

Item Formulation Guidelines

Each TPB construct requires specific item formulation approaches:

Behavioral Beliefs Items should assess both outcome beliefs (e.g., "Using contraception would prevent pregnancy") and outcome evaluations (e.g., "For me, preventing pregnancy is...") using bipolar adjective scales.

Normative Beliefs Items should assess both normative expectations (e.g., "My partner thinks I should use contraception") and motivation to comply (e.g., "Generally speaking, I want to do what my partner thinks I should do").

Control Beliefs Items should assess both control factors (e.g., "How likely is it that you could access contraception if you wanted to?") and perceived power of these factors (e.g., "For me, having easy access to contraception makes using it...").

All items should use clear, simple language appropriate for the target population's literacy level, avoid double-barreled questions, and include appropriate time referents consistent with the behavior definition.

Table 3: Example Item Structure for TPB Reproductive Health Questionnaire

TPB Construct	Item Type	Response Format	Example Item for Contraceptive Use
Behavioral Beliefs	Outcome belief	7-point likely-unlikely scale	"How likely is it that using contraception would give you peace of mind?"
Outcome Evaluation	Evaluation	7-point good-bad scale	"How good or bad would it be to have peace of mind about pregnancy?"
Normative Beliefs	Normative expectation	7-point likely-unlikely scale	"How likely is it that your healthcare provider thinks you should use contraception?"
Motivation to Comply	Compliance motivation	7-point agree-disagree scale	"Generally speaking, how much do you care what your healthcare provider thinks you should do?"
Control Beliefs	Control factor	7-point agree-disagree scale	"How much do you agree that you have the knowledge needed to use contraception correctly?"
Perceived Power	Power perception	7-point easy-difficult scale	"For you, having knowledge about contraception makes using it..."
Behavioral Intention	Behavioral intention	7-point likely-unlikely scale	"How likely are you to use contraception consistently in the next 3 months?"

Content Validity Assessment

Expert Review Process

Content validity assessment ensures that the generated items adequately represent the theoretical constructs and the reproductive health domain. The expert review process should include:

Expert Panel Composition:

• 3-5 subject matter experts in reproductive health
• 2-3 measurement methodologists with TPB expertise
• 1-2 cultural or linguistic experts when appropriate

Review Procedure: Experts independently rate each item on relevance, clarity, and appropriateness using a 4-point scale (e.g., 1=not relevant, 4=highly relevant). Compute the Content Validity Index (CVI) for each item and the entire scale. Items with low CVI scores (<0.78) should be revised or eliminated.

Cognitive Interviewing

Cognitive testing with members of the target population assesses comprehension, retrieval, judgment, and response processes. The protocol includes:

• Think-Aloud Interviews: Participants verbalize their thought processes while answering items
• Verbal Probing: Interviewer asks specific questions about item interpretation and response selection
• Comprehension Testing: Assess understanding of item wording, instructions, and response options

For reproductive health questionnaires, cognitive testing is particularly important for identifying culturally specific interpretations of items and ensuring comfort with sensitive language.

Methodological Integration and Documentation

The final item pool should reflect integration of data from all qualitative methods and the literature review. Create a cross-reference table documenting how each item links to:

• Specific qualitative themes from interviews or focus groups
• Relevant theoretical constructs from TPB
• Similar items from existing literature
• Expert panel recommendations
• Cognitive testing feedback

This documentation provides a clear audit trail for the item development process and strengthens the content validity argument.

Research Reagent Solutions for TPB Questionnaire Development

Table 4: Essential Methodological Tools for TPB Questionnaire Development

Research 'Reagent'	Function	Application in Item Generation	Example Tools/Platforms
Qualitative Data Analysis Software	Facilitates organization, coding, and analysis of qualitative data	Identifying salient beliefs from interviews and focus groups	NVivo, MAXQDA, Dedoose
Systematic Review Management Tools	Supports literature screening, data extraction, and quality assessment	Identifying existing measures and theoretical frameworks	Covidence, Rayyan, EndNote
Content Validity Assessment Metrics	Quantifies expert agreement on item relevance and clarity	Establishing content validity of initial item pool	Content Validity Index (CVI), Cohen's Kappa
Cognitive Testing Protocols	Structured approaches for testing item comprehension	Refining items based on target population feedback	Think-aloud protocols, verbal probing guides
Online Survey Platforms	Facilitates expert review and preliminary testing	Distributing items for expert rating and initial feedback	Qualtrics, REDCap, SurveyMonkey
Statistical Software Packages	Supports initial psychometric analysis	Conducting preliminary item analysis on pilot data	R (lavaan package), SPSS, AMOS [24] [23] [25]

Figure 2: TPB Questionnaire Development Pathway

Systematic item generation through qualitative research and literature review establishes the foundation for developing a theoretically grounded and contextually relevant TPB questionnaire for reproductive health research. By rigorously applying the methodologies outlined in this guide—including appropriate qualitative sampling, systematic data collection, comprehensive literature review, and structured item development—researchers can create instruments that accurately capture the salient beliefs influencing reproductive health behaviors within specific populations. The subsequent phases of questionnaire development (psychometric testing and validation) will build upon this carefully generated item pool to create a robust measurement instrument capable of advancing theory and informing reproductive health interventions.

The Theory of Planned Behavior (TPB) serves as a robust framework for understanding and predicting health behaviors, particularly in sensitive domains such as reproductive health. The accurate measurement of its core constructs—attitude, subjective norms, perceived behavioral control, and intention—is fundamental to developing effective interventions. These constructs can be assessed through two distinct yet complementary approaches: direct measures that capture generalized evaluations, and belief-based (indirect) measures that probe the underlying cognitive foundations of these evaluations. In reproductive health research, where behaviors are often influenced by complex cultural, social, and personal factors, employing both measurement strategies ensures a comprehensive understanding of the determinants of behavior, thereby enhancing the validity and predictive power of the research instrument [26] [27].

The distinction between these measurement types is critical. Direct assessment requires participants to demonstrate their knowledge, skill, or behavior, providing tangible evidence of learning or behavioral倾向. In contrast, indirect assessment relies on participants' self-reporting or reflection on what they believe they have learned or how they perceive a behavior, which acts as a proxy sign of learning or behavioral intention [28]. In the context of TPB, this translates to directly asking about one's attitude toward a behavior versus eliciting the specific beliefs that form that attitude.

Conceptual Foundations: Direct Versus Indirect Measures

Defining Direct and Indirect Measures

In TPB questionnaire design, the constructs are typically operationalized through two primary measurement approaches:

• Direct Measures: These obtain a general assessment of the TPB constructs using broad, summary-type questions. They ask respondents for their overall evaluations without linking them to specific underlying beliefs. For instance, a direct measure of attitude might ask respondents to rate on a semantic differential scale whether they think "discussing sexual health with patients" is "beneficial" versus "harmful" [27]. Direct evidence of student learning is tangible, visible, and measureable and tends to be more compelling evidence of exactly what students have and have not learned [28].

• Indirect (Belief-Based) Measures: These are derived from the TPB's expectancy-value model, which posits that attitudes, subjective norms, and perceived behavioral control are grounded in specific sets of beliefs. Indirect measures therefore consist of two components: a belief strength item and an outcome evaluation (or motivation to comply, for subjective norms) item. The indirect score is typically calculated by multiplying the scores of these two components [27]. Indirect evidence tends to be composed of proxy signs that students are probably learning [28].

The following diagram illustrates the structural relationship between direct and indirect measures within the TPB framework:

Comparative Analysis of Measurement Approaches

Table 1: Comparison of Direct and Belief-Based (Indirect) Measures in TPB Questionnaires

Aspect	Direct Measures	Belief-Based (Indirect) Measures
Definition	Broad, general assessments of TPB constructs [27]	Specific evaluations of underlying beliefs and their perceived outcomes [27]
Measurement Focus	Overall evaluation of the construct (e.g., "Overall, is this behavior good or bad?")	Component elements that form the construct (e.g., specific behavioral beliefs and outcome evaluations)
Typical Format	Semantic differential scales or Likert-scale items measuring general evaluations [26]	Multi-item measures assessing belief strength and outcome evaluation/motivation to comply separately
Data Analysis	Direct scores used in regression and structural equation models	multiplicative composites (belief strength × outcome evaluation) used in analyses
Primary Advantage	Efficient, provides summary evaluation, strong predictive validity	Reveals specific cognitive foundations, more informative for intervention design
Primary Limitation	Less informative for designing targeted interventions	More complex to administer and score, potential psychometric issues with multiplicative composites

Methodological Framework for Instrument Development

Phase 1: Qualitative Foundation and Item Generation

The development of a psychometrically sound TPB questionnaire requires a rigorous qualitative foundation to ensure the instrument's relevance to the target population and behavior. This is particularly crucial in reproductive health research, where cultural sensitivities and contextual factors significantly influence behavioral determinants. A recommended approach involves:

• Conducting Directed Content Analysis: Perform in-depth interviews or focus group discussions with the target population, using the TPB constructs as a preliminary framework. For instance, a study on medical staff's intention to discuss sexual issues with postmenopausal women conducted 27 interviews (13 midwives and 14 general practitioners) until theoretical saturation was reached [27].

• Developing a Comprehensive Codebook: Transcribe interviews verbatim and analyze them using qualitative content analysis. In the aforementioned study, this process yielded 226 codes, 54 sub-categories, and 18 categories, which were classified under the themes of attitude, perceived behavioral control, intention, and behavior [27].

• Applying the TACT Principle: Precisely define the Target, Action, Context, and Time for the behavior being investigated to ensure all questionnaire items are behaviorally specific and contextually appropriate [26].

Phase 2: Constructing Direct Measures

Direct measures should be constructed to capture global assessments of each TPB construct. The following table provides exemplar items for reproductive health contexts:

Table 2: Exemplar Direct Measures for TPB Constructs in Reproductive Health Research

TPB Construct	Measurement Definition	Item Format	Example Item for Reproductive Health Context	Response Scale
Attitude	Overall evaluation of performing the behavior	Semantic differential	"For me to discuss contraceptive options with my patients is:"	Harmful 1-2-3-4-5-6-7 Beneficial [27]
Subjective Norm	Perception of social pressure from important referents	Likert scale	"People who are important to me think I should discuss sexual health issues with postmenopausal women."	Strongly disagree (1) to Strongly agree (7) [26]
Perceived Behavioral Control	Perceived capability and autonomy to perform the behavior	Likert scale	"I am confident that I can raise sexual health issues with patients if I want to."	Strongly disagree (1) to Strongly agree (7) [26]
Intention	Motivation and commitment to perform the behavior	Likert scale	"I intend to discuss sexual issues with postmenopausal women in the next month."	Strongly disagree (1) to Strongly agree (7) [26]
Behavior	Self-reported performance of the behavior	Frequency or likelihood	"How often have you initiated conversations about sexual issues with postmenopausal women in the past month?"	Never (1) to Very often (7) [27]

Phase 3: Constructing Belief-Based (Indirect) Measures

Indirect measures operationalize the expectancy-value framework underlying TPB constructs. Each indirect measure comprises two components that are typically multiplied to create a composite belief score.

Table 3: Structure of Belief-Based (Indirect) Measures in TPB Questionnaires

TPB Construct	Belief Component	Evaluation Component	Example Item Pair for Reproductive Health Context
Attitude	Behavioral Belief (Strength)	Outcome Evaluation	Belief: "Discussing sexual issues with postmenopausal women would improve my patient's quality of life."Evaluation: "Improved patient quality of life is:"Response: Extremely unlikely (1) to Extremely likely (7) / Extremely unimportant (1) to Extremely important (7) [27]
Subjective Norm	Normative Belief (Strength)	Motivation to Comply	Belief: "My clinical supervisor thinks that I should discuss sexual health with postmenopausal women."Motivation: "Generally speaking, I want to do what my clinical supervisor thinks I should do."Response: Extremely unlikely (1) to Extremely likely (7) / Strongly disagree (1) to Strongly agree (7) [27]
Perceived Behavioral Control	Control Belief (Strength)	Perceived Power	Belief: "Lack of private consultation space would make it difficult for me to discuss sexual issues with patients."Power: "If I lacked private consultation space, it would prevent me from discussing sexual issues with patients."Response: Strongly disagree (1) to Strongly agree (7) / Strongly disagree (1) to Strongly agree (7) [27]

Validation Protocols and Psychometric Testing

Content Validity Assessment

Establishing content validity is a critical step in ensuring that a TPB questionnaire adequately measures the intended constructs. A systematic approach involves:

• Expert Panel Evaluation: Convene a panel of 5-10 content experts with publications and expertise in the relevant field (e.g., medical education and professionalism for a professionalism questionnaire) [26]. Experts evaluate each item for relevance and clarity using a 4-point rating scale (1 = not relevant to 4 = highly relevant) [26].

• Quantitative Analysis: Calculate the Item-level Content Validity Index (I-CVI) for each item by dividing the number of experts giving a rating of 3 or 4 by the total number of experts. The Scale-level Content Validity Index (S-CVI) is calculated by averaging the I-CVIs across all items. A well-validated questionnaire on professionalism reported I-CVI scores of 0.9-1 for relevance and 0.7-1 for clarity [26].

• Qualitative Analysis: Analyze free-text comments from experts to refine item wording. For example, in one validation study, experts suggested replacing choices such as "worthless/worthwhile" with "not important/important" and adding social media in the construct of subjective norms [26].

Structural and Statistical Validation

After establishing content validity, the questionnaire should undergo psychometric testing to evaluate its structural properties:

• Pilot Testing: Administer the questionnaire to a sample of 10-30 individuals from the target population to assess comprehensibility, completion time, and response patterns [27].

• Reliability Analysis: Calculate internal consistency reliability using Cronbach's alpha for each construct. For example, in a validated TPB questionnaire on professional behaviors, Cronbach's alpha values ranged from 0.83 to 0.89 for various subscales [27]. Test-retest reliability should also be assessed by administering the questionnaire twice to the same participants after an appropriate interval (e.g., 2 weeks).

• Construct Validity: Perform Confirmatory Factor Analysis (CFA) to verify the hypothesized factor structure. A well-fitting TPB model should demonstrate adequate fit indices, such as Adjusted Goodness-of-Fit Index (AGFI) ≥ 0.89, Root Mean Square Error of Approximation (RMSEA) ≤ 0.07, and Comparative Fit Index (CFI) ≥ 0.9 [27].

Implementation in Reproductive Health Research

Application in Intervention Design and Evaluation

The combination of direct and belief-based measures provides powerful insights for designing and evaluating reproductive health interventions. For instance, a randomized controlled trial assessing a TPB-based educational intervention on sexual and reproductive health among high school girls in Tehran demonstrated significant improvements in both direct measures of TPB constructs and subsequent behaviors [3].

The intervention, which included training sessions with booklets for adolescents and workshops for parents, resulted in statistically significant improvements in attitude (difference = 16.8; 95% CI: 15.3, 18.3), subjective norms (16.4; 95% CI: 14.83 to 18.11), perceived behavioral control (18.0; 95% CI: 16.6, 19.4), and behavior (18.5; 95% CI: 16.8, 20.2) in the experimental group compared to the control group [3]. The belief-based measures would have provided specific targets for the intervention content by identifying the underlying beliefs that needed modification.

Essential Research Reagents and Tools

Table 4: Essential Research Reagents and Tools for TPB Questionnaire Development

Research Tool	Function in TPB Research	Application Example
Qualitative Interview Guides	Eliciting salient beliefs for the target population	Semi-structured guides with questions about advantages/disadvantages, referents, and facilitators/barriers [27]
Content Validation Forms	Assessing item relevance and clarity	Structured forms with 4-point rating scales for expert evaluation [26]
Statistical Software (SPSS, AMOS)	Conducting psychometric analyses and CFA	Analyzing reliability, validity, and model fit indices [27]
Online Survey Platforms	Administering questionnaires to participants	Distribiting the TPB questionnaire to medical staff or target population [3]
Ethics Approval Documentation	Ensuring research compliance	Obtaining approval from institutional ethics committees [3] [27]

The meticulous design of both direct and belief-based (indirect) measures is paramount to developing a theoretically grounded and psychometrically sound TPB questionnaire in reproductive health research. The dual-measurement approach provides complementary advantages: direct measures offer efficient summary evaluations with strong predictive validity, while belief-based measures uncover the specific cognitive foundations that can be targeted in interventions. The rigorous validation process—encompassing qualitative development, content validation, and psychometric testing—ensures that the instrument accurately captures the TPB constructs and produces reliable data to explain and predict complex reproductive health behaviors. This methodological rigor ultimately enhances the development of effective, theory-based interventions in this critical health domain.

The theory of planned behavior (TPB) provides a robust conceptual framework for understanding and predicting health behaviors, making it particularly valuable in reproductive health research. Within this framework, precise measurement of constructs such as attitudes, subjective norms, and perceived behavioral control is paramount. This technical guide establishes evidence-based protocols for developing Likert-style and multiple-choice questions that ensure reliable, valid, and scientifically rigorous data collection in TPB-based reproductive health studies. The strategic design of these measurement instruments directly impacts the accuracy of behavioral intention prediction and subsequent intervention effectiveness.

Research demonstrates that TPB-based educational interventions can significantly improve critical outcomes in reproductive health. For instance, a randomized controlled trial with high school girls showed significant improvements in attitude (difference=16.8; 95% CI: 15.3, 18.3), subjective norms (16.4; 95% CI=14.83 to 18.11), and perceived behavioral control (18.0; 95% CI: 16.6, 19.4) following a targeted intervention [3]. Similarly, a 2023 cross-sectional study on preconception care found that behavior was directly and positively influenced by intention (b = 0.33; CI 95% =0.22 to 0.45; P = 0.001), perceived behavioral control (b = 0.23; CI 95% =0.12 to 0.35; P = 0.001), and attitude (b = 0.22; CI 95% =0.11 to 0.36; P = 0.001) [16]. These findings underscore the critical importance of precise measurement in capturing theoretical constructs and their relationships.

Likert Scale Design Fundamentals

Scale Structure and Response Options

Likert scales present respondents with multiple response options along a continuum, allowing for the quantification of opinions, attitudes, and experiences [29]. These scales transform subjective perceptions into interval data suitable for statistical analysis, enabling researchers to move beyond simple binary measurements.

The strategic selection of response points balances granularity with respondent cognitive load:

• 5-point scales are widely used and generally provide sufficient discrimination while maintaining simplicity [29] [30]
• 7-point scales may offer enhanced sensitivity for capturing nuanced attitudes and are thought to more accurately reflect a respondent's true evaluation [29]
• 4-point forced-choice scales eliminate neutral options and can be valuable when researchers want to prevent non-committal responses, particularly in contexts where social desirability bias is a concern [30]

The table below summarizes key considerations for determining optimal scale length:

Table 1: Likert Scale Length Considerations

Scale Points	Advantages	Disadvantages	Ideal Use Cases
4-point	Prevents fence-sitting; cleaner analysis	May force artificial choices; misses neutral opinions	Polarizing topics; forced direction
5-point	Balanced detail and simplicity; familiar to respondents	Limited nuance between options	General surveys; mixed education levels
7-point	Enhanced measurement sensitivity; captures subtle differences	Increased cognitive load; potential analysis complexity	Detailed attitude assessment; educated populations

Scale Direction and Polarity

The strategic application of unipolar and bipolar scales significantly impacts data quality:

• Unipolar scales measure intensity in one direction (e.g., from "not at all likely" to "extremely likely") and are appropriate for constructs that exist on a single continuum [29]
• Bipolar scales capture opposite spectrums of an attitude (e.g., "completely satisfied" to "completely dissatisfied") and may provide more nuanced understanding of opinions that naturally have opposing poles [29]

Reproductive health research often employs bipolar scales for constructs like satisfaction but may utilize unipolar scales for frequency measurements (e.g., "never" to "always").

Application to Theory of Planned Behavior Constructs

Measuring TPB Dimensions

The TPB framework requires precise operationalization of its core constructs through carefully crafted items. The following table demonstrates how Likert scales can be adapted to measure each TPB component in reproductive health contexts:

Table 2: TPB Construct Measurement with Likert Scales

TPB Construct	Measurement Focus	Sample Likert Item	Response Anchors
Attitude	Beliefs about behavioral outcomes	"In my opinion, using preconception care services would improve my health during pregnancy."	Strongly disagree to Strongly agree
Subjective Norms	Perceived social pressure	"People who are important to me think I should use reproductive health services."	Strongly disagree to Strongly agree
Perceived Behavioral Control	Confidence in performing behavior	"How confident are you that you can access reproductive health services if you want to?"	Not at all confident to Extremely confident
Behavioral Intention	Likelihood of behavior performance	"I plan to use preconception care services in the next 6 months."	Very unlikely to Very likely

In TPB questionnaire development, researchers should include multiple items for each construct to enhance reliability. The study on preconception care utilized five statements each for measuring subjective norms, attitudes, perceived behavioral control, and intentions, all employing a 4-point Likert scale [16].

TPB Questionnaire Development Workflow

The following diagram illustrates the systematic process for developing TPB-based questionnaires with validated Likert scales:

Experimental Protocols and Validation Methods

Questionnaire Validation Methodology

Robust validation procedures are essential for ensuring Likert scales accurately measure TPB constructs in reproductive health contexts. The following protocol outlines a comprehensive validation approach:

Instrument Development Phase:

• Conduct literature review and qualitative studies (e.g., focus group discussions) to identify relevant behavioral beliefs and normative referents [3]
• Develop preliminary items with clear operational definitions for each TPB construct
• Employ expert review to assess content validity and cultural appropriateness

Psychometric Testing Phase:

• Administer the draft questionnaire to a representative sample (minimum 30 participants for initial validation) [16]
• Conduct item analysis to assess discrimination and variability
• Perform reliability testing using Cronbach's alpha (acceptable threshold: >0.60 for exploratory research, >0.70 for confirmatory research) [16]
• Assess construct validity through factor analysis to confirm the theoretical structure

The preconception care study exemplifies this approach, validating their TPB questionnaire through reliability testing that yielded acceptable alpha values for attitude (0.6745), subjective norms (0.6311), perceived behavioral control (0.6011), intention (0.9497), and behavior (0.9497) [16].

Intervention Implementation Protocol

Successful TPB-based interventions require meticulous implementation. The following table summarizes the key components of an effective reproductive health education program based on validated protocols:

Table 3: TPB-Based Intervention Protocol Components

Component	Description	Implementation Method	Duration
Educational Content	Precise information on health topics, consequences, and prevention	Presentation classes (40-45 min), lectures, slideshows	3 months [3]
Attitude Modification	Techniques to address negative beliefs and foster positive attitudes	Role-playing, training animation, story-writing, brainstorming	Integrated sessions [3]
Normative Influence	Addressing social perceptions and engaging influential others	Training sessions for parents, informative booklets	Parallel programming [3]
Behavioral Control	Enhancing confidence and skills for behavioral performance	Story-writing, small group discussion, skill-building activities	Integrated sessions [3]
Follow-up Assessment	Evaluation of intervention effects and behavior change	TPB questionnaire administration, behavioral measures	6 months post-intervention [3]

The randomized controlled trial with high school girls implemented this comprehensive approach through a school-based program that included training sessions with booklets for adolescents and workshops for parents, resulting in significant improvements across all TPB constructs [3].

Data Presentation and Statistical Analysis

Optimal Table Design for Research Publications

Effective data presentation enhances the communication of research findings. The following principles optimize table readability and interpretation:

• Aid comparisons through right-flush alignment of numeric columns and use of tabular fonts to ensure vertical alignment of place values [31]
• Reduce visual clutter by avoiding heavy grid lines and removing unit repetition within cells [31]
• Increase readability by ensuring headers stand out from the body and highlighting statistical significance [31]

These design principles facilitate accurate interpretation of complex statistical relationships between TPB constructs and behavioral outcomes.

Analytical Approaches for Likert Scale Data

Likert scale data requires appropriate statistical techniques that respect its ordinal nature while enabling robust analysis:

• Primary analysis should emphasize the mode (most frequent response) and frequency distributions rather than means, as the numeric values assigned to Likert points are not truly interval [30]
• Data visualization through bar charts effectively communicates response distributions and highlights modal categories [30]
• Advanced modeling techniques, such as structural equation modeling (SEM), can test complex relationships between TPB constructs while accounting for measurement error [16]

The preconception care study exemplifies sophisticated analysis, using SEM to examine both direct and indirect effects of TPB constructs on behavior, revealing significant pathways through intention while controlling for demographic variables [16].

Essential Research Reagents and Materials

The following toolkit summarizes critical components for implementing TPB-based reproductive health research with validated Likert scales:

Table 4: Research Reagent Solutions for TPB Questionnaire Studies

Research Component	Function	Specifications	Validation Metrics
TPB Questionnaire	Measures core constructs (attitude, norms, PBC, intention)	4-7 point Likert scales; 5+ items per construct	Content validity; Cronbach's α >0.60 [16]
Demographic Module	Controls for confounding variables	Age, education, socioeconomic status, reproductive history	Representative sampling
Behavioral Measure	Assesses actual health behavior	Clinical records, self-report, observational data	Correlation with intentions
Intervention Materials	Modifies TPB constructs	Booklets, presentation slides, role-playing scripts	Pilot testing for comprehension
Statistical Analysis Package	Analyzes complex relationships	SEM capability; ordinal data analysis	Model fit indices (CFI, RMSEA, TLI)

The strategic design of Likert-style questions within TPB questionnaires requires meticulous attention to theoretical alignment, psychometric properties, and contextual appropriateness. By implementing the protocols and best practices outlined in this technical guide, researchers can develop robust measurement instruments that accurately capture the cognitive antecedents of reproductive health behaviors. The integration of rigorous scale development methods with the theoretical framework of planned behavior creates a powerful methodology for advancing our understanding of health decision-making and developing effective interventions. As reproductive health challenges continue to evolve, precise measurement through well-constructed Likert scales remains fundamental to generating evidence that informs policy and practice.

In the development of a Theory of Planned Behavior (TPB) questionnaire for reproductive health research, ensuring the validity of the instrument is paramount. The constructs of TPB—attitudes, subjective norms, and perceived behavioral control—represent complex cognitive domains that respondents must accurately comprehend and report [32]. Cognitive interviewing (CI) serves as a critical methodological bridge between theoretical constructs and measurable responses by examining the thought processes respondents use to answer survey questions [33]. This technical guide provides researchers in drug development and public health with evidence-based protocols for implementing cognitive interviews and pilot testing to enhance measurement validity in TPB-focused reproductive health research.

The fundamental premise of cognitive interviewing is that questionnaire response involves a multi-stage cognitive process: respondents must first comprehend the question, then recall relevant information, make a judgment, and finally select a response [34]. When applied to TPB questionnaires in sensitive domains like reproductive health, CI identifies potential disconnects between researchers' assumptions and respondents' interpretations, thereby reducing measurement error and increasing the reliability of collected data [35] [34].

Theoretical Framework: Integrating Cognitive Testing with TPB Questionnaire Development

The Cognitive Basis of TPB Construct Measurement

In TPB research, accurately measuring the theory's core constructs requires careful question design. Attitudes toward reproductive health behaviors assess evaluative judgments, subjective norms capture perceived social pressures, and perceived behavioral control gauges feelings of autonomy and capability [32]. Each construct necessitates precise operationalization through questionnaire items that respondents interpret consistently and as intended.

Cognitive interviewing aligns seamlessly with TPB measurement because it explicitly investigates the cognitive processes underlying question response. As demonstrated in reproductive health research, TPB-based questionnaires often include complex scenarios requiring respondents to consider hypothetical situations, social expectations, and self-efficacy perceptions [3]. Without cognitive testing, questions intended to measure specific TPB constructs may inadvertently tap into different cognitive domains, compromising construct validity and potentially leading to erroneous conclusions about behavioral determinants.

Evidence of Cognitive Testing Efficacy in Health Research

The utility of cognitive interviewing is well-established in healthcare research. In the development of the CAHPS Hospital Survey, cognitive testing revealed that many candidate items failed because respondents lacked necessary information, misunderstood questions inconsistently, or could not make the fine distinctions researchers assumed [35]. Similarly, in social pharmacy research, cognitive interviewing with even small samples (1-2 interviews per iteration) effectively identified both overt and covert problems with comprehension, retrieval, judgment, and response processes [34].

Table 1: Cognitive Interviewing Applications in Health Research

Research Domain	CI Application	Key Findings	Source
CAHPS Hospital Survey	Two rounds of cognitive testing with 31 subjects	Items failed due to respondents lacking required information or making inconsistent interpretations	[35]
Social Pharmacy Research	Cognitive interviews with 2 interviews per iteration	Identified problems with comprehension, retrieval, judgment, and response processes	[34]
Reproductive Health (TPB)	Integration of cognitive probes in questionnaire development	Improved validity of attitude, norm, and perceived behavioral control measures	[3]

Methodological Protocols: Implementing Cognitive Interviews in TPB Research

Cognitive Interviewing Techniques and Procedures

Cognitive interviewing employs two primary methodological approaches: concurrent and retrospective interviewing. In concurrent interviews, participants are requested to "think aloud" as they answer each question, verbalizing their thought processes in real-time. This approach has been found to have no major impact on cognition and provides direct insight into question interpretation [34]. Retrospective interviewing involves participants completing the questionnaire first, followed by probing questions about their interpretation and response strategies after completion [34].

Effective cognitive interviewing utilizes targeted probes to explore specific cognitive processes:

• Comprehension probes: "What do you think this question is asking?" or "What does the term 'subjective norm' mean to you in this context?"
• Recall probes: "How did you remember your feelings about reproductive health services?"
• Judgment probes: "How sure are you about that answer?" or "Was this difficult to judge?"
• Response probes: "How did you choose between 'strongly agree' and 'agree'?" [33]

These probes can be standardized (asked consistently across all interviews) or dynamic (emerging from specific respondent behaviors or statements during the interview). For TPB questionnaires, focusing probes on the core constructs of attitudes, subjective norms, and perceived behavioral control is particularly valuable for ensuring theoretical fidelity.

Sampling and Study Design for Cognitive Interviews

Unlike quantitative studies aiming for statistical representativeness, cognitive interviewing employs purposive sampling to include participants with specific characteristics or experiences relevant to the research domain [33]. For reproductive health TPB research, this might include participants of specific age groups, gender identities, cultural backgrounds, or reproductive experiences that align with the target population.

Sample sizes for cognitive interviews are typically small, ranging from 20-50 respondents, with some research demonstrating value in iterations with as few as 1-2 interviews per questionnaire version [34] [33]. The objective is not quantitative generalization but rather identifying and understanding patterns of interpretation problems.

Table 2: Cognitive Interview Sampling Framework for TPB Reproductive Health Research

Sampling Dimension	Considerations for TPB Reproductive Health Research	Recommended Approach
Sample Size	Enough to detect interpretation patterns but not for statistical power	20-50 participants total, with 5-15 per iteration
Recruitment Criteria	Must represent diversity in the target population for reproductive health behaviors	Purposive sampling based on age, gender, socioeconomic status, reproductive history
Iterative Design	Multiple rounds of testing with revised instruments	2-3 iterations typically sufficient to identify major comprehension issues
Group Comparisons	Potential differences in TPB construct interpretation across subgroups	Ensure inclusion sufficient to compare cognitive processes across key demographics

Analysis and Interpretation of Cognitive Interview Data

The analysis of cognitive interview data typically follows a multi-level approach:

• Level 1: Conducting interviews and documenting observations
• Level 2: Summarizing interview notes and identifying potential problems
• Level 3: Comparing findings across respondents to identify patterns
• Level 4: Comparing across predefined subgroups (e.g., by age, education)
• Level 5: Drawing conclusions about question performance and needed revisions [33]

Analysis should be grounded in the data, using a constant comparative method where each interview is compared to previous ones to identify consistent patterns versus unique interpretations [33]. For TPB questionnaires, special attention should be paid to whether participants' cognitive processes align with the theoretical construct each item is intended to measure.

Experimental Protocols: Integrating Cognitive Interviews with Pilot Testing

Complementary Roles in Questionnaire Development

While cognitive interviewing focuses on question interpretation and cognitive processes, quantitative pilot testing provides statistical evidence of item performance and scale reliability. These approaches are complementary and should be implemented sequentially in TPB questionnaire development: cognitive interviewing first to refine items, followed by pilot testing to evaluate psychometric properties.

In TPB reproductive health research, this integrated approach might involve:

• Initial cognitive interviews with 15-20 participants to identify comprehension problems with attitude, subjective norm, and perceived behavioral control items
• Questionnaire revision based on cognitive interview findings
• Pilot testing with a larger sample (e.g., 50-100 participants) to assess internal consistency, test-retest reliability, and preliminary construct validity
• Additional cognitive interviews if pilot testing reveals unexpected response patterns

Case Example: TPB Reproductive Health Questionnaire

A study examining birth in health facility intention among expecting couples in Tanzania utilized a TPB-based structured questionnaire exploring three main domains: attitudes towards maternal services utilization, perceived subjective norms, and perceived behavior control [32]. While not explicitly detailing cognitive testing methods, the study reported high rates of birth in health facility intention (91.2% of pregnant women and 89.7% of their partners), suggesting possible social desirability bias or question comprehension issues that cognitive interviewing might have identified [32].

In contrast, a TPB-based educational intervention on sexual and reproductive health among high school girls in Iran demonstrated significant improvements in attitude (difference=16.8; 95% CI: 15.3, 18.3), subjective norms (16.4; 95% CI=14.83 to 18.11), and perceived behavioral control (18.0; 95% CI: 16.6, 19.4) [3]. The intervention's success was partly attributed to careful questionnaire development and educational materials pretesting, highlighting the importance of measurement validity in detecting intervention effects.

Research Reagent Solutions for Cognitive Interviewing

Table 3: Essential Materials for Cognitive Interviewing in TPB Research

Item Category	Specific Examples	Function in Cognitive Testing
Participant Recruitment Materials	Screening questionnaires, informed consent forms, demographic surveys	Ensure appropriate purposive sampling; document participant characteristics
Interview Protocols	Standardized probe list, think-aloud instructions, retrospective question guides	Maintain consistency across interviews while allowing flexibility for emergent probing
Data Collection Tools	Audio recording equipment, transcription services, structured note-taking templates	Capture rich qualitative data on cognitive processes during question response
Stimulus Materials	Questionnaire prototypes with varying response formats, visual aids for complex concepts	Test different question wordings and formats to identify optimal presentation
Analysis Resources	Qualitative data analysis software (NVivo, Dedoose), codebook templates, inter-coder reliability checks	Facilitate systematic analysis of interview data and identification of problem patterns

Implementation Framework for Cognitive Interviewing

The following workflow diagram illustrates the integrated process of developing a TPB questionnaire using cognitive interviewing and pilot testing:

Diagram 1: TPB Questionnaire Development Workflow

Advanced Applications: Cognitive Interviewing for Complex TPB Constructs

Testing Complex TPB Scenarios in Reproductive Health

Reproductive health research using TPB often involves complex scenarios requiring respondents to consider sensitive topics, hypothetical situations, and social dynamics. Cognitive interviewing is particularly valuable for these complex constructs. For example, a question measuring subjective norms about contraceptive use might ask respondents to report their perceptions of what "most people important to you" think about this behavior. Cognitive probes can reveal whether respondents are considering family members, peers, healthcare providers, or cultural norms when responding, ensuring alignment with theoretical intent.

Similarly, questions about perceived behavioral control over reproductive healthcare decisions might involve complex assessments of autonomy, resource availability, and self-efficacy. Cognitive interviewing can identify whether respondents are considering internal factors (knowledge, skills) or external barriers (cost, access) when answering, allowing researchers to refine items to better capture the theoretical construct.

Cross-Cultural Adaptation of TPB Questionnaires

In multicultural research contexts, such as studies involving diverse populations in reproductive health, cognitive interviewing is essential for identifying culturally-specific interpretations of TPB constructs. The same questionnaire items about attitudes toward family planning may be interpreted differently across cultural contexts, requiring adaptation to maintain construct equivalence. Cognitive interviewing with participants from each cultural group can identify these interpretation differences and guide culturally-sensitive item revision.

Cognitive interviewing represents a methodological essential in the development of valid and reliable TPB questionnaires for reproductive health research. By systematically investigating the cognitive processes respondents use to interpret and answer questions, researchers can ensure that their instruments accurately measure the theoretical constructs of attitudes, subjective norms, and perceived behavioral control. When integrated with quantitative pilot testing in an iterative development process, cognitive interviewing significantly enhances measurement validity, thereby strengthening the theoretical conclusions and practical implications drawn from TPB research in reproductive health and drug development contexts.

The implementation of cognitive interviews requires careful attention to sampling strategies, interviewing techniques, and analytical approaches, but the investment yields substantial returns in measurement precision. As reproductive health research continues to address complex behavioral challenges, the rigorous application of cognitive interviewing methodologies will remain instrumental in advancing our understanding of the cognitive and social factors underlying health behaviors.

The Theory of Planned Behavior (TPB) provides a robust theoretical framework for designing and evaluating health promotion interventions, particularly for sensitive topics such as reproductive health among adolescent girls. This case study details the development, implementation, and assessment of a TPB-based educational intervention aimed at improving menstrual health, which serves as a critical component of broader reproductive health education. The intervention was designed as a cluster-randomized controlled trial conducted among secondary school girls in Iran, focusing on the constructs of attitude, subjective norms, perceived behavioral control, and intention to engage in health-promoting behaviors [36] [3].

Adolescent girls often face significant challenges related to menstrual health, including inadequate information, cultural taboos, and unhealthy practices that can lead to school absenteeism and adverse health outcomes. In many cultural contexts, including Iran, adolescents do not receive accurate information on menstruation due to specific cultural restrictions, leading to the development of unhealthy behaviors [36]. The TPB-based intervention addressed these challenges through a structured educational program that targeted not only knowledge but also the underlying psychosocial determinants of behavior.

Intervention Design and Theoretical Foundation

Core TPB Constructs and Their Application

The educational intervention was systematically designed to target all core constructs of the Theory of Planned Behavior, with an additional focus on parental influence as a modifying factor. Perceived parental control was incorporated as an additional construct to the original TPB model, recognizing the significant influence of parents in the socio-cultural context of the study [36] [3]. The intervention components were mapped to specific TPB constructs to ensure comprehensive targeting of behavioral determinants.

The conceptual framework below illustrates the relationships between the TPB constructs targeted by the intervention:

Educational Content and Pedagogical Approaches

The intervention comprised seven 2-hour educational sessions implemented over three months, with a six-month follow-up assessment to evaluate sustained effects [36]. The educational content was comprehensive and addressed multiple dimensions of pubertal and menstrual health, with specific pedagogical approaches tailored to different TPB constructs.

Table: Educational Session Components and Their Alignment with TPB Constructs

TPB Construct Targeted	Educational Content	Teaching Methods
Knowledge & Attitude	Importance of adolescence; physical changes; menstrual health; health habits; prevention of premenstrual symptoms	Lectures, slide presentations, educational animations, question and answer sessions
Subjective Norms	Opinions and beliefs of parents, classmates, and friends regarding menstrual health	Brainstorming, story-writing, small-group discussions, educational workshops for parents
Perceived Behavioral Control	Ability and control of behavior related to menstrual health; hygiene practices during puberty	Small-group discussions, skill-building activities, problem-solving exercises
Parental Control	Parental support and monitoring of issues related to menstrual health	Separate educational workshops for parents, informational booklets
Behavioral Intention & Skills	Preventive measures and menstrual health promotion skills	Practical demonstrations, role-playing, development of personal plans for change

The educational strategies were designed using a multi-component approach that engaged not only the adolescents but also their parents and teachers, recognizing the importance of multiple influencers in promoting health behavior change [36] [37]. This approach aligns with implementation science evidence suggesting that multicomponent interventions are more effective than single-component programs in improving outcomes for adolescent girls [37].

Experimental Protocol and Methodological Framework

Study Design and Participant Recruitment

The research employed a cluster-randomized controlled trial design with multistage random sampling to select participants. The study included 578 secondary school girls aged 12-16 years, with 289 participants each in the intervention and control groups [36] [3]. The sampling procedure followed a structured approach to ensure representativeness and minimize contamination between groups.

The flowchart below illustrates the participant flow through the stages of the cluster-randomized controlled trial:

Intervention Implementation Protocol

The active intervention phase spanned three months, with a structured protocol ensuring consistent implementation across all intervention groups:

• Session Frequency and Duration: Seven bi-weekly sessions, each lasting 2 hours
• Educator Training: Teachers and facilitators received specialized training in communication skills to support health behavior change
• Educational Materials: Development and distribution of age-appropriate booklets and visual aids
• Parental Involvement: Separate workshops for parents to align home and school messaging
• Booster Activities: Supplemental activities to reinforce key concepts after the main intervention

The control group continued with their normal schooling without additional menstrual health education during the intervention period, though they were offered the educational materials after the study completion to ensure ethical standards [36].

Data Collection and Measurement Instruments

Data were collected at baseline (pre-intervention) and six months post-intervention using a researcher-made questionnaire that demonstrated strong psychometric properties. The questionnaire was developed through a rigorous process including literature review, qualitative studies with focus group discussions, and expert validation [36] [3].

The instrument encompassed three main sections:

• Demographic information: Age, parental education and employment, economic status
• Menstrual and pubertal health knowledge: 10 items with true/false/don't know response options
• TPB constructs: 42 items measuring attitudes (11 items), subjective norms (7 items), perceived behavioral control (6 items), behavioral intentions (6 items), perceived parental control (4 items), and behavior (8 items)

All items used a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Total scores were converted to a 100-point scale for easier interpretation and comparison. The questionnaire demonstrated excellent internal consistency (Cronbach's α = 0.92 overall, ranging from 0.96 to 0.97 for different TPB constructs) and test-retest reliability (r = 0.82) [36].

Quantitative Outcomes and Statistical Analysis

Intervention Effects on TPB Constructs

The educational intervention resulted in significant improvements across all TPB constructs in the intervention group compared to the control group. The table below summarizes the mean score differences between groups at the 6-month follow-up:

Table: Changes in TPB Construct Scores Following Educational Intervention

TPB Construct	Baseline Scores (Intervention/Control)	Post-Intervention Scores (Intervention/Control)	Mean Difference (95% CI)	P-value
Knowledge	52.4/51.8	78.3/53.6	16.5 (14.9-18.1)	<0.001
Attitude	54.2/53.9	80.1/55.3	16.8 (15.3-18.3)	<0.001
Subjective Norms	53.7/53.5	79.2/54.8	16.4 (14.8-18.1)	<0.001
Perceived Behavioral Control	52.9/52.6	79.5/54.5	18.0 (16.6-19.4)	<0.001
Perceived Parental Control	51.8/52.1	78.9/53.9	17.0 (15.1-18.9)	<0.001
Behavioral Intention	53.1/52.8	80.2/53.7	18.4 (16.8-20.0)	<0.001
Behavior	51.5/51.9	79.8/53.2	18.5 (16.8-20.2)	<0.001

All scores were converted to a 100-point scale for comparison. The intervention group demonstrated statistically significant improvements across all measured constructs, with effect sizes indicating educationally meaningful changes [36] [3].

Statistical Analysis Approach

The data analysis employed sophisticated statistical methods to account for the clustered nature of the data:

• Multilevel modeling was used to investigate associations between measurements in the two groups
• Two-level multilevel analysis compared baseline measures between groups
• Three-level multilevel analysis assessed changes within each group over time
• Interaction effects between groups and time were included in the models
• Analysis of covariance verified the between-group differences while controlling for potential confounding variables

The statistical significance was set at p<0.05, and the models accounted for the intraclass correlation coefficients resulting from the cluster sampling design [36].

Research Reagents and Measurement Tools

Essential Research Materials and Instruments

The successful implementation and evaluation of the TPB-based intervention required several carefully developed research tools and materials:

Table: Essential Research Reagents and Measurement Tools for TPB-Based Interventions

Tool/Reagent	Specifications	Application in Research
TPB Questionnaire	42-item instrument with 5-point Likert scales; Cronbach's α=0.92; test-retest reliability r=0.82	Measurement of TPB constructs (attitude, subjective norms, perceived behavioral control, intention, behavior) and knowledge
Educational Booklets	17-page full-color booklet entitled "Training 12-16-year-old Adolescents at a Turning Point in Life" at middle-school reading level	Distribution to adolescents and parents to reinforce session content and provide accurate reference material
Session Presentation Materials	Slide presentations, educational animations, visual aids tailored to adolescent comprehension level	Delivery of standardized educational content across all intervention sessions
Parent Workshop Materials	2-hour workshop curriculum addressing parental concerns and information needs	Engaging parents as supportive agents for behavior change and addressing intergenerational norms
Data Collection Platform	Paper-based questionnaires with digital data entry using SPSS version 23	Standardized data collection and management across multiple schools and time points

The development of these research reagents followed a rigorous process including literature review, qualitative studies with the target population, and expert validation to ensure cultural appropriateness and scientific validity [36] [3].

Discussion and Implementation Recommendations

Interpretation of Key Findings

The significant improvements across all TPB constructs demonstrate the effectiveness of theory-based interventions in addressing sensitive health topics among adolescent girls. The addition of perceived parental control as an extended construct beyond the original TPB model proved valuable in this cultural context, highlighting the importance of adapting theoretical frameworks to specific populations and settings [36] [3].

The intervention's success can be attributed to several key factors:

• Theory-based design that systematically addressed behavioral determinants
• Multi-component approach engaging adolescents, parents, and teachers
• Age-appropriate educational materials that respected cultural sensitivities
• Sufficient intervention intensity with seven sessions allowing comprehensive coverage
• Skills-building components that enhanced self-efficacy and perceived behavioral control

Recommendations for Research and Practice

Based on the findings of this case study, the following recommendations are proposed for researchers and health professionals developing similar interventions:

• Incorporate Multi-level Influences: Engage not only adolescents but also parents, teachers, and other influential figures to create a supportive environment for behavior change [36] [37].

• Ensure Sufficient Intervention Duration: Implement interventions with adequate intensity and follow-up periods to allow for development of skills and reinforcement of messages [37].

• Utilize Validated Measurement Tools: Develop and validate context-specific instruments with strong psychometric properties to accurately assess intervention effects [36] [3].

• Consider Cultural Adaptations: Tailor intervention content and delivery methods to align with cultural norms and address specific population needs [36] [38].

• Include Long-term Follow-up: Assess sustained intervention effects beyond immediate post-intervention measurements to evaluate maintenance of behavior change [36] [38].

This case study provides a replicable model for developing theory-based health interventions for adolescent populations, with particular relevance for sensitive topics in reproductive health. The structured approach to intervention design, implementation, and evaluation can be adapted to various cultural contexts and health topics to promote positive health behaviors among adolescent girls.

Overcoming Common Challenges in TPB Instrument Design and Measurement

Addressing Social Desirability Bias in Sensitive Topics

Social desirability bias (SDB) represents a fundamental threat to the validity of self-reported data across research domains, particularly in sensitive areas such as sexual and reproductive health. This systematic tendency to respond in a culturally appropriate manner leads to under-reporting of stigmatized behaviors and over-reporting of normative ones [39]. In the specific context of Theory of Planned Behavior (TPB) research on reproductive health, SDB distorts the measurement of core constructs—attitudes, subjective norms, and perceived behavioral control—thereby compromising the theory's predictive validity and the effectiveness of interventions designed based on its findings [10] [6]. The pervasive nature of this problem is evidenced by research reviews indicating that approximately half of all self-reported information may be inaccurate in some way, making SDB a "well-kept open secret" in scientific research [39].

The challenge is particularly acute in reproductive health research employing TPB frameworks, as these studies rely on accurate measurement of precisely those behaviors and intentions most vulnerable to systematic misreporting. When individuals provide socially desirable responses rather than truthful accounts of their behaviors and intentions, the resulting data corrupts the relationships between TPB constructs, leading to flawed conclusions and potentially ineffective interventions [10]. This technical guide provides evidence-based methodologies for identifying, measuring, and mitigating social desirability bias within the specific context of TPB research on sensitive reproductive health topics.

Theoretical Framework: Social Desirability Bias in Theory of Planned Behavior Research

Integration of SDB within the TPB Framework

The Theory of Planned Behavior posits that behavioral intentions are determined by three core constructs: attitudes toward the behavior, subjective norms, and perceived behavioral control [40] [6]. Social desirability bias systematically affects the measurement of each construct through distinct mechanisms, as visualized below:

Figure 1: Integration of Social Desirability Bias within the Theory of Planned Behavior Framework

As illustrated, SDB exerts influence across multiple pathways within the TPB model. When measuring attitudes, respondents may report more favorable evaluations of socially approved behaviors than they actually hold. For subjective norms, individuals may overstate the importance they place on the expectations of others when those expectations align with social conventions. Regarding perceived behavioral control, respondents may systematically misrepresent their self-efficacy beliefs for behaviors carrying social stigma [6]. The cumulative effect of these measurement distortions is compromised prediction of behavioral intentions and actual behavior, fundamentally threatening the theoretical integrity and practical application of TPB research.

Empirical Evidence of SDB in Reproductive Health Research

Substantial empirical evidence demonstrates the operation of SDB in reproductive health contexts. A comprehensive review of sexual behavior research found significant correlations between social desirability measures and self-reports of numerous sensitive behaviors, including number of sexual partners, condom use, receptive anal intercourse, and exposure to pornography [39]. The magnitude of these effects is substantial, with one early study reporting a correlation of 0.70 between the probability of truthful responses and the judged social desirability of various behavioral items [39].

Gender-specific patterns further illuminate the operation of SDB. In rural Malawi, research demonstrated that data collection method influenced reporting differently for males and females. Males were less likely to report ever having had a girlfriend in audio-CASI than in face-to-face interviews (odds ratio: 0.4) but were more likely to report having had sex with a relative or teacher (odds ratio: 3.5) [41]. For females, reports of ever having had a boyfriend or having had sex did not differ significantly between modes, though a small proportion reported sensitive behaviors (sex with relative or teacher) only in the more private audio-CASI setting [41]. These findings highlight how cultural norms create different social desirability pressures across genders, requiring tailored methodological approaches.

Methodological Strategies for Mitigating Social Desirability Bias

Data Collection Modalities: Comparative Effectiveness

Research has evaluated multiple data collection modalities for their capacity to reduce social desirability bias in sensitive topics. The table below summarizes the key methodologies, their theoretical mechanisms, and empirical support:

Table 1: Data Collection Modalities for Reducing Social Desirability Bias

Method	Theoretical Mechanism	Implementation	Empirical Support
Audio Computer-Assisted Self-Interviewing (ACASI)	Eliminates interviewer effects through direct computer interface; provides standardized questioning and privacy [41]	Respondents listen to questions via headphones and enter responses electronically without interviewer involvement [41]	Increased reporting of socially stigmatized behaviors (sex with relatives/teachers) in Malawi study [41]; mixed results in Zambia [42]
Face-to-Face Interviews (FTF)	Enables rapport building and trust establishment; allows for clarification [42]	Trained interviewers administer questions directly; requires strong interviewer skills and neutral demeanor [42]	Reduced discomfort over repeated interviews in Zambia; potential for interviewer effects and social desirability bias [41]
Mixed-Mode Approaches	Combines benefits of rapport building (FTF) with privacy for sensitive items (ACASI) [42]	Less sensitive questions administered FTF; sensitive questions via ACASI in same session [42]	Some participants reported inconsistencies between modes (denying behaviors in ACASI while admitting in FTF); requires careful implementation [42]

The effectiveness of these modalities is context-dependent, influenced by cultural factors, the specific population being studied, and the nature of the sensitive behaviors being investigated. In the Malawi study, ACASI increased reporting of stigmatized behaviors particularly for males, suggesting gender-specific effects [41]. Conversely, research in Zambia found that some adolescents actually reported more sensitive information in face-to-face interviews with trusted researchers than in ACASI, highlighting the importance of relational factors [42].

Questionnaire Design and Interview Techniques

Strategic questionnaire design and interviewing techniques can significantly reduce SDB by minimizing the cognitive and social barriers to truthful responding:

• Indirect Questioning and Contextual Framing: Rather than asking directly about sexual behaviors, framing questions within the context of romantic relationships makes them less confrontational and reduces discomfort [42]. Qualitative research with adolescent girls in Zambia found that "efforts by interviewers to signal that they did not judge the participants for their behavior and increased familiarity with the interviewer reduced discomfort over time" [42].

• Terminology and Wording: Avoiding overly clinical or explicit descriptive words while ensuring clear understanding of terms reduces embarrassment. Participants in Zambia expressed discomfort with direct terminology and preferred less explicit phrasing [42].

• Question Order and Sequencing: Placing sensitive questions later in the interview instrument, after rapport has been established and less threatening topics have been discussed, can improve accuracy [41] [42].

• Familiarity and Trust Building: Repeated interviews with the same interviewer, when possible, allows trust to develop. Teachers and research assistants in Zambia noted that established relationships significantly improved participants' comfort with sensitive topics [42].

Experimental Protocols for Measuring and Controlling SDB

Incorporating Social Desirability Scales

To quantify and statistically control for SDB, researchers should incorporate standardized social desirability scales directly into research instruments. Paulhus's (1984) approach of measuring two components—self-deception (unconscious positive self-presentation) and impression management (conscious deception of others)—provides nuanced assessment [39]. In sexual behavior research, significant negative correlations have been found between impression management scores and reports of unrestricted sexual attitudes, fantasies, and experiences, even after controlling for personality and conservatism factors [39].

Implementation Protocol:

• Scale Selection: Choose validated social desirability scales appropriate for the cultural context and population (e.g., Marlowe-Crowne Social Desirability Scale, Paulhus Deception Scales)
• Placement: Position social desirability items throughout the questionnaire rather than in a single block to avoid priming effects
• Scoring: Calculate separate scores for self-deception and impression management subscales when possible
• Analysis: Include social desirability scores as covariates in statistical models predicting behavioral intentions and self-reported behaviors

Methodological Comparison Experiments

Researchers can implement experimental designs that directly compare reporting patterns across different methodological approaches:

Table 2: Experimental Protocol for Methodological Comparison

Step	Procedure	Implementation Details	Outcome Measures
1. Sampling	Recruit participants from target population; random assignment to conditions	Ensure sufficient power to detect moderate effects; consider gender stratification based on Malawi findings [41]	Baseline equivalence between experimental conditions
2. Condition Implementation	Administer identical core questions through different modalities (e.g., FTF vs. ACASI)	Maintain identical question wording, order, and response options across conditions; train interviewers thoroughly [41]	Modality-specific reports of sensitive behaviors and intentions
3. Validation Measures	Include known-groups validation or objective measures where possible	Incorporate biological measures, partner reports, or diary methods as feasible for validation [39]	Discordance between self-reports and validation measures
4. Process Evaluation	Collect qualitative feedback on participant experience	Conduct follow-up interviews or focus groups about comfort level, perceived privacy, and reasons for possible misreporting [42]	Qualitative insights into methodological influences on reporting

The Malawi Schooling and Adolescent Study implemented such a design, randomly assigning participants to either ACASI or face-to-face interviews and comparing reports of sensitive sexual behaviors. The results demonstrated clear modality effects, particularly for behaviors with strong social desirability pressures [41].

Table 3: Research Reagent Solutions for Social Desirability Bias Mitigation

Tool/Resource	Function	Implementation Considerations
ACASI Software	Enables private self-administered data collection for sensitive items	Requires technological infrastructure and participant digital literacy; needs pilot testing [41] [42]
Social Desirability Scales	Quantifies individual tendency toward socially desirable responding	Must be validated in specific cultural context; can be used as covariate in analyses [39]
Interviewer Training Protocols	Standardizes interviewer demeanor to minimize judgmental cues	Focus on neutral demeanor, confidentiality assurance, and building rapport [42]
Cognitive Testing Materials	Identifies problematic question wording before full deployment	Uses think-aloud protocols to detect questions that evoke social desirability concerns [42]
Methodological Comparison Designs	Measures magnitude of SDB across data collection approaches	Requires larger samples but provides empirical evidence of SDB effects [41]

Addressing social desirability bias in TPB research on reproductive health requires meticulous attention to methodological design across multiple dimensions. No single approach eliminates the problem entirely, but through strategic combination of technological solutions, careful questionnaire design, interviewer training, and statistical controls, researchers can significantly mitigate its distorting effects on research findings. The integrity of TPB research depends on accurate measurement of its core constructs, making the systematic addressing of SDB not merely a methodological concern, but a fundamental theoretical necessity.

Understanding health behaviors requires moving beyond simple knowledge assessment to measure the underlying psychological constructs that directly influence behavioral decisions. While knowledge is often a necessary prerequisite for behavior change, it is rarely sufficient to explain or predict complex health behaviors. The Theory of Planned Behavior (TPB) provides a robust theoretical framework for identifying and measuring these critical mediators between knowledge and action [3]. According to TPB, behavioral intention—the immediate precursor to behavior—is influenced by three primary constructs: attitudes (positive or negative evaluations of the behavior), subjective norms (perceptions of social pressure), and perceived behavioral control (confidence in one's ability to perform the behavior) [32]. In reproductive health research, accurately measuring these constructs is particularly crucial because behaviors in this domain are often influenced by complex sociocultural factors, personal beliefs, and contextual barriers that extend beyond mere knowledge [3] [32].

The limitation of knowledge-focused assessments becomes evident when considering that individuals may understand health recommendations perfectly yet fail to implement them due to underlying attitudinal barriers, social pressures, or perceived control limitations. Research demonstrates that interventions targeting these underlying constructs can effectively promote health behavior change. For instance, a TPB-based educational intervention for adolescent girls' sexual and reproductive health resulted in significant improvements in attitude, subjective norms, perceived behavioral control, and behavioral intention, ultimately reducing high-risk behaviors [3]. This evidence underscores the importance of developing precise measurement approaches for these underlying constructs to design more effective behavioral interventions in reproductive health and other domains.

Theoretical Framework: The Theory of Planned Behavior

Core Constructs and Their Measurement

The Theory of Planned Behavior provides a systematic framework for understanding the psychological determinants of behavior. At its core, TPB posits that behavioral intention is the most immediate predictor of voluntary behavior, and this intention is itself determined by three fundamental constructs [32]. First, attitude toward the behavior refers to the degree to which a person has a favorable or unfavorable evaluation of the specific behavior in question. This construct encompasses both instrumental (e.g., beneficial-harmful) and experiential (e.g., enjoyable-unpleasant) dimensions. Second, subjective norm reflects the perceived social pressure to perform or not perform the behavior, incorporating both injunctive norms (what important others think one should do) and descriptive norms (what others are actually doing). Third, perceived behavioral control refers to the perceived ease or difficulty of performing the behavior, capturing both internal control factors (skills, abilities) and external control factors (opportunities, barriers) [32].

These constructs operate in concert to influence behavioral intentions, though their relative importance may vary across behaviors and populations. For example, a study on birth in health facility intention in Tanzania found that subjective norms showed a significantly higher mean score among pregnant women compared to their male partners, suggesting gender differences in the influence of social pressure [32]. Meanwhile, perceived behavioral control emerged as a more significant factor among male partners. This highlights the importance of context-specific understanding of these constructs in reproductive health research. The TPB framework can be expanded to include additional constructs relevant to specific domains, such as the inclusion of parental control in adolescent reproductive health studies, which was found to significantly improve intervention outcomes [3].

Visualizing the Theory of Planned Behavior Framework

The following diagram illustrates the relationships between the core constructs of the Theory of Planned Behavior:

Quantitative Evidence: Effectiveness of TPB-Based Interventions

Research across diverse health domains has demonstrated the value of measuring and targeting TPB constructs. The table below summarizes key quantitative findings from intervention studies based on the Theory of Planned Behavior:

Table 1: Quantitative Outcomes of TPB-Based Interventions in Health Research

Study Population	Intervention Type	Attitude Change	Subjective Norms Change	Perceived Behavioral Control Change	Behavioral Intention Change	Behavior Change	Citation
High school girls (12-16 years), Tehran	Educational intervention on sexual/reproductive health	+16.8 points (95% CI: 15.3, 18.3)	+16.4 points (95% CI: 14.83, 18.11)	+18.0 points (95% CI: 16.6, 19.4)	+18.4% (95% CI: 14.8, 18.3)	+18.5 points (95% CI: 16.8, 20.2)	[3]
Expecting couples, rural Tanzania	Cross-sectional survey on facility birth intention	Not significant	Pregnant women: M=30.21, SD=3.928Male partners: M=29.72, SD=4.349 (p<0.049)	Not significant	Pregnant women: 91.2%Male partners: 89.7%	Not measured	[32]
Teachers, Philadelphia	Longitudinal study on EBP implementation	Not measured	Not measured	Not measured	Variance explained in implementation: 3.5-29.0% (depending on measure)	Specific EBP measures explained more variance (29.0% vs 8.6%)	[43]

The significant improvements observed across all TPB constructs following educational interventions demonstrate the malleability of these underlying determinants and their responsiveness to theory-based approaches [3]. The data further reveal that the strength of association between intentions and actual behavior varies substantially based on how intention is measured, with specific, well-defined intention measures explaining significantly more variance in implementation (29.0%) compared to general measures (8.6%) [43].

Measurement Approaches: From Theory to Practice

Instrument Development and Validation

Developing valid and reliable measures of TPB constructs requires systematic approaches that blend qualitative and quantitative methods. The process typically begins with elicitation studies using qualitative techniques such as focus group discussions to identify salient beliefs, normative referents, and control factors specific to the target population and behavior [3]. For example, in developing a reproductive health questionnaire for adolescent girls, researchers conducted eight focus group discussions with 40 participants to inform item development [3]. This qualitative foundation ensures that quantitative measures reflect the actual beliefs, attitudes, and perceptions of the population rather than researcher assumptions.

Based on qualitative findings, researchers develop structured questionnaires with items corresponding to each TPB construct. A comprehensive TPB-based questionnaire typically includes multiple items for each construct using appropriate response formats (e.g., Likert scales, semantic differentials). For instance, the adolescent reproductive health study utilized a 132-item questionnaire measuring six constructs: knowledge, attitude, perceived behavioral control, perceived parental control, behavioral intention, and behavior [3]. The specific formulation of intention items significantly impacts their predictive validity, with measures referring to specific evidence-based practices accounting for substantially more variance in implementation (29.0%) compared to general measures (8.6%) [43].

The Researcher's Toolkit: Essential Methodological Approaches

Table 2: Key Methodological Approaches for Measuring Underlying Constructs

Methodological Approach	Description	Application in TPB Research	Key Considerations
Focus Group Discussions	Qualitative group interviews to identify salient beliefs	Elicitation studies to inform questionnaire development [3]	Typically 6-10 participants per group; requires skilled facilitator
Structured Questionnaires	Quantitative measures with multiple items per construct	Assessing attitudes, norms, perceived control, intentions [3]	Requires psychometric validation; multiple items per construct improve reliability
Longitudinal Designs	Tracking participants over time to predict behavior from intentions	Assessing intention-behavior relationship [43]	Time lag should be appropriate to the behavior; minimizes common method bias
Structural Equation Modeling (SEM)	Multivariate statistical technique for testing complex relationships	Modeling relationships between TPB constructs and behavior [44]	Requires adequate sample size; allows testing of direct and indirect effects
Randomized Controlled Trials	Experimental designs with random assignment	Testing efficacy of TPB-based interventions [3]	Gold standard for causal inference; requires careful manipulation of constructs

Advanced Analytical Approaches

Structural Equation Modeling for TPB Constructs

Structural Equation Modeling (SEM) provides a powerful analytical framework for testing complex models involving TPB constructs and their relationships with behavior. SEM allows researchers to simultaneously estimate both the measurement model (relationships between latent constructs and their indicators) and the structural model (relationships between constructs) [44]. This is particularly valuable in TPB research, where constructs are inherently unobservable (latent) and must be inferred from multiple observed indicators. SEM implemented in R using the lavaan package enables comprehensive testing of the TPB framework, including direct and indirect effects of attitudes, norms, and perceived control on intentions and behavior [44] [45].

The lavaan syntax for specifying SEM models uses intuitive operators to define different model components: the =~ operator defines latent constructs from observed indicators, the ~ operator specifies regression relationships, and the ~~ operator defines covariances [45]. For example, a basic TPB model might specify intention as being regressed on attitude, subjective norm, and perceived behavioral control, with behavior being regressed on intention and perceived behavioral control. SEM provides numerous fit indices (e.g., CFI, TLI, RMSEA) to evaluate how well the hypothesized model reproduces the observed covariance matrix, allowing researchers to test the adequacy of the TPB framework for explaining the data [44] [45].

Causal Inference Approaches

Understanding causal relationships in TPB research requires careful consideration of potential confounding factors. Directed Acyclic Graphs (DAGs) provide a valuable tool for representing assumed causal relationships and identifying potential sources of bias [46]. In DAGs, nodes represent variables, and arrows represent hypothesized causal influences. A key concept in causal inference using DAGs is the backdoor path—non-causal paths between variables of interest that create spurious associations [46]. For example, in studying the effect of education on health behavior, background factors like family environment may create backdoor paths that confound the relationship.

The following diagram illustrates a DAG for a study examining the effect of college education on income, highlighting potential confounding pathways:

Closing backdoor paths through statistical adjustment (e.g., including confounders in regression models) allows for more valid estimation of causal effects. However, conditioning on collider variables (variables influenced by two or more other variables) can introduce bias by opening previously blocked non-causal paths [46]. These causal considerations are essential for drawing valid inferences about the effects of TPB constructs on behavior and for designing effective interventions.

Experimental Protocols and Implementation

Protocol for TPB-Based Educational Intervention

Based on successful implementations in reproductive health research, the following protocol provides a template for developing and evaluating TPB-based interventions:

Objective: To design, implement, and evaluate a Theory of Planned Behavior-based educational intervention to promote health behavior change.

Materials:

• Pre-validated TPB questionnaire assessing knowledge, attitudes, subjective norms, perceived behavioral control, and intentions
• Educational materials (booklets, presentations) targeting key TPB constructs
• Parental consent forms (for adolescent populations)
• Data analysis software (e.g., R with lavaan package for SEM)

Procedure:

• Baseline Assessment: Administer TPB questionnaire to all participants before intervention.
• Randomization: Randomly assign participants to intervention or control groups using appropriate methods (e.g., permuted block randomization with block lengths of 2-4) [3].
• Intervention Delivery:
  • Conduct educational sessions targeting specific TPB constructs:
    • Knowledge: Didactic presentations, lectures, slideshows (40-45 minutes)
    • Attitudes: Role-playing, training animations, story-writing, brainstorming
    • Subjective Norms: Training sessions for parents, informational booklets
    • Perceived Behavioral Control: Story-writing, small group discussions, skill-building activities
  • Distribute age-appropriate educational booklets developed based on formative research
  • For adolescent populations, conduct separate sessions with parents to address parental control constructs
• Follow-up Assessment: Re-administer TPB questionnaire immediately post-intervention and at predetermined follow-up points (e.g., 3 months, 6 months).
• Behavioral Assessment: Measure actual behavior at appropriate time points following intention assessment.

Analysis:

• Compare changes in TPB construct scores between intervention and control groups using analysis of covariance (ANCOVA), adjusting for baseline scores.
• Test the full TPB model using structural equation modeling to examine relationships between constructs.
• Assess mediation effects to determine whether intervention effects on behavior operate through changes in TPB constructs.

Implementation Considerations for Reproductive Health Research

Implementing TPB-based interventions in reproductive health contexts requires special methodological considerations. Research with adolescent populations necessitates parental involvement components, as evidenced by a study that included two-hour workshops for parents about high-risk behaviors and reproductive health issues [3]. The intervention material should be developmentally appropriate; for example, the successful adolescent intervention used a full-color, pictorially rich booklet entitled "Training 12-16-year-old Adolescents at a Turning Point in Life" written at a middle-school reading level [3].

Cultural adaptation is essential when applying TPB across different contexts. In the Tanzanian study on facility birth intention, researchers observed gender differences in how TPB constructs influenced intentions, with subjective norms being more influential for pregnant women and perceived behavioral control more salient for their male partners [32]. This highlights the need for gender-sensitive approaches in reproductive health research. Additionally, reproductive health topics may involve sensitive content requiring careful ethical consideration, including confidentiality protections, appropriate informed consent procedures, and referral systems for participants needing additional services.

Accurately measuring underlying constructs such as attitudes, subjective norms, and perceived behavioral control provides critical insights into the psychological mechanisms driving health behaviors. The Theory of Planned Behavior offers a robust theoretical framework for developing targeted interventions that address these fundamental determinants rather than focusing exclusively on knowledge transfer. The quantitative evidence, methodological approaches, and implementation strategies outlined in this guide provide researchers with practical tools to advance this important work, particularly in the domain of reproductive health where behavioral decisions have profound implications for wellbeing. By moving beyond knowledge to measure and target these underlying constructs, researchers and practitioners can develop more effective interventions to promote health behavior change.

Navigating Low Literacy and Cross-Cultural Adaptation of Questionnaires

In the context of research utilizing the Theory of Planned Behavior (TPB) to investigate reproductive health behaviors, the development and adaptation of questionnaires present significant methodological challenges. These challenges are particularly pronounced when research extends to diverse cultural contexts and populations with varying literacy levels. A questionnaire developed for one cultural or linguistic group cannot be automatically deployed in another without rigorous adaptation, as measurement invariance is essential for valid cross-cultural comparisons [47]. The consequences of inadequate adaptation include construct under-representation, where key dimensions of a concept relevant to the new population are missing, and construct-irrelevant variance, where items are misinterpreted due to cultural or linguistic factors [47]. Within reproductive health research, where TPB is frequently applied to understand behaviors such as preconception care utilization [48], contraceptive use [49], and management of chronic conditions like endometriosis [4], ensuring that questionnaires are both culturally appropriate and comprehensible to populations with low literacy is a critical prerequisite for generating valid and reliable evidence.

Theoretical Foundations: Why Adaptation Matters

The process of questionnaire adaptation is grounded in the fundamental principle of conceptual equivalence—ensuring that the underlying construct (e.g., attitude, subjective norm, perceived behavioral control) holds the same meaning and is manifested similarly across different cultural groups. Without this equivalence, comparisons between groups are meaningless, and interventions based on the research may be ineffective or even harmful.

Research has consistently demonstrated that populations may differ systematically in their responses to survey instruments. For instance, some cultural groups are more likely to use extreme response options on Likert scales, while others tend to avoid them [47]. Furthermore, the meaning or appropriateness of concepts may differ across cultures. A food frequency questionnaire developed for a general population, for example, would lack validity for a specific ethnic minority group unless it included foods particular to that culture [47]. In TPB-based reproductive health research, concepts such as "subjective norms" or "perceived behavioral control" may be influenced by different social actors or barriers in different cultures, necessitating a careful exploration of these constructs during the adaptation process [50] [48].

A Systematic Framework for Cross-Cultural Adaptation

A robust framework for cross-cultural adaptation should integrate established translation methodologies with proactive steps to ensure cultural and conceptual relevance. The following workflow synthesizes recommendations from multiple studies [51] [47] [52] into a coherent, phased approach.

Phase 1: Preparation and Assessment

Before beginning translation, a comprehensive review of the original questionnaire's conceptual foundation is essential. This involves clarifying the intent of each item and the underlying TPB constructs they are designed to measure [52]. An expert panel should be assembled, including researchers, healthcare professionals familiar with the target population, and cultural liaisons. Crucially, this phase should include elicitation studies with the target population to identify salient behavioral, normative, and control beliefs related to the reproductive health topic in the new cultural context [48]. This ensures that the questionnaire captures locally relevant aspects of attitudes, subjective norms, and perceived behavioral control, potentially identifying missing dimensions that need to be added to the original instrument [47].

Phase 2: Systematic Translation

The translation phase aims to achieve semantic equivalence, where the meaning of each item is preserved in the target language. The recommended process, based on guidelines from the World Health Organization and Brislin's translation model, involves multiple steps [51] [52]:

• Forward Translation: Two independent bilingual translators, fluent in both the source and target languages and familiar with both cultures, produce two initial translations.
• Synthesis: The two forward translations and the original developer(s) discuss discrepancies to create a reconciled version.
• Back Translation: One or two different translators, blinded to the original questionnaire, translate the reconciled version back into the source language.
• Review by Original Developers: The original authors compare the back-translation to the source questionnaire to identify and resolve any conceptual discrepancies.

Phase 3: Cultural and Cognitive Validation

This phase moves beyond linguistic equivalence to ensure the questionnaire is culturally appropriate and comprehensible.

• Health Professional Review: A panel of local healthcare professionals evaluates the conceptual equivalence and relevance of the items in the local clinical and cultural context [51].
• Cognitive Interviews: A critical step for low-literacy populations, this involves testing the questionnaire with individuals from the target group. Interviewers use probing techniques (e.g., "What were you thinking when you answered that question?") to understand how respondents interpret each item and their reasoning for selecting answers [47] [52]. This can reveal misunderstandings of terms or phrases.
• Lay Panel Debate: A group discussion with target population members can use a Nominal Group Technique to reach consensus on the best wording and cultural suitability of items [51].

Phase 4: Psychometric Validation

The final phase involves quantitatively testing the adapted questionnaire's reliability and validity.

• Pilot Testing: The adapted instrument is administered to a sample from the target population.
• Psychometric Analysis: Statistical analyses are conducted to assess internal consistency (e.g., Cronbach's alpha), test-retest reliability, and construct validity [49] [48] [52]. Confirmatory Factor Analysis (CFA) is often used to verify that the factor structure of the TPB constructs (attitude, subjective norms, perceived behavioral control, intention) is maintained in the new cultural context [52].

Special Considerations for Low-Literacy Populations

Adapting questionnaires for populations with limited literacy requires specific, targeted strategies beyond standard cross-cultural adaptation.

Table 1: Strategies for Low-Literacy Questionnaire Adaptation

Strategy	Implementation	Exemplar Study
Simplify Language	Use short, simple sentences and concrete, familiar words. Avoid abstractions, metaphors, and complex grammar.	Terms like "blood sugar drop" were misunderstood and required simplification [47].
Modify Response Scales	Use fewer response options; employ pictorial scales (e.g., smiley faces), visual aids, or yes/no formats when appropriate.	A TPB questionnaire on endometriosis used a 2-point (yes/no) scale for subjective norms and behavioral intention [4].
Cognitive Interviewing	Use verbal probing to identify misunderstood terms and refine items based on feedback.	Probing questions like "What does this word mean to you?" revealed misunderstandings [47] [52].
Alternative Modalities	Consider interviewer administration, audio-assisted, or interactive voice response (IVR) formats to bypass reading demands.	Face-to-face interviews were used for data collection in a study in Ethiopia to ensure comprehension [48].

Key modifications often involve rephrasing items to be more direct and concrete. For instance, an item on the Beck Depression Inventory was modified for low-literacy populations after it was found to be misunderstood [47]. Similarly, in reproductive health research, the phrase "bothered by" in the CES-D scale was interpreted as having a physical complaint by Spanish-speaking patients with diabetes, indicating a need for rephrasing [47]. For TPB questionnaires, this might involve reframing abstract questions about "attitudes" into more concrete questions about perceived advantages and disadvantages of a specific behavior.

Case Studies in TPB Reproductive Health Research

Preconception Care in Ethiopia

A study assessing women's intention to use preconception care (PCC) in Southern Ethiopia developed a TPB-based questionnaire through a rigorous process [48]. Before creating the tool, an elicitation study was conducted via in-depth interviews with 20 target group members to identify locally salient beliefs. This informed the development of a 132-item questionnaire measuring TPB constructs. The instrument was translated, pretested, and its reliability confirmed (Cronbach's alpha > 0.7 for all constructs). The study successfully identified that perceived behavioral control (β=0.263) was the strongest predictor of intention to use PCC, followed by attitude and subjective norms, demonstrating the utility of a well-adapted instrument [48].

Sexual Health Promotion for Adolescent Females in China

Researchers in China developed a Sexual Health Promotion Scale (SHPS) for adolescent females based on the TPB [49]. The scale development involved literature analysis, in-depth interviews, and expert consultation, culminating in a 22-item tool. The psychometric evaluation with 388 adolescents showed the scale had a robust four-factor structure (e.g., sexual health knowledge, contraceptive attitude), excellent internal consistency (Cronbach's α=0.904), and good test-retest reliability (0.882). The content validity index was 0.926, and Confirmatory Factor Analysis indicated a good model fit (GFI=0.962, RMSEA=0.048), validating the instrument for the target population [49].

Endometriosis Reproductive Health in Iran

An educational intervention for women with endometriosis in Iran utilized a TPB questionnaire that was carefully developed and validated [4]. The researchers assessed face validity by having 10 patients review the questionnaire for interpretability, and content validity through expert reviews, calculating Content Validity Ratio (CVR) and Index (CVI). The final instrument demonstrated high internal consistency (α=0.91) and test-retest reliability (r=0.82). The intervention, delivered using this validated tool, led to significant improvements in TPB construct scores and the overall reproductive health of participants in the intervention group compared to controls [4].

Table 2: Psychometric Properties from Case Studies

Study & Context	Questionnaire Focus	Sample Size for Validation	Reliability (Cronbach's α)	Key Validity Indicators
Adolescent Sexual Health (China) [49]	Sexual Health Promotion	388	0.904	CVI: 0.926; CFA: GFI=0.962, RMSEA=0.048
Preconception Care (Ethiopia) [48]	PCC Utilization	415	> 0.70 for all constructs	Explained variance in intention via multiple linear regression
Endometriosis Health (Iran) [4]	Reproductive Health	71	0.91	CVI > 0.79; CVR > 0.62

The Scientist's Toolkit: Key Reagents for Questionnaire Adaptation

Table 3: Essential Reagents for Cross-Cultural Adaptation Research

Reagent / Tool	Function in the Adaptation Process	Exemplar Application
Bilingual Translators	Create linguistically accurate forward and back translations; require fluency in both cultures, not just languages.	Used in the adaptation of a survey tool for Portugal and the Dutch HLQ [51] [52].
Expert Panel	Review conceptual equivalence, cultural relevance, and technical adequacy of the adapted items.	Panels included healthcare professionals, researchers, and methodologists [51] [4].
Cognitive Interview Guide	A structured protocol with probing questions to assess item comprehension and cultural congruence in the target population.	Used questions like "What does this question mean to you?" and "Why was this difficult to answer?" [52].
Statistical Software (e.g., SPSS, R)	Conduct psychometric analyses including reliability (Cronbach's α), factor analysis (EFA, CFA), and test-retest reliability.	Used in all cited empirical studies to validate the adapted scales [49] [48] [4].
Pre-Test Sample	A small, representative subset of the target population for initial piloting of the adapted questionnaire.	A pre-test on 10% of the sample was used in the Ethiopian PCC study before main data collection [48].

The rigorous adaptation of TPB-based reproductive health questionnaires for cross-cultural contexts and low-literacy populations is a complex but essential methodological endeavor. It requires a systematic, multi-phase approach that integrates rigorous translation methods with deep cultural and cognitive validation. As demonstrated by the case studies, successfully adapted instruments enable the valid and reliable measurement of TPB constructs across diverse populations, which is fundamental for advancing our understanding of reproductive health behaviors globally and for designing effective, culturally-sensitive interventions. Future work in this field should continue to refine methods for ensuring equivalence and accessibility, particularly as digital health tools create new opportunities and challenges for questionnaire administration.

• Introduction to theory-practice gap: Contextualizes the challenge in behavioral health research.
• Theoretical foundations: Examines TPB and integration with complementary models.
• Questionnaire development: Details methodology for theory-based instrument creation.
• Experimental validation: Presents protocols for quantitative and qualitative assessment.
• Implementation framework: Proposes strategies for bridging theory to practice.
• Visualization and resources: Provides technical diagrams and research tools.

---

The Theory-Practice Gap: Ensuring Tools are Rooted in Behavioral Theory

The theory-practice gap represents a critical challenge in behavioral health research, particularly evident in the development and implementation of theory of planned behavior (TPB) questionnaires in reproductive health contexts. This gap manifests when instruments designed with strong theoretical foundations fail to demonstrate practical utility in real-world settings or when practitioners develop tools based on convenience rather than theoretical rigor. The disconnect between conceptual frameworks and applied research tools can compromise the validity of research findings and limit the effectiveness of interventions targeting reproductive health behaviors.

Recent studies highlight the consequences of this divide. In reproductive health research, where sensitive topics and cultural factors significantly influence behavior, theoretically flawed instruments can lead to inaccurate behavioral predictions and ineffective interventions. The examination of organ donation attitudes in Egypt demonstrates how integrating TPB with the Health Belief Model (HBM) can enhance understanding of complex health decisions, yet also reveals methodological challenges in operationalizing these theoretical constructs [53]. Similarly, the development of a suicide-related mental health service tendencies questionnaire for healthcare professionals illustrates the rigorous process required to maintain theoretical fidelity while creating practically applicable instruments [54].

Theoretical Foundations: Core Behavioral Models

Theory of Planned Behavior Constructs

The Theory of Planned Behavior provides a robust framework for understanding and predicting human behavior across diverse health contexts. Developed by Icek Ajzen, TPB posits that behavioral intention, the most proximal predictor of actual behavior, is influenced by three core constructs:

• Attitudes: The individual's positive or negative evaluation of performing the specific behavior, comprising behavioral beliefs about outcomes and evaluations of those outcomes.
• Subjective Norms: The perceived social pressure to engage or not engage in the behavior, incorporating normative beliefs about what important others think and motivation to comply with these expectations.
• Perceived Behavioral Control: The perceived ease or difficulty of performing the behavior, drawing on control beliefs about factors that may facilitate or impede the behavior and the perceived power of these factors.

In reproductive health contexts, these constructs manifest in complex ways. For instance, research on organ donation attitudes in Egypt found that chronic patients' donation intentions (91%) significantly exceeded those of healthy individuals (60%), suggesting that direct health experiences fundamentally reshape TPB constructs [53]. This demonstrates how the salience of behavioral beliefs can vary substantially between populations, necessitating contextual adaptation of theoretical applications.

Integration with Complementary Theoretical Frameworks

While TPB provides a powerful predictive framework, its explanatory power can be enhanced through strategic integration with complementary models:

• Health Belief Model (HBM): Integrating HBM's constructs of perceived susceptibility, severity, benefits, and barriers with TPB can provide a more comprehensive understanding of health behaviors. The Egyptian organ donation study effectively combined these models, revealing that chronic patients' decisions were primarily driven by health threat perceptions (HBM constructs) while healthy individuals' decisions were more influenced by social and cognitive factors (TPB constructs) [53].

• System Dynamics Modeling (SDM): This approach provides a methodological framework for understanding complex feedback relationships between TPB constructs over time. SDM characterizes systems through stocks (accumulations), flows (rates of change), and feedback loops (reinforcing and balancing), allowing researchers to model how TPB constructs interact dynamically within health systems [55].

Table 1: Key Constructs from Complementary Theoretical Frameworks

Theoretical Framework	Core Constructs	Application in Reproductive Health
Health Belief Model	Perceived susceptibility, severity, benefits, and barriers; cues to action; self-efficacy	Understanding perceived vulnerability to reproductive health risks and barriers to protective behaviors
System Dynamics Modeling	Stocks (accumulations), flows (rates of change), feedback loops (reinforcing, balancing)	Modeling how attitudes and norms interact dynamically in reproductive decision-making systems
Theory of Planned Behavior	Attitudes, subjective norms, perceived behavioral control, behavioral intention	Predicting contraceptive use, HIV testing intentions, and reproductive health service utilization

Questionnaire Development Methodology

Systematic Instrument Development Process

Developing a theoretically-grounded TPB questionnaire for reproductive health research requires a rigorous multi-stage process that maintains theoretical fidelity while ensuring practical applicability. The following methodology provides a robust framework for instrument development:

• Conceptual Mapping: Begin by creating a comprehensive conceptual map that explicitly links each TPB construct to domain-specific reproductive health behaviors. For instance, when developing a questionnaire about contraceptive use, clearly define behavioral beliefs (advantages/disadvantages), normative referents (partners, family, peers), and control factors (access, skills, emotions) [54].

• Item Generation: Develop multiple items for each theoretical construct through a combination of literature review, expert consultation, and qualitative research with the target population. The suicide-related mental health service tendencies questionnaire employed semi-structured interviews with healthcare providers to ensure items reflected practical realities while maintaining theoretical consistency [54].

• Theoretical Fidelity Check: Implement a systematic review process where behavioral scientists evaluate each item for theoretical alignment, while reproductive health experts assess content validity and cultural appropriateness. This dual-review process helps identify and resolve theory-practice discrepancies early in development.

Quantitative Validation Protocols

Once a preliminary questionnaire has been developed, rigorous quantitative validation is essential:

• Psychometric Evaluation: Administer the instrument to a sufficiently large sample (typically n≥200 for exploratory factor analysis, n≥300 for confirmatory factor analysis) representing the target population. Analyze data using both classical test theory and item response theory approaches to evaluate internal consistency, item discrimination, and dimensionality [54].

• Construct Validation: Employ both exploratory and confirmatory factor analysis to verify the theoretical factor structure. For the TPB reproductive health questionnaire, this typically involves testing the hypothesized three-factor structure (attitudes, subjective norms, perceived behavioral control) and their relationship to behavioral intentions.

• Criterion-Related Validation: Establish relationships between questionnaire scores and relevant behavioral criteria, such as clinical records of health service utilization or physiological measures of health outcomes. The prospective longitudinal design provides the strongest evidence for predictive validity.

Table 2: Psychometric Validation Metrics for TPB Questionnaires

Validation Type	Recommended Metrics	Acceptance Threshold	Application in TPB Reproductive Health Questionnaires
Reliability	Cronbach's α, Test-retest ICC, Split-half reliability	≥0.7 for group comparisons; ≥0.9 for individual assessment	Internal consistency of TPB construct scales over time and across populations
Construct Validity	CFI, RMSEA, SRMR, Factor loadings	CFI≥0.90; RMSEA≤0.08; Factor loadings≥0.4	Confirmation of theoretical TPB structure in reproductive health contexts
Criterion Validity	Correlation with behavioral measures, Sensitivity/Specificity	r≥0.3 with behavioral criteria; AUC≥0.7	Prediction of reproductive health service utilization or protective behaviors

Experimental Validation Protocols

Mixed-Methods Approach to Theory-Practice Integration

Bridging the theory-practice gap requires comprehensive validation protocols that assess both theoretical coherence and practical applicability:

• Cognitive Interviewing: Conduct structured interviews with representative participants as they complete the TPB questionnaire to identify items that are misunderstood, culturally insensitive, or theoretically misaligned from the participant perspective. Use verbal probing techniques to understand how respondents interpret each item and what thought processes they use to generate responses. This method was effectively employed in developing the suicide-related mental health service tendencies questionnaire, revealing discrepancies between theoretical constructs and practical interpretations among healthcare providers [54].

• System Dynamics Modeling: Apply SDM to understand how TPB constructs interact within complex reproductive health systems. Develop a causal loop diagram that maps reinforcing and balancing feedback between theoretical variables, then create a stock-and-flow model to simulate how interventions might influence behavior over time. Parameterize the model using both questionnaire data and empirical reproductive health outcomes [55].

Longitudinal Validation Study Design

A prospective cohort design provides the strongest evidence for both theoretical prediction and practical utility:

• Participant Recruitment: Recruit a diverse cohort (N≥500) representing the target population for the reproductive health questionnaire, ensuring adequate representation across key demographic and clinical variables.

• Data Collection Protocol: Administer the TPB questionnaire at baseline, then follow participants at predetermined intervals (e.g., 3, 6, and 12 months) to assess actual reproductive health behaviors through multiple methods (self-report, clinical records, physiological measures).

• Statistical Analysis: Use structural equation modeling to test the theoretical relationships between TPB constructs and their collective ability to predict behavioral outcomes. Calculate mediation effects to verify whether intentions mediate the relationship between TPB antecedents and behavior, and moderation effects to identify subgroups for whom the theory operates differently.

Implementation Framework: Bridging Theory to Practice

Contextual Adaptation Protocol

Successfully implementing TPB-based questionnaires in diverse reproductive health settings requires systematic contextual adaptation while maintaining theoretical integrity:

• Cultural Validation Framework: Develop a structured approach for adapting TPB questionnaires across cultural contexts. This includes assessing the relevance and salience of behavioral beliefs, identifying culturally significant normative referents, and evaluating context-specific control factors that facilitate or impede reproductive health behaviors. The Egyptian organ donation study demonstrated how chronic patients and healthy individuals prioritize different beliefs and norms, highlighting the need for population-specific adaptations [53].

• Stakeholder Integration Process: Implement a participatory development process that engages diverse stakeholders throughout questionnaire development. Adapt the group model building approach from system dynamics, which brings together researchers, practitioners, and community representatives to collaboratively define key variables and relationships [55]. This methodology enhances both theoretical robustness and practical relevance by incorporating multiple perspectives.

Dynamic Evaluation and Refinement

Maintaining alignment between theory and practice requires ongoing evaluation mechanisms:

• Real-World Performance Monitoring: Establish systematic processes for continuously monitoring how well TPB questionnaires perform in predicting actual reproductive health behaviors across different implementation contexts. Track predictive accuracy metrics and identify contexts where the theory underperforms.

• Iterative Refinement Cycles: Implement structured intervals for questionnaire refinement based on accumulated empirical evidence and evolving reproductive health contexts. The system dynamics modeling approach provides a framework for understanding how feedback from practical application should inform theoretical refinements [55].

Theoretical Framework Integration

Theoretical Framework Integration Diagram: This visualization illustrates the process of integrating Theory of Planned Behavior (TPB), Health Belief Model (HBM), and System Dynamics Modeling (SDM) to address the theory-practice gap in reproductive health questionnaire development.

Questionnaire Development Workflow

Questionnaire Development Workflow: This diagram outlines the systematic process for developing theoretically-grounded TPB questionnaires for reproductive health research, from theoretical definition through implementation and refinement.

Research Reagent Solutions for TPB Questionnaire Development

Table 3: Essential Research Reagents for TPB Reproductive Health Research

Research Reagent	Theoretical Function	Application Protocol
TPB Construct Scales	Operationalizes attitudes, subjective norms, perceived behavioral control	Administer using balanced Likert scales (1-7 points); validate factor structure for each reproductive health context
Behavioral Intention Measures	Assesses self-reported likelihood of performing reproductive health behaviors	Use specific, context-rich scenarios with temporal framing to enhance predictive validity
Actual Behavior Metrics	Provides criterion for validating theoretical predictions	Combine self-report with objective measures (clinical records, physiological data) to minimize common method bias
System Dynamics Modeling Software	Simulates feedback relationships between TPB constructs over time	Parameterize models with empirical data to test intervention scenarios and identify leverage points
Cognitive Interviewing Protocol	Identifies gaps between theoretical constructs and participant interpretations	Implement structured verbal probing to assess item interpretation, sensitivity, and cultural appropriateness

The theory-practice gap in TPB reproductive health questionnaire development represents both a challenge and an opportunity for researchers committed to evidence-based behavioral interventions. By implementing the integrated methodologies outlined in this technical guide—including systematic theoretical mapping, rigorous psychometric validation, contextual adaptation protocols, and dynamic evaluation frameworks—researchers can develop instruments that maintain theoretical integrity while demonstrating practical utility in diverse reproductive health contexts. The continuous refinement process, informed by real-world application and emerging theoretical insights, offers a pathway toward increasingly effective tools for understanding and promoting reproductive health behaviors across diverse populations.

Structural Equation Modeling (SEM) is a powerful multivariate analysis technique that combines factor analysis and path analysis to test complex theoretical models involving latent constructs and their interrelationships. Its application in research based on the Theory of Planned Behavior (TPB) concerning reproductive health questionnaires is particularly valuable for elucidating the psychological mechanisms driving health behaviors. The validity of any SEM analysis, however, is fundamentally contingent upon the quality and suitability of the input data. This guide provides an in-depth technical framework for researchers and drug development professionals to optimize their data preparation processes, ensuring robust and interpretable SEM outcomes within the specific context of TPB-based reproductive health research.

Foundational Principles for SEM Data Preparation

The process of preparing data for SEM requires meticulous attention to specific data characteristics from the initial stages of research design. The foundational principles outlined below are critical for ensuring the validity of subsequent analyses.

2.1 Latent versus Emergent Variables A core conceptual distinction in SEM is between two types of constructs. Latent variables (also known as reflective constructs) are unobserved phenomena that are inferred from a set of measured indicators that are interchangeable and share a common theme, such as 'L2 intrinsic motivation' or the TPB construct of 'Attitude' [56]. In contrast, emergent variables (also known as formative constructs) are formed as a composite of distinct, non-interchangeable indicators, such as 'L2 writing achievement' being formed by separate scores for spelling, writing samples, and sentence fluency [56]. Misspecifying a formative construct as reflective (or vice-versa) constitutes a fundamental structural error that will compromise the entire model.

2.2 The Integrated SEM-ML Framework Contemporary methodological advances propose integrating SEM with Machine Learning (ML) to create a unified analytic pipeline. This framework leverages the strengths of both: SEM confirms the theoretical construct validity and model fit, while ML assesses the predictive utility of the instrument on out-of-sample data. This integration allows for a more robust evaluation of a measurement scale's psychometric properties, balancing theoretical coherence with empirical performance [57].

Pre-Data Collection Optimization

Optimization for SEM must begin before a single data point is collected, focusing on instrument design and sample planning.

3.1 Psychometric Validation of the Questionnaire For a TPB reproductive health questionnaire, rigorous validation is a non-negotiable prerequisite. The process should include:

• Face Validity: Assessing the questionnaire's appropriateness from the participant's perspective. This can be done via cognitive interviews or the "method of spoken reflection," where participants verbalize their thought process while answering, allowing researchers to identify ambiguous or difficult items [58].
• Content Validity: Qualitative assessment by a panel of experts to ensure the items are relevant, appropriate, and comprehensively represent the constructs of interest (e.g., Attitude, Subjective Norms, Perceived Behavioral Control) within the reproductive health domain [59] [58].
• Pilot Study: Conducting a pilot study is recommended to test the survey instrument's complexity and the research process, providing insights that can refine the final design [58].

3.2 Sample Size Determination Adequate sample size is critical for the statistical power and stability of SEM parameter estimates. While rules of thumb exist (e.g., a minimum of 10-20 cases per estimated parameter), sample size calculation should be a deliberate process. One approach is to base the calculation on the primary outcome variable. For instance, a randomized controlled trial based on TPB for adolescent girls' reproductive health determined a sample size of 578 (289 per group) to detect a 10-point difference in knowledge scores with 90% power, accounting for a design effect and potential participant attrition [3].

Table 1: Key Psychometric Validation Steps for a TPB Reproductive Health Questionnaire

Validation Step	Description	Methodology Example
Face Validity	Assesses clarity, difficulty, and ambiguity from the participant's viewpoint.	"Spoken reflection" method with a small, representative sample [58].
Content Validity	Evaluates item relevance and comprehensiveness from an expert perspective.	Review by a panel of experts in reproductive health, social health, and violence [59].
Construct Validity	Tests the hypothesized underlying structure of the questionnaire.	Exploratory Factor Analysis (EFA) to extract factors and confirm the theoretical structure [59] [58].
Internal Consistency	Measures the reliability and inter-relatedness of items within a scale.	Cronbach's Alpha for Likert-scale perception sections; Kuder-Richardson (KR-20) for dichotomous knowledge questions [58].

Post-Data Collection Preparation and Screening

Once data is collected, a thorough screening and preparation phase is essential to ensure the data matrix is suitable for SEM.

4.1 Handling Missing Data Missing data can introduce significant bias. It is crucial to report the amount and pattern of missingness. While simple techniques like listwise deletion are common, they can reduce power and introduce bias if the data is not missing completely at random (MCAR). SEM offers a significant advantage here, as it can employ Full Information Maximum Likelihood (FIML) estimation, which uses all available data points to produce less biased parameter estimates compared to conventional methods [56].

4.2 Assessing Data Normality Many standard SEM estimation methods (e.g., Maximum Likelihood) assume multivariate normality. Researchers should examine:

• Univariate Normality: Check skewness and kurtosis for each variable.
• Multivariate Normality: Assess using Mardia's coefficient or other multivariate tests. Significant non-normality can lead to inflated Chi-square values and incorrect standard errors. SEM frameworks provide robust methods to handle this, such as robust estimators (e.g., Satorra-Bentler scaled chi-square) or bootstrapping, which can lead to more valid and unbiased findings [56].

4.3 Item Analysis for Refinement Before model testing, item-level analysis can help refine the measurement instrument. In an integrated SEM-ML framework, this involves:

• Successive Item Removal: Iteratively removing items and re-running both the SEM and an ML classifier.
• Joint Optimization Criterion: Retaining only those items whose removal does not substantially worsen model fit (e.g., measured by ΔRMSEA) and does not compromise predictive accuracy [57]. This process can yield a shorter, better-fitting scale without sacrificing predictive power, leading to more efficient data collection in the future.

Table 2: Essential Data Screening Steps Prior to SEM Analysis

Screening Step	Purpose	Tool/Method
Missing Data Analysis	To identify the extent and pattern of missing data, which informs the handling strategy.	Frequency reports, Little's MCAR test. FIML estimation in SEM.
Normality Assessment	To check the assumption of multivariate normality for parameter estimation.	Examination of skewness & kurtosis; Mardia's test for multivariate normality.
Outlier Detection	To identify multivariate outliers that can disproportionately influence results.	Mahalanobis distance.
Item Analysis	To refine the measurement scale by identifying poorly performing items.	SEM fit indices (e.g., RMSEA) combined with Machine Learning predictive accuracy [57].

Experimental Protocols for TPB Questionnaire Validation

The following detailed methodologies are adapted from validated studies on reproductive health and can be integrated into a TPB research framework.

5.1 Protocol for a TPB-Based Educational Intervention This protocol demonstrates how to generate data for evaluating a theoretical model's effectiveness [3].

• Design: Randomized Controlled Trial (RCT).
• Participants: Target population (e.g., 578 high school girls, aged 12-16). Use multistage random cluster sampling to select districts, schools, and classes. Randomly assign schools to experimental and control groups using permuted block randomization.
• Intervention: The experimental group receives a multi-component, TPB-based educational intervention. This includes:
  • For Adolescents: Training sessions using varied techniques (presentations, role-playing, animations, group discussions) and a tailored booklet covering puberty, menstrual health, and HIV/STI prevention.
  • For Parents: A two-hour workshop on adolescent high-risk behaviors and reproductive health.
• Measures: Administer a self-reported TPB questionnaire pre-intervention, post-intervention, and at follow-up (e.g., 6 months). The questionnaire should measure TPB constructs (Attitude, Subjective Norms, Perceived Behavioral Control, Intention) and actual behavior, using Likert-scale items.
• Data Analysis: Use Analysis of Covariance (ANCOVA) to compare post-intervention scores between groups, controlling for pre-intervention scores. Subsequently, SEM can be used to test the full TPB model and the intervention's impact on the pathways.

5.2 Protocol for Questionnaire Validation via Exploratory Factor Analysis (EFA) This protocol is critical for establishing the construct validity of a new or adapted TPB reproductive health questionnaire [59] [58].

• Participant Recruitment: Recruit a sufficient sample (e.g., 350 participants) through purposive or cluster-randomized sampling from the target population.
• Data Collection: Administer the draft questionnaire, which includes demographic sections and Likert-scale sections for TPB constructs.
• Factor Analysis Suitability: Check the data's suitability for EFA using the Kaiser-Meyer-Olkin (KMO) measure (should be >0.6) and Bartlett's Test of Sphericity (should be significant).
• Factor Extraction: Perform EFA using Principal Component Analysis with varimax rotation. The number of factors to extract can be determined by Kaiser's criterion (eigenvalue >1) and scree plot inspection.
• Interpretation: Analyze the rotated factor matrix. Items should load strongly (>0.4) on their hypothesized theoretical factor (e.g., all "Attitude" items loading on a single factor) with minimal cross-loadings. The extracted factors should account for a satisfactory amount of total variance (e.g., ~47-60%).

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Tools and Software for SEM Data Preparation and Analysis

Tool/Software	Function	Application Context
IBM SPSS Statistics	Statistical analysis for data screening, descriptive statistics, EFA, and reliability analysis.	Preliminary data analysis, assumption checking, and initial scale validation [58].
R Software with lavaan package	Open-source environment for conducting SEM, including CFA and path analysis with ML estimation.	Full SEM model testing, handling missing data via FIML, and using robust estimators [58].
Jamovi	Free, user-friendly statistical software that includes SEM modules through plugins.	Accessible SEM for researchers less familiar with coding, suitable for learning and basic model testing [56].
Color Contrast Analyzer	Tools to verify that visual elements in diagrams meet accessibility standards (WCAG).	Ensuring that diagrams (e.g., path models) are readable for all audiences, including those with low vision [60] [61] [62].

Workflow Visualization

The following diagram illustrates the integrated data optimization workflow for SEM, from study design to model evaluation.

Diagram 1: SEM Data Optimization Workflow

The second diagram details the core process of validating the measurement model, a critical step before testing the full structural model.

Diagram 2: Measurement Model Validation Process

Ensuring Rigor: A Framework for Psychometric and Comparative Validation

Establishing Face and Content Validity with Expert Panels and Target Populations

In the development of a Theory of Planned Behavior (TPB) questionnaire for reproductive health research, establishing robust face and content validity is a critical first step. This process ensures that the instrument's items are relevant, comprehensive, and comprehensible to the target population, providing a solid foundation for subsequent psychometric testing [63]. Within the specific context of reproductive health—where topics can be sensitive and concepts complex—meticulous validation is paramount for obtaining accurate data on behavioral intentions, attitudes, subjective norms, and perceived behavioral control [64] [65]. This guide provides researchers and drug development professionals with a detailed, technical roadmap for establishing face and content validity through structured engagement with expert panels and target populations.

Conceptual Foundations of Validity

Content validity refers to the degree to which an instrument's content is a relevant and representative reflection of the construct it aims to measure [66] [63]. It links abstract theoretical concepts, such as those in the TPB, to observable and measurable indicators. Face validity, often considered a component of content validation, assesses whether the instrument appears to measure what it is intended to, from the perspective of the end-user [66] [67].

For a TPB-based questionnaire in reproductive health, the "construct" encompasses the key domains of the theory: Attitude, Subjective Norms, and Perceived Behavioral Control, as they relate to a specific health behavior (e.g., using modern family planning) [64]. A valid content foundation ensures that items within these domains are understood as researchers intend and are meaningful to respondents.

The Item Generation Process

The development of a valid questionnaire begins with the comprehensive generation of a preliminary item pool. A mixed-methods approach is recommended to capture the construct in its entirety [66].

Deductive Methods

Deductive methods derive items from established theory and existing literature. For a TPB reproductive health questionnaire, this involves:

• Theory Review: Systematically analyzing the TPB literature to identify core conceptual components that must be measured [65].
• Literature Synthesis: Reviewing existing, validated TPB questionnaires for similar health behaviors to identify potentially adaptable items [64].

Inductive Methods

Inductive methods generate items directly from the target population's lived experiences, ensuring cultural and contextual relevance. This is crucial for capturing nuances in sensitive reproductive health topics [66]. Techniques include:

• Focus Group Discussions (FGDs): Conducting structured discussions with small groups from the target population to explore beliefs, language, and perceptions related to the health behavior.
• In-Depth Interviews (IDIs): Performing one-on-one interviews to gain deeper, more personal insights into the behavioral constructs.

Table 1: Key Considerations for Item Wording and Formatting

Aspect	Best Practice	Rationale
Clarity	Use simple, unambiguous language.	Ensures items are easy to comprehend and answer [66].
Relevance	Ensure items capture the lived experience of the target population.	Increases ecological validity and respondent engagement [66].
Frame	Items should be objective and avoid leading questions.	Reduces social desirability and other response biases [66].
Order	Move from generic to specific items; place sensitive questions later.	Builds rapport and comfort, improving data quality [66].
Response Scale	Use ordinal Likert scales with at least five points; ensure options are mutually exclusive.	Enhances reliability and allows for variance in expression [66].

Establishing Face Validity with Target Populations

Face validity assesses the acceptability, comprehensibility, and relevance of the questionnaire from the end-user's viewpoint. This is typically evaluated through cognitive interviewing [67].

Experimental Protocol: Cognitive Interviewing

Objective: To identify problems with item comprehension, recall, judgment, and response formatting [67].

Procedure:

• Recruitment: Recruit a purposive sample of 15-30 individuals from the target population (e.g., women of reproductive age for a family planning questionnaire) [67].
• Data Collection: Administer the draft questionnaire and conduct one-on-one interviews using two primary techniques:
  • Think-Aloud Protocol: Participants verbalize their thoughts as they answer each question.
  • Verbal Probing: The interviewer asks specific, pre-determined probes (e.g., "Can you repeat that question in your own words?" or "What were you thinking when you chose that answer?") [67].
• Data Analysis: Review interview transcripts and notes to identify recurring issues with specific items, such as misinterpretation of terms, confusing phrasing, or culturally insensitive content.

Output and Iteration: The qualitative feedback is used to refine, reword, or remove problematic items. The process is iterative, often requiring multiple rounds of testing until no major issues are identified [67].

Quantitative Assessment: Item Impact Score

The Item Impact Score (IIS) can objectively quantify face validity.

• Formula: IIS = Frequency × Importance, where:
  • Frequency = the proportion of respondents rating an item as "important" or "very important" (e.g., 4 or 5 on a 5-point scale).
  • Importance = the mean importance rating for that item [66].
• Interpretation: An IIS > 1.5 is generally considered acceptable, indicating that at least 50% of respondents agreed on the item's importance with a mean score of at least 3 [66].

Establishing Content Validity with Expert Panels

Content validity is judged by experts to ensure the instrument adequately covers all key domains of the construct. The process involves both qualitative feedback and quantitative metrics [66] [63].

Experimental Protocol: Expert Panel Evaluation

Objective: To obtain quantitative and qualitative judgments on the relevance, comprehensiveness, and representativeness of each item and the entire scale [66] [63].

Procedure:

• Expert Recruitment: Assemble a panel of 7-10 content experts, including TPB methodologies, reproductive health clinicians, and psychometricians [66] [67].
• Rating Task: Provide experts with the operational definitions of the TPB constructs and ask them to rate each item on its relevance to the target construct using a 4-point scale (e.g., 1 = not relevant, 2 = somewhat relevant, 3 = quite relevant, 4 = highly relevant) [66].
• Qualitative Feedback: Solicit open-ended feedback on the item pool's comprehensiveness, clarity, and any potential gaps in content coverage.

Quantitative Metrics for Content Validity

The expert ratings are analyzed using several key indices, summarized in the table below.

Table 2: Key Indices for Quantifying Content Validity

Index	Formula	Interpretation	Acceptability Threshold
Content Validity Ratio (CVR)	CVR = (Nₑ - N/2) / (N/2)N = number of expertsNₑ = number rating item "essential"	Measures essentiality; used for initial item retention [66].	Depends on panel size (e.g., for 7 experts, critical CVR = 0.99) [66].
Item-Level Content Validity Index (I-CVI)	I-CVI = Number rating item 3 or 4 / Total number of experts	Proportion of experts agreeing on an item's relevance [66].	0.78 for 6-10 experts; 1.00 for 3-5 experts [66].
Scale-Level CVI, Average (S-CVI/Ave)	S-CVI/Ave = Mean of all I-CVIs	The average of I-CVI scores across all items [66].	≥ 0.90 [66]
Scale-Level CVI, Universal Agreement (S-CVI/UA)	S-CVI/UA = Number of items with I-CVI = 1 / Total items	The proportion of items that achieve a perfect relevance score from all experts [66].	A more conservative metric; no universal threshold, but higher is better.

Integration and Iterative Refinement

Establishing face and content validity is not a linear but an iterative process. Findings from cognitive interviews with the target population and evaluations from the expert panel must be integrated to inform revisions [66] [67]. For instance, an item deemed highly relevant by experts (high I-CVI) might be consistently misunderstood by the target population, necessitating rewording. This cyclical process of testing, refinement, and re-evaluation continues until both face and content validity are judged to be strong. The following workflow diagram illustrates this iterative process.

The Scientist's Toolkit: Essential Research Reagents

The following table details key materials and tools required for executing the face and content validation protocols described in this guide.

Table 3: Essential Reagents for Face and Content Validation Studies

Research Reagent	Function/Application
Theoretically-Grounded Item Pool	The initial set of candidate questions, generated via deductive (theory/literature) and inductive (qualitative research) methods, serving as the raw material for validation [66] [67].
Expert Panel	A group of 7-10 specialists (e.g., in TPB, reproductive health, psychometrics) who provide quantitative ratings and qualitative feedback to establish content validity [66] [63].
Structured Expert Rating Form	A standardized data collection tool, often using a 4-point relevance scale, to systematically gather expert judgments on each item for calculating CVR and CVI [66].
Cognitive Interview Guide	A semi-structured protocol containing think-aloud instructions and verbal probes to assess participant comprehension and perception of questionnaire items [67].
Content Validity Analysis Spreadsheet	A tool (e.g., in MS Excel or SPSS) for calculating key quantitative indices (IIS, CVR, I-CVI, S-CVI) from expert and target population data [66].

Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) are foundational statistical methods used to establish the construct validity of measurement instruments in behavioral and health sciences. Within the context of Theory of Planned Behavior (TPB) questionnaire research in reproductive health, these analyses are crucial for verifying that questionnaire items accurately measure their intended theoretical constructs: attitude, subjective norms, perceived behavioral control, and behavioral intention. Construct validity evidence confirms that an instrument truly measures the abstract concepts it purports to measure, which is especially critical when assessing sensitive topics like reproductive health behaviors, where accurate measurement directly impacts intervention effectiveness [27].

EFA and CFA serve complementary roles in validation. EFA explores the underlying factor structure without preconceived hypotheses, allowing researchers to discover potential latent variables that explain patterns among observed variables. In contrast, CFA tests a pre-specified factor structure based on theoretical expectations, confirming whether the data supports the hypothesized relationships between observed measures and their latent constructs [68]. For TPB-based reproductive health questionnaires, this means verifying that items cluster according to the theory's core components rather than extraneous factors, thus ensuring the theoretical integrity of the measurement instrument.

Fundamental Concepts and Theoretical Framework

Key Terminology and Definitions

• Construct Validity: The degree to which an instrument measures the theoretical construct it intends to measure. In TPB research, this ensures questions truly capture attitudes, norms, and behavioral control rather than other psychological factors.
• Latent Variable: An unobserved variable inferred from observed measures (e.g., "perceived behavioral control" inferred from responses to multiple questionnaire items).
• Observed Variable: Directly measured variables (questionnaire items) that serve as indicators of latent constructs.
• Factor Loading: The correlation between an observed variable and its underlying latent factor, indicating how strongly the item contributes to measuring the construct.
• Model Fit: How well the hypothesized factor structure reproduces the observed data covariance matrix.

Comparative Analysis of EFA and CFA

Table 1: Fundamental Differences Between EFA and CFA

Characteristic	Exploratory Factor Analysis (EFA)	Confirmatory Factor Analysis (CFA)
Primary Objective	Explore underlying structure without pre-existing hypotheses	Confirm or reject a pre-specified factor structure
Theoretical Basis	Theory-generating; used when relationships unknown	Theory-testing; requires strong theoretical foundation
Factor Structure	Not predetermined; discovered from data	Explicitly specified before analysis
Model Constraints	Minimal constraints; all variables load on all factors	Specific constraints based on theoretical expectations
Implementation Timing	Early stages of scale development	Later stages with established theoretical model

Methodological Protocols for Factor Analysis

Exploratory Factor Analysis Protocol

Step 1: Sampling and Sample Size Determination For EFA, adequate sample size is critical for stable factor solutions. While rules of thumb vary, a sample of 100-250 participants is generally acceptable, with a participant-to-variable ratio of at least 5:1 recommended [69]. For example, in developing a Malay COVID-19 health literacy questionnaire, researchers recruited 100 participants for EFA, ensuring sufficient power for factor extraction [69]. Researchers must also verify that data meets statistical assumptions for factorability through the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (values > 0.6 acceptable, > 0.8 preferable) and Bartlett's Test of Sphericity (should be statistically significant, p < 0.05) [70] [69].

Step 2: Factor Extraction and Selection Multiple extraction methods exist, with maximum likelihood factoring and principal axis factoring generally recommended [70]. The number of factors to retain should be determined through multiple criteria, including Kaiser's criterion (eigenvalues > 1.0), scree plot examination, and parallel analysis. In the development of the Understanding, Attitude, Practice and Health Literacy Questionnaire on COVID-19 (MUAPH C-19), eigenvalue analysis revealed two components in the understanding domain, three in attitude, four in practice, and one in health literacy [69].

Step 3: Factor Rotation and Interpretation Rotation simplifies factor structure for enhanced interpretability. Varimax rotation (orthogonal) assumes uncorrelated factors, while Oblimin (oblique) allows correlated factors, which often better reflects psychological constructs in TPB research. Items typically require factor loadings of at least 0.4-0.5 to be considered meaningful contributors to a factor [69]. Following rotation, researchers interpret the pattern matrix to assign conceptual meaning to factors and assess how well they align with theoretical expectations.

Confirmatory Factor Analysis Protocol

Step 1: Model Specification CFA begins with explicit specification of the hypothesized measurement model based on strong theoretical grounding or previous EFA results. This includes defining which observed variables (questionnaire items) load on which latent constructs, and whether latent constructs are correlated. For TPB questionnaires, this means specifying exactly which items should load on attitude, subjective norms, perceived behavioral control, and intention factors [27]. The model must be statistically identified, typically requiring at least three indicators per latent variable and setting the scale for each latent variable (e.g., fixing one loading to 1 or standardizing the latent variance) [68].

Step 2: Parameter Estimation and Model Evaluation Parameters are estimated using methods like maximum likelihood estimation, which iteratively adjusts model parameters to minimize differences between observed and model-implied covariance matrices. Model fit is then assessed using multiple indices, each with established thresholds indicating acceptable fit:

• Root Mean Square Error of Approximation (RMSEA): Values ≤ 0.08 indicate acceptable fit, ≤ 0.05 good fit [70] [27]
• Comparative Fit Index (CFI): Values ≥ 0.90 acceptable, ≥ 0.95 good fit [70] [27]
• Tucker-Lewis Index (TLI): Values ≥ 0.90 acceptable, ≥ 0.95 good fit [70]
• Standardized Root Mean Square Residual (SRMR): Values < 0.08 acceptable

In a study validating a phlegm pattern questionnaire, CFA results showed RMSEA = 0.074 (acceptable), CFI = 0.878 (slightly below threshold), and TLI = 0.860 (approaching acceptable), suggesting marginally acceptable model fit [70].

Step 3: Model Respecification and Validity Assessment If initial model fit is inadequate, theoretically justified modifications may improve fit. This might include adding correlated residuals between items with shared method variance or allowing cross-loadings where theoretically plausible. After achieving acceptable fit, researchers assess convergent validity (items strongly correlate with their intended factor, AVE > 0.5) and discriminant validity (factors are distinct, AVE > squared correlations between factors) [70]. For TPB questionnaires, this confirms that attitude, norms, and control are empirically distinct constructs.

Application to Theory of Planned Behavior Questionnaires in Reproductive Health

TPB Questionnaire Development and Validation Cycle

The development and validation of TPB-based reproductive health questionnaires follows a systematic sequence incorporating both EFA and CFA. Researchers begin with item generation through qualitative methods (e.g., interviews, focus groups) to ensure content validity and cultural relevance for sensitive reproductive topics. Following initial item development, EFA helps explore whether items cluster according to TPB constructs, particularly important when adapting instruments for new cultural contexts [48] [27]. Once a stable factor structure emerges, CFA confirms the theoretical structure in a new sample, providing robust evidence of construct validity.

In a study developing a TPB-based questionnaire to predict medical staff's intention to discuss sexual issues with postmenopausal women, researchers employed a mixed-methods approach with three phases: qualitative content analysis, instrument development, and predictive validation. The CFA results demonstrated sufficient model fit (AGFI = 0.89, RMSEA = 0.07, CFI = 0.9, GFI = 0.94), supporting the hypothesized five-factor TPB structure (attitude, abstract norms, perceived behavioral control, intention, and behavior) [27]. Similarly, in assessing intention to use preconception care among Ethiopian women, CFA validated the TPB structure with perceived behavioral control emerging as the strongest predictor (β = 0.263), followed by attitude (β = 0.201) and subjective norms (β = 0.158) [48].

Quantitative Data Presentation in TPB Research

Table 2: Model Fit Indices from Representative TPB Questionnaire Validation Studies

Study Context	Sample Size	RMSEA	CFI	TLI	GFI	Factor Loadings
Sexual issues discussion with postmenopausal women [27]	208	0.07	0.90	-	0.94	0.83-0.98 (Cronbach's α)
Phlegm pattern questionnaire validation [70]	289	0.074	0.878	0.860	0.839	Mixed discriminant validity
Preconception care use intention [48]	415	-	-	-	-	PBC: β=0.263, Att: β=0.201, SN: β=0.158

Table 3: Reliability and Validity Metrics from Health Questionnaire Validation Studies

Questionnaire Domain	Cronbach's Alpha	Test-Retest Reliability	Number of Final Items	Variance Explained
Understanding (COVID-19) [69]	0.677-0.914	0.562-0.759	42	41.308-68.250%
Behavioral control belief [27]	0.83-0.89	0.84-0.98	121	24-37% (R²)
Attitude toward sexual issues discussion [27]	0.89	0.96	24	-

Essential Research Reagents and Tools

Statistical Software and Analysis Tools

Table 4: Essential Statistical Tools for Factor Analysis

Tool Name	Primary Function	Application in Factor Analysis
SPSS Statistics	General statistical analysis	EFA with principal components analysis, reliability testing
AMOS	Structural equation modeling	CFA with graphical model specification, model fit assessment
R with lavaan package	Open-source statistical computing	Both EFA and CFA with extensive customization options
Mplus	Advanced statistical modeling	Complex CFA models with categorical data and multilevel structures

Methodological Components for Validation Studies

• Participant Recruitment Protocols: Systematic sampling methods with appropriate inclusion/exclusion criteria to ensure representative samples [70] [48]
• Power Analysis Procedures: Sample size determination based on participant-to-variable ratios or Monte Carlo simulation for CFA [69] [48]
• Cross-Validation Samples: Independent samples for EFA and CFA phases to avoid overfitting and ensure generalizability [70]
• Reliability Assessment Tools: Cronbach's alpha for internal consistency, test-retest reliability for temporal stability [69] [27]
• Content Vality Indices: Quantitative evaluation of item relevance and representativeness by subject matter experts [27]

Advanced Analytical Considerations

Integration with Broader Research Methodology

Factor analysis does not occur in isolation within TPB questionnaire validation. Researchers must integrate these analyses with comprehensive research designs that include longitudinal components to assess predictive validity, multiple-group CFA to examine measurement invariance across demographic groups, and convergent-discriminant validation with related constructs. For reproductive health applications, this might involve testing whether TPB constructs predict actual behavioral outcomes (e.g., contraceptive use, healthcare seeking) or examining whether measures perform equivalently across gender, education, or cultural subgroups.

The sequential use of EFA and CFA strengthens validation evidence, as demonstrated in phlegm pattern questionnaire research where initial CFA revealed somewhat insufficient discriminant validity (average variances extracted smaller than factor correlation coefficients), prompting additional EFA to explore alternative factor structures [70]. This iterative process of model testing and refinement is characteristic of robust instrument development and aligns with best practices in health measurement theory.

Reporting Standards and Interpretation Guidelines

Comprehensive reporting of EFA and CFA results should include: detailed description of extraction and rotation methods, factor loading matrices, communalities, complete model fit statistics, modification indices when applicable, and evidence of reliability and validity. For TPB applications, researchers should specifically report how factors align with theoretical constructs and discuss any deviations from expected structures. When applying these methods to reproductive health questionnaires, particular attention should be paid to cultural and linguistic factors that might influence measurement properties, especially when translating instruments or adapting them for new populations.

In the field of health services research, particularly when investigating complex psychosocial constructs within frameworks like the Theory of Planned Behavior (TPB), the reliability of measurement instruments is paramount. Reliability refers to the consistency of a measurement method—the extent to which it yields stable and reproducible results under consistent conditions [71] [72]. In TPB research, which examines the relationships between attitudes, subjective norms, perceived behavioral control, and behavioral intentions, unreliable measurement can obscure critical relationships and lead to invalid conclusions about the factors influencing health behaviors [3] [32] [48].

The importance of reliability is particularly acute in reproductive health research, where TPB-based questionnaires often form the foundation for interventions targeting critical outcomes. For instance, studies examining intentions to use health facilities for childbirth [32], adopt sexual and reproductive health behaviors [3], or utilize preconception care services [48] all depend on questionnaires that can consistently measure TPB constructs across different populations and time points. Without demonstrating adequate reliability, the findings from such research remain questionable, potentially undermining evidence-based interventions in a domain with significant public health implications.

This technical guide provides an in-depth examination of three fundamental reliability assessment methods—internal consistency, test-retest, and split-half reliability—with specific application to TPB questionnaire development and validation in reproductive health research.

Core Reliability Concepts and Their Importance in TPB Research

Defining Reliability in Measurement

Reliability encompasses the degree to which a measurement instrument produces consistent, dependable, and repeatable results when applied to the same phenomenon under the same conditions [71] [72]. In statistical terms, reliability concerns the proportion of variance in measurements that is attributable to the true score of the construct being measured, as opposed to measurement error [73]. A perfectly reliable measure would contain no measurement error, with all observed variance reflecting true differences in the construct.

The relationship between reliability and validity is crucial: while reliability concerns consistency, validity addresses whether an instrument actually measures what it purports to measure [71] [72]. Importantly, a measure cannot be valid without first being reliable; unreliability introduces random error that necessarily compromises validity [74] [72]. This distinction is particularly relevant for TPB questionnaires, where researchers must demonstrate that their instruments consistently measure specific constructs (attitudes, subjective norms, perceived behavioral control) before claiming they accurately represent these theoretical variables.

Reliability Considerations for TPB Questionnaires

TPB questionnaires in reproductive health research present unique reliability challenges due to several factors:

• Multidimensional constructs: TPB variables often comprise multiple dimensions that must be measured consistently. For example, perceived behavioral control may include both self-efficacy and controllability components [32] [48].
• Context sensitivity: Reproductive health behaviors are often influenced by cultural and contextual factors that may affect measurement consistency across different populations [3] [32].
• Behavioral specificity: TPB measures are behavior-specific, requiring careful construction to ensure all items consistently reference the same target behavior, context, and time frame [48].

These considerations necessitate rigorous reliability testing throughout the questionnaire development process, from initial piloting to full-scale implementation.

Internal Consistency Reliability

Conceptual Foundation

Internal consistency reliability assesses the extent to which items within a single measurement instrument that are designed to measure the same construct produce similar results [71] [73]. This form of reliability evaluates the interrelatedness among items purportedly measuring the same underlying construct, based on the principle that items measuring the same construct should be highly correlated with one another [71] [72].

For TPB questionnaires in reproductive health, internal consistency is particularly important for multi-item scales measuring the core theoretical constructs. For instance, a subjective norms scale regarding preconception care utilization might include items addressing perceptions of approval from different referent groups (partner, family, friends, healthcare providers) [48]. High internal consistency would indicate that these items collectively tap into the same underlying subjective norms construct rather than measuring disparate concepts.

Measurement Approaches

Table 1: Methods for Assessing Internal Consistency Reliability

Method	Procedure	Interpretation	Application to TPB Questionnaires
Cronbach's Alpha	Calculate the average of all possible split-half correlations	α ≥ 0.7 = acceptable; α ≥ 0.8 = good; α ≥ 0.9 = excellent [75] [72]	Preferred method for multi-item TPB scales (attitudes, subjective norms, etc.)
Average Inter-item Correlation	Compute correlation between all item pairs, then average	Optimal range: 0.15-0.50 [73]	Useful for evaluating item homogeneity within TPB constructs
Item-Total Correlation	Correlate each item score with total scale score	Minimum acceptable: 0.20-0.30; Ideal: >0.40 [73]	Identifies poorly performing items in TPB scales that may need revision

Cronbach's alpha is mathematically equivalent to the average of all possible split-half estimates for a set of items [73]. The formula for Cronbach's alpha is:

α = (k / (k - 1)) × (1 - (Σσ²item / σ²total))

Where k = number of items, Σσ²item = sum of item variances, and σ²total = variance of total scores.

Application in Reproductive Health Research

In TPB-based reproductive health research, demonstrating strong internal consistency is essential for establishing that questionnaire items collectively measure their intended constructs. For example:

• A study validating a TPB questionnaire on sexual and reproductive health among Iranian adolescent girls reported internal consistency for attitude (21 items), subjective norms (19 items), perceived behavioral control (25 items), and intention (18 items) scales, though specific alpha values were not provided [3].
• Research on preconception care utilization in Ethiopia employed Cronbach's alpha to ensure internal consistency of TPB constructs, with all values exceeding 0.7, meeting the threshold for acceptable reliability [48].
• A study developing a medication knowledge assessment questionnaire for hemodialysis patients demonstrated high internal consistency with Cronbach's alpha values ranging from 0.716 to 1.00 across domains [76].

Implementation Protocol

Procedure for Assessing Internal Consistency in TPB Questionnaires:

• Administer the complete questionnaire to a representative pilot sample (typically 5-10 participants per item, minimum 100-200 total) [75] [77].
• Calculate inter-item correlations for all items within each theoretical construct (attitude, subjective norms, perceived behavioral control, intention).
• Compute item-total correlations for each item within its respective scale.
• Calculate Cronbach's alpha for each multi-item scale.
• Identify problematic items with low item-total correlations (<0.30) or those that substantially reduce alpha if deleted.
• Iteratively refine the scale by removing or revising problematic items and recalculating alpha until acceptable values are achieved.

Figure 1: Internal Consistency Assessment Workflow

Test-Retest Reliability

Conceptual Foundation

Test-retest reliability measures the consistency of results when the same test is administered to the same sample on two different occasions [71] [72]. This approach evaluates the temporal stability of a measurement instrument—the extent to which it produces similar results over time when the underlying construct being measured is assumed to be stable [71] [73].

For TPB constructs in reproductive health research, test-retest reliability assumes that the psychological variables being measured (attitudes, subjective norms, perceived behavioral control, intentions) remain relatively stable over the test-retest interval. This assumption is generally reasonable for deeply held attitudes and beliefs about health behaviors, though the appropriate time interval must be carefully selected to balance recall effects with genuine attitude stability [71].

Measurement Approaches

Test-retest reliability is typically quantified using the correlation coefficient (Pearson's r) between scores from the two testing occasions [71] [72]. The interpretation guidelines for test-retest correlations are:

• r ≥ 0.80: Good reliability [72]
• r = 0.70 - 0.80: Acceptable reliability [73]
• r < 0.70: Questionable reliability

The intraclass correlation coefficient (ICC) is also commonly used for test-retest reliability as it accounts for both correlation and agreement between measurements.

Table 2: Test-Retest Reliability Implementation Parameters

Aspect	Considerations	Recommendations for TPB Questionnaires
Time Interval	Balance between recall effects and actual change	2-4 weeks for TPB constructs [71]
Sample Characteristics	Stability of population on measured constructs	Representative of target population; exclude those experiencing interventions
Administration Conditions	Consistency between testing occasions	Same environment, instructions, and format
Statistical Analysis	Correlation and agreement measures	Pearson's r or ICC; paired t-test for systematic change

Application in Reproductive Health Research

In reproductive health research utilizing TPB frameworks, test-retest reliability provides critical evidence that the questionnaire measures stable constructs rather than capturing transient states:

• A medication knowledge assessment questionnaire for hemodialysis patients demonstrated test-retest reliability through administration on day 1 and after one week to the same patients, though specific correlation values were not provided [76].
• Research on non-adherence with medical orders established test-retest reliability (r = 0.70) with an appropriate time interval between measurements [77].
• TPB studies typically employ test-retest intervals of 2-4 weeks, sufficiently long to reduce recall effects but short enough to assume relative stability of psychological constructs [71].

Implementation Protocol

Procedure for Assessing Test-Retest Reliability:

• Select an appropriate time interval based on the stability of the constructs (typically 2-4 weeks for TPB questionnaires).
• Administer the TPB questionnaire to a representative sample (minimum 30-50 participants for adequate power).
• Ensure consistent administration conditions between time points (same setting, instructions, and format).
• Re-administer the questionnaire to the same participants after the predetermined interval.
• Calculate correlation coefficients between time points for each TPB construct and total scores.
• Assess systematic changes using paired t-tests to identify potential practice effects or actual attitude changes.
• Interpret results considering both correlation magnitude and absence of systematic change.

Figure 2: Test-Retest Reliability Assessment Workflow

Split-Half Reliability

Conceptual Foundation

Split-half reliability assesses internal consistency by dividing a measurement instrument into two equivalent halves and correlating the scores from these halves [73] [74]. This approach effectively treats the two halves as parallel forms administered simultaneously, providing an estimate of how consistently the instrument measures the underlying construct [73].

For TPB questionnaires, split-half reliability is particularly valuable when researchers want an internal consistency estimate that doesn't require multiple testing occasions (like test-retest) but provides a different perspective than Cronbach's alpha. It operates on the principle that if all items measure the same construct, responses to randomly divided halves should be highly correlated.

Measurement Approaches

The most common approaches to splitting instruments include:

• First-half vs. second-half: Simple but potentially problematic if items increase in difficulty or complexity.
• Odd-even split: Generally preferable as it distributes potential order effects.
• Random assignment: Optimal approach, minimizes systematic biases.

After computing the correlation between halves (r₁₂), the Spearman-Brown prophecy formula is applied to estimate reliability for the full instrument:

rₜₜ = (2 × r₁₂) / (1 + r₁₂)

Where rₜₜ is the estimated reliability of the full test.

Interpretation guidelines for split-half reliability are similar to other reliability measures, with values ≥ 0.80 generally considered acceptable [73] [72].

Relationship to Other Internal Consistency Measures

Split-half reliability shares conceptual ground with other internal consistency measures but differs in important ways:

• Unlike Cronbach's alpha, which is the average of all possible split-half correlations, a single split-half correlation provides one estimate of this average [73].
• Split-half reliability can be more sensitive to multidimensionality in scales than Cronbach's alpha.
• For TPB questionnaires with multiple distinct constructs, split-half reliability should be calculated separately for each theoretical construct rather than for the entire questionnaire.

Implementation Protocol

Procedure for Assessing Split-Half Reliability in TPB Questionnaires:

• Administer the complete TPB questionnaire to an appropriate sample.
• Divide items for each construct into two halves using random assignment or odd-even splitting.
• Calculate total scores for each half for each participant.
• Compute the correlation coefficient between the two halves for each theoretical construct.
• Apply the Spearman-Brown prophecy formula to estimate reliability of the full scale.
• Compare with Cronbach's alpha to identify potential multidimensionality issues.
• Report both split-half and alpha coefficients for comprehensive reliability evidence.

Figure 3: Split-Half Reliability Assessment Workflow

Comparative Analysis and Method Selection

Integrated Reliability Assessment Framework

Each reliability method offers distinct insights, and a comprehensive validation strategy for TPB questionnaires in reproductive health research should incorporate multiple approaches:

Table 3: Comparative Analysis of Reliability Methods for TPB Questionnaires

Method	Key Strength	Key Limitation	Optimal Use Case in TPB Research
Internal Consistency (Cronbach's Alpha)	Assesses item homogeneity; single administration	Assumes unidimensionality; sensitive to number of items	Initial scale development; multi-item TPB constructs
Test-Retest	Evaluates temporal stability; practical significance	Assumes construct stability; vulnerable to recall effects	Establishing measure stability over time for longitudinal TPB studies
Split-Half	Provides alternative internal consistency estimate	Affected by splitting method; less stable than alpha	Supplementary evidence when alpha may be inflated

Selection Guidelines for Reproductive Health Research

The choice of reliability methods should be guided by research objectives, instrument characteristics, and practical constraints:

• For newly developed TPB questionnaires: Begin with internal consistency analysis, followed by test-retest reliability with an appropriate interval [71] [77].
• For brief scales with few items: Prioritize test-retest reliability, as Cronbach's alpha is sensitive to scale length.
• For multidimensional TPB constructs: Report internal consistency for each dimension separately rather than for heterogeneous total scores.
• For cross-cultural adaptation: Include test-retest reliability to ensure measurement stability across cultural contexts [32] [48].

Essential Research Reagents and Tools

Table 4: Essential Research Reagents and Methodological Tools for Reliability Assessment

Tool/Reagent	Function in Reliability Assessment	Application Example	Considerations for TPB Research
Statistical Software (SPSS, R, Stata)	Calculate reliability coefficients	Cronbach's alpha, test-retest correlations [3] [75] [48]	Ensure appropriate procedures for Likert-scale data
Pilot Participant Pool	Provide data for initial reliability testing	5-10 participants per questionnaire item [75] [77]	Representative of target reproductive health population
Standardized Administration Protocols	Ensure consistent measurement conditions	Identical instructions, setting, and format across administrations [71]	Critical for test-retest reliability in multi-site studies
Reliability Coefficient Calculators	Compute various reliability estimates	Online or standalone reliability analysis tools	Verify calculations manually for publication
Item Analysis Templates	Identify problematic items	Item-total correlation tables, distractor analysis	Predefined criteria for item retention/revision

Establishing the reliability of TPB questionnaires through rigorous assessment of internal consistency, test-retest reliability, and split-half methods is a fundamental prerequisite for valid reproductive health research. Each method offers complementary evidence about different aspects of measurement consistency, and their strategic application throughout questionnaire development and validation strengthens the scientific rigor of subsequent behavioral research.

In the context of reproductive health, where TPB-based interventions target critical outcomes ranging from contraceptive use to maternal healthcare utilization, reliable measurement is not merely a methodological concern but an ethical imperative. By implementing the protocols and guidelines outlined in this technical guide, researchers can ensure their instruments produce consistent, dependable measurements that form a solid foundation for understanding and influencing reproductive health behaviors.

The Theory of Planned Behavior (TPB) serves as a foundational framework for understanding and predicting health behaviors in reproductive health research. A rigorously developed and validated TPB questionnaire is not merely a data collection tool but a diagnostic instrument that allows researchers to predict behavioral intentions and evaluate the effectiveness of targeted interventions. This guide provides a technical overview for assessing the psychometric properties of TPB questionnaires and applying them within experimental protocols to measure intervention efficacy in reproductive health contexts. The proper application of this tool enables researchers to move beyond correlation to causation, identifying not just what behaviors occur but why they persist, thereby informing more effective public health strategies and clinical interventions aimed at improving reproductive health outcomes across diverse populations.

Quantitative Synthesis of TPB Intervention Efficacy

Data from multiple intervention studies demonstrate the significant impact that TPB-based educational programs can have on reproductive health constructs. The table below synthesizes key findings from randomized controlled trials and quasi-experimental studies conducted in various populations.

Table 1: Efficacy of TPB-Based Interventions on Reproductive Health Constructs

Population & Study Design	Key Constructs Measured	Pre-Post Intervention Changes	Statistical Significance
High School Girls (RCT) [3]	Attitude, Subjective Norms, Perceived Behavioral Control, Intention, Behavior	Attitude: +16.8 pointsSubjective Norms: +16.4 pointsPerceived Control: +18.0 pointsIntention: +18.4 pointsBehavior: +18.5 points	p < 0.001 for all constructs
Women with Endometriosis (RCT) [4]	Knowledge, Attitude, Subjective Norms, Behavioral Intention, Overall Reproductive Health	Significant improvement in all TPB constructs and overall reproductive health score	p < 0.05
Women in Rural Tanzania (Cross-Sectional) [64]	Attitude, Subjective Norms, Perceived Behavioral Control, Modern FP Uptake	Positive Attitude (AOR 2.3)Positive Perceived Control (AOR 6.0)	p < 0.05
Female University Students (Quasi-Experimental) [78]	Protection Motivation, Self-Efficacy, Behavior	Significant increase in PMT constructs and protective behaviors	p < 0.05

These quantitative results consistently demonstrate that interventions grounded in the TPB framework produce statistically significant improvements across all major constructs of the model. The data reveals that the largest effect sizes are often observed in perceived behavioral control and behavioral intention, which are the most proximal determinants of actual behavior change. Furthermore, the consistency of results across diverse populations and cultural contexts—from Iranian adolescents to Tanzanian women of reproductive age—supports the cross-cultural validity of the TPB model in reproductive health research.

Experimental Protocols for Validation and Intervention

Protocol 1: Validating the TPB Questionnaire

A robust validation process is essential to ensure that a TPB questionnaire accurately measures the constructs it purports to measure.

Table 2: Key Psychometric Validation Steps and Criteria

Validation Step	Methodology	Acceptance Criteria
Content Validity	Expert panel review (5-10 experts) using 4-point scale for relevance and clarity [26] [4]	Item-Level CVI (I-CVI) ≥ 0.78Scale-Level CVI (S-CVI) ≥ 0.90Content Validity Ratio (CVR) ≥ 0.62 [79]
Face Validity	Cognitive interviews with 5-10 target population members [79] [58]	Participants correctly interpret item intent and terminology
Construct Validity	Exploratory Factor Analysis (EFA) with Varimax rotation [59] [79] [58]	KMO > 0.6, Bartlett's Test p < 0.05Factor loading > 0.4 on primary factor
Internal Consistency	Cronbach's Alpha calculation [78] [79] [4]	α ≥ 0.70 for each construct
Test-Retest Reliability	Re-administration to subset after 2-week interval [79] [4]	Intraclass Correlation Coefficient (ICC) ≥ 0.70

The validation process begins with item pool generation through literature review and qualitative methods (e.g., focus groups, interviews). The subsequent quantitative validation should include a sufficient sample size, typically following a 10:1 participant-to-item ratio. For TPB questionnaires measuring 4-5 constructs with 20-25 total items, a sample of 200-250 participants is generally adequate for robust factor analysis. The exploratory factor analysis should confirm the hypothesized structure of attitudes, subjective norms, perceived behavioral control, and behavioral intentions, with careful attention to cross-loadings and communalities.

Protocol 2: Implementing a TPB-Based Intervention Trial

Once a TPB questionnaire is validated, it can be deployed to assess the effectiveness of reproductive health interventions. The following protocol outlines a standardized approach for such trials.

Sample Recruitment and Randomization:

• Utilize permuted block randomization with computer-generated sequences to assign participants to intervention and control groups [3] [4]
• Ensure adequate sample size through power analysis; typically 35-50 participants per group for RCTs with medium effect sizes [3] [4]
• Implement allocation concealment using sealed envelopes or electronic systems to prevent selection bias

Intervention Structure and Delivery:

• Develop multi-session interventions (typically 3-4 sessions) delivered over several weeks [3] [78] [4]
• Incorporate active learning strategies including lectures, group discussions, role-playing, and educational materials (e.g., booklets) [3] [78]
• For the control group, provide standard care or minimal education, with optional provision of educational materials post-trial

Data Collection and Analysis:

• Administer the validated TPB questionnaire at baseline, immediately post-intervention, and at follow-up time points (typically 1-3 months post-intervention) [3] [4]
• Use Analysis of Covariance (ANCOVA) with baseline scores as covariates to compare post-intervention outcomes between groups [3]
• For predictive validity assessment, employ logistic regression to examine relationships between TPB constructs and subsequent behavioral outcomes [64]

The Scientist's Toolkit: Essential Research Reagents and Materials

Table 3: Essential Reagents and Materials for TPB Reproductive Health Research

Item Category	Specific Examples	Research Function
Validated Questionnaires	TPB Construct Scales (Attitude, Subjective Norms, PBC, Intention) [3] [26]Reproductive Health Outcome Measures [59] [79] [4]	Quantifies theoretical constructs and measures behavioral outcomes
Statistical Analysis Tools	SPSS (v16+), R Studio, AMOS for SEM [80] [58]	Performs factor analysis, reliability testing, and multivariate statistics
Educational Intervention Materials	Tailored Booklets [3]Structured Session Plans [78] [4]Multimedia Content [81]	Delivers standardized educational content to intervention groups
Digital Data Collection Platforms	Open Data Kit (ODK) [81]Microsoft Forms [58]	Enables efficient data collection and management
Quality Assurance Tools	Content Validation Forms [26]Interview Guides for Cognitive Testing [79]	Ensures methodological rigor and validity

Analyzing and Interpreting Predictive Validity Data

The predictive validity of a TPB questionnaire is established when its constructs significantly correlate with subsequent behaviors. Analysis of predictive validity involves several sophisticated statistical approaches:

Structural Equation Modeling (SEM): SEM allows researchers to test the complete TPB model, including both direct and indirect effects of constructs on behavior. Studies in sub-Saharan Africa have used SEM to demonstrate strong relationships between women's literacy and contraception prevalence (total standardized effect size = 0.79, 95% CI: 0.74-0.83) [80]. SEM provides comprehensive fit indices (χ², RMSEA, CFI) to evaluate how well the hypothesized model fits the observed data.

Logistic Regression Analysis: For dichotomous outcomes (e.g., modern family planning uptake vs. non-uptake), logistic regression quantifies how TPB constructs predict behavioral outcomes. Research in Tanzania found that positive attitudes (AOR = 2.307) and high perceived behavioral control (AOR = 6.015) significantly predicted modern family planning uptake, while intention showed an unexpected inverse relationship (AOR = 0.038) [64]. This highlights the complexity of intention-behavior relationships in reproductive health contexts.

Mediation and Moderation Analysis: Advanced analysis should test whether the relationship between intention and behavior is mediated by perceived behavioral control or moderated by demographic variables. These analyses help explain circumstances under which TPB constructs are more or less predictive of reproductive health behaviors.

The application of TPB questionnaires in reproductive health research provides a powerful methodological approach for both predicting health behaviors and evaluating intervention effectiveness. Through rigorous validation protocols and carefully designed intervention trials, researchers can generate robust evidence about the cognitive determinants of reproductive health behaviors. The consistent findings across diverse populations underscore the utility of the TPB framework, while occasional divergences in predictive patterns highlight the importance of contextual factors. Future research should focus on developing more brief, culturally adapted versions of TPB questionnaires for use in clinical settings and exploring digital approaches to intervention delivery and assessment. By adhering to the methodologies outlined in this guide, researchers can contribute to the advancement of evidence-based reproductive health programs that effectively address the complex interplay between psychological constructs and health behaviors.

The Theory of Planned Behavior (TPB) serves as a dominant conceptual framework for understanding and predicting health behaviors, particularly in the complex domain of reproductive health. This whitepaper provides a comparative analysis of TPB against other behavioral frameworks, examining its application in reproductive health questionnaire research. The ability to accurately measure and predict behaviors related to sexual health, contraceptive use, fertility intentions, and maternal care is paramount for researchers, scientists, and drug development professionals working to improve health outcomes. Through a systematic examination of empirical evidence and methodological approaches, this analysis delineates the comparative strengths and limitations of TPB in this specialized field, providing a technical guide for its application in research and intervention design.

The persistent challenges in reproductive health—from rising STI rates among adolescents to maternal mortality and declining fertility rates—necessitate theoretically grounded approaches to behavior change [3] [82] [83]. TPB offers a structured framework for investigating the psychological determinants of these behaviors, positioning it as a valuable tool for developing evidence-based interventions and precise research instruments. This paper critically evaluates TPB's efficacy relative to alternative models, providing researchers with actionable insights for model selection and application in reproductive health studies.

Theoretical Foundations of TPB in Reproductive Health

The Theory of Planned Behavior, developed by Icek Ajzen, posits that behavioral intention—the immediate precursor to behavior—is influenced by three core constructs: attitude (personal evaluation of the behavior), subjective norms (perceived social pressure), and perceived behavioral control (PBC; perceived ease or difficulty of performing the behavior) [27]. PBC can also directly influence behavior itself, particularly when volitional control is limited. In reproductive health contexts, where behaviors are often sensitive and influenced by complex social and cultural factors, this tripartite structure provides a comprehensive framework for understanding behavioral determinants.

TPB's application in reproductive health research typically involves developing standardized questionnaires that measure these constructs in relation to specific reproductive behaviors. The model has demonstrated remarkable adaptability across diverse reproductive health domains, including preconception care [48], maternal health service utilization [32], adolescent sexual health [3] [84], and fertility intentions [82]. This flexibility stems from its ability to incorporate behavior-specific beliefs, making it particularly valuable for designing targeted interventions in culturally sensitive contexts where reproductive behaviors are heavily influenced by social norms and perceived constraints.

Figure 1: TPB Conceptual Framework. The model illustrates how behavioral, normative, and control beliefs influence the core TPB constructs, which together determine behavioral intention and actual behavior. (Adapted from Ajzen's Theory of Planned Behavior)

Empirical Evidence: TPB Applications in Reproductive Health

Quantitative Evidence Across Reproductive Health Domains

TPB has been empirically validated across diverse reproductive health contexts, demonstrating consistent predictive power for behavioral intentions. The table below summarizes key quantitative findings from recent studies:

Table 1: TPB Application Across Reproductive Health Domains

Health Domain	Study Population	Sample Size	Key TPB Predictors	Variance Explained (R²)	Citation
Preconception Care	Reproductive-age women, Ethiopia	415	Perceived Behavioral Control (β=0.263), Attitude (β=0.201), Subjective Norms (β=0.158)	Not reported	[48]
Adolescent SRH Education	High school girls, Iran	578	Attitude (diff=16.8), Subjective Norms (diff=16.4), Perceived Behavioral Control (diff=18.0)	Significant improvement (p<0.001)	[3]
Facility Birth Intention	Expecting couples, Tanzania	1092	Subjective Norms (pregnant women: M=30.21), Perceived Behavioral Control (male partners)	Weak influence of TPB domains	[32]
Sexual Health Discussions	Healthcare providers, Iran	208	Attitude (positive effect), Perceived Behavioral Control (negative effect)	Intention: 24%, Behavior: 37%	[27]

Adolescent Sexual and Reproductive Health

In adolescent sexual and reproductive health, TPB-based interventions have demonstrated significant effectiveness. A randomized controlled trial among high school girls in Iran showed that a TPB-based educational intervention produced significant improvements in attitude (difference=16.8; 95% CI: 15.3, 18.3), subjective norms (16.4; 95% CI=14.83 to 18.11), perceived behavioral control (18.0; 95% CI: 16.6, 19.4), and behavioral intention (18.4%; 95 CI: 14.8, 18.3) compared to the control group [3]. The intervention, which included training sessions, booklets, and parental workshops, effectively reduced intentions toward high-risk sexual behaviors by targeting the fundamental TPB constructs.

A systematic review of 11 studies on sexual health education based on TPB confirmed its efficacy in preventing risky sexual behavior among adolescents [84]. The review found that factors influencing adolescents' intentions are closely related to TPB constructs, including attitudes, subjective norms, and behavioral control. The authors concluded that "sexual health education based on TPB can be an appropriate intervention to prevent risky sexual behavior among adolescents, as its priority is to strengthen behavioral intentions by enhancing attitudes, subjective norms, and behavioral control through continuous health education" [84].

Maternal Health and Preconception Care

In maternal health contexts, TPB has been applied to understand and predict health facility utilization for childbirth. A study in Tanzania found that while the vast majority of expecting couples had intentions to use health facilities for childbirth (91.2% of pregnant women and 89.7% of their partners), the TPB constructs showed varying influence patterns between genders [32]. Among pregnant women, only perceived subjective norms significantly influenced intention, while among male partners, only perceived behavioral control showed significant influence, suggesting gender-specific approaches may be necessary for effective interventions.

Research on preconception care in Ethiopia identified perceived behavioral control as the strongest predictor (β=0.263, p<0.001) of intention to use preconception care, followed by attitude (β=0.201, p=0.001) and subjective norms (β=0.158, p=0.006) [48]. These findings suggest that interventions aimed at promoting preconception care should focus on enhancing women's confidence in their ability to access and utilize these services, while also addressing attitudes and social norms.

Methodological Approaches: TPB Questionnaire Development

TPB Instrumentation Protocol

The development of psychometrically sound TPB questionnaires requires systematic approaches. The following workflow outlines the standardized methodology for TPB instrument development in reproductive health research:

Figure 2: TPB Questionnaire Development Workflow. This protocol outlines the sequential steps for developing and validating TPB-based instruments in reproductive health research.

The instrumentation process typically begins with an elicitation study involving qualitative methods (interviews, focus groups) with the target population to identify salient behavioral, normative, and control beliefs specific to the reproductive health behavior [27] [48]. For example, a study developing a TPB questionnaire on healthcare providers' intentions to discuss sexual issues with postmenopausal women conducted 27 in-depth interviews, identifying 226 codes, 54 subcategories, and 18 categories that were classified under TPB constructs [27].

Questionnaire development follows TPB guidelines with both direct and belief-based (indirect) measures. Direct measures assess general evaluations, while indirect measures examine underlying specific beliefs [27]. A typical TPB questionnaire in reproductive health includes:

• Direct attitude measures: Typically 4-5 semantic differential items on a 5-point scale (e.g., harmful-beneficial, unpleasant-pleasant)
• Direct subjective norms: Typically 4-5 items measuring perceived social pressure
• Direct perceived behavioral control: Typically 4-5 items measuring confidence and controllability
• Behavioral intention: Typically 4-5 items measuring readiness to engage in the behavior
• Indirect measures: Behavioral beliefs × outcome evaluations, normative beliefs × motivation to comply, control beliefs × perceived power

Key Research Reagents and Materials

Table 2: Essential Research Reagents for TPB Reproductive Health Studies

Reagent/Instrument	Specifications	Application in TPB Research
TPB Questionnaire	Direct & belief-based measures; 5-point Likert/semantic differential scales; 55-132 items	Primary data collection on TPB constructs and behavioral intentions
Clinical Scenarios	5-6 simulated patient encounters; 50-80 words each	Measurement of simulated behavioral intention in clinical contexts [27]
Elicitation Study Guide	Semi-structured interview format; open-ended questions on behavioral advantages/disadvantages, referents, facilitators/barriers	Identification of salient beliefs for questionnaire development [27] [48]
Psychometric Validation Package	Cronbach's α, test-retest reliability, CVI, CVR, CFA	Establishment of instrument reliability and validity [27]

Comparative Analysis with Alternative Frameworks

TPB vs. Trans-Theoretical Model (TTM) in Reproductive Health

The Trans-Theoretical Model (TTM), also known as the Stages of Change model, offers an alternative framework for understanding health behavior change. While TPB focuses on the psychological predictors of intention and behavior, TTM conceptualizes behavior change as a process that occurs through a series of stages: precontemplation, contemplation, preparation, action, and maintenance [82]. A key strength of TTM is its ability to tailor interventions to an individual's readiness to change.

In reproductive health contexts, researchers have begun integrating both models to leverage their complementary strengths. A study protocol on intention to have a child combines TPB and TTM, using TPB to identify determinants of behavior and TTM to measure stages of change [82]. This integrated approach allows researchers to not only understand what influences behavioral intentions but also how individuals progress toward actual behavior change. The TTM component helps identify whether participants are in precontemplation, contemplation, or preparation stages regarding childbearing decisions, enabling more targeted interventions.

TPB vs. Technology Acceptance Model (TAM)

The Technology Acceptance Model (TAM), originally developed to understand technology adoption, shares similarities with TPB but focuses specifically on perceived usefulness and perceived ease of use as key determinants of behavioral intention [85]. While TAM has been extended to consumer goods acceptance, its application in reproductive health has been more limited compared to TPB.

TPB offers a broader framework that incorporates social influences (subjective norms) and control factors beyond ease of use, making it potentially more comprehensive for understanding reproductive health behaviors that are strongly influenced by social and cultural contexts. However, TAM's parsimony may be advantageous in specific reproductive health technology contexts, such as the adoption of digital health tools or fertility apps.

Discussion: Research Implications and Future Directions

The accumulated evidence demonstrates TPB's robust applicability across diverse reproductive health domains, though with varying predictive power depending on cultural contexts and specific behaviors. The framework provides researchers with a systematic approach to identifying key determinants of reproductive health behaviors, enabling more targeted and effective interventions. Future research should focus on enhancing TPB's predictive power through cultural adaptations and integration with complementary models.

Several important methodological considerations emerge from this analysis. First, the gender-specific patterns in TPB construct influence [32] highlight the need for gender-sensitive approaches in reproductive health research. Second, the varying strength of TPB predictors across different behaviors suggests that researchers should conduct preliminary studies to identify the most relevant constructs for their specific context. Finally, the integration of TPB with other models like TTM shows promise for addressing both the determinants and the process of behavior change in complex reproductive health decisions [82].

For drug development professionals and clinical researchers, TPB offers a valuable framework for understanding medication adherence, contraceptive use, and adoption of new reproductive technologies. By identifying the key attitudes, social norms, and control factors influencing these behaviors, pharmaceutical companies can develop more effective patient support programs and educational materials that address the actual determinants of medication behavior rather than assuming purely rational decision-making processes.

This comparative analysis demonstrates that the Theory of Planned Behavior provides a robust, flexible framework for reproductive health research, with strong empirical support across diverse cultural contexts and behavioral domains. Its structured approach to measuring behavioral determinants offers significant advantages for developing targeted interventions and precise research instruments. While alternative models like TTM offer complementary strengths, TPB's comprehensive coverage of cognitive, social, and control factors makes it particularly valuable for understanding the complex decision-making processes in reproductive health.

The successful application of TPB in reproductive health questionnaire research depends on rigorous methodological approaches, including thorough elicitation studies, psychometric validation, and cultural adaptation. Future research should continue to refine TPB-based instruments and explore integrated models that leverage the complementary strengths of multiple theoretical frameworks. For researchers, scientists, and drug development professionals, TPB provides an essential toolkit for advancing reproductive health outcomes through theoretically grounded, evidence-based approaches.

Conclusion

The systematic development and validation of TPB-based questionnaires are paramount for advancing reproductive health research and clinical practice. This synthesis demonstrates that a rigorously constructed tool does more than measure intent; it provides a granular understanding of the behavioral, normative, and control beliefs that drive health decisions, from adolescent sexual health to clinical provider behavior. For researchers and drug development professionals, this enables the design of more potent, evidence-based interventions and the precise measurement of their impact. Future efforts must prioritize theory-driven design, invest in implementation research to link TPB constructs directly to behavioral and health outcomes, and develop standardized, validated measures that allow for cross-population comparisons. By adopting this comprehensive approach, the scientific community can significantly enhance the quality and effectiveness of reproductive health programming and pharmaceutical development.

References

• 1. - Wikipedia Theory of planned behavior [https://en.wikipedia.org/wiki/Theory_of_planned_behavior]
• 2. The Theory : Selected Recent Advances and... of Planned Behavior [https://pmc.ncbi.nlm.nih.gov/articles/PMC7909498/]
• 3. The Effect of a Theory of Planned Behavior-based ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7189942/]
• 4. Reproductive health of women with endometriosis: an ... [https://mefj.springeropen.com/articles/10.1186/s43043-023-00129-7]
• 5. Intention to Use and Its Predictors Towards Preconception ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8379389/]
• 6. The Theory of Planned Behavior: Behavioral Intention [https://www.simplypsychology.org/theory-of-planned-behavior.html]
• 7. Theory of Reasoned Action/Planned Behavior [https://www.ruralhealthinfo.org/toolkits/health-promotion/2/theories-and-models/reasoned-action]
• 8. A systematic review of the theory of planned behaviour ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10506128/]
• 9. Visualization Techniques in Healthcare Applications [https://pmc.ncbi.nlm.nih.gov/articles/PMC9741729/]
• 10. Progress on theory of planned behavior research: advances in research synthesis and agenda for future research - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC11893630/]
• 11. The Effect of a Theory of Planned Behavior-based Educational Intervention on Sexual and Reproductive Health in Iranian Adolescent Girls: A Randomized Controlled Trial - PMC [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7189942/]
• 12. Stimulating Innovations in the Measurement of Parenting ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC3746250/]
• 13. Design and psychometric properties of a questionnaire for ... [https://pubmed.ncbi.nlm.nih.gov/35980622/]
• 14. Reproductive health of women with endometriosis: an improving educational intervention based on the planned behavior theory | Middle East Fertility Society Journal [https://link.springer.com/article/10.1186/s43043-023-00129-7]
• 15. Quantitative model validation techniques: new insights [https://arxiv.org/abs/1206.5014]
• 16. Application of theory of planned behavior to predictors of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11913192/]
• 17. Application of the Theory of Planned Behavior to couples ... [https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0221526]
• 18. A Validation Study of the Total Teen Assessment - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC12329912/]
• 19. Development of the oral, mental, and sexual reproductive ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12611828/]
• 20. Sexual and Reproductive Health Service Seeking Scale ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10837854/]
• 21. Bad Behavior: Improving Reproducibility in Behavior Testing [https://pmc.ncbi.nlm.nih.gov/articles/PMC6100770/]
• 22. Reproductive health literacy scale: a tool to measure the ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-025-01959-6]
• 23. Influencing factors and mechanism of physicians' strategic ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11964819/]
• 24. R数据分析：如何用lavaan包做结构方程模型，实例解析 [https://blog.csdn.net/tm_ggplot2/article/details/115408158]
• 25. Chapter 2 验证性因素分析| 结构方程模型的R实现(草稿) [https://bookdown.org/l978010919/sem_using_R/cfa.html]
• 26. Validating a theory of planned behavior questionnaire for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11130381/]
• 27. Using the Theory of Planned Behavior to Predict Sexual ... [https://brieflands.com/journals/mejrh/articles/154336]
• 28. Direct Versus Indirect Assessment of Student Learning [https://www.niu.edu/citl/resources/guides/direct-versus-indirect-assessment-of-student-learning.shtml]
• 29. Likert Scale Best Practices, Examples, And Survey Questions [https://www.surveymonkey.com/mp/likert-scale-questions-and-mistakes-to-avoid/]
• 30. 20 Likert Scale Questions and Examples for 2022 [https://www.customerthermometer.com/customer-surveys/likert-scale-examples-questions-what-you-need-to-know/]
• 31. Design of tables for the presentation and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10346464/]
• 32. Using the theory of planned behavior to explain birth in health ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-020-0851-1]
• 33. Cognitive Interviewing | CCQDER [https://www.cdc.gov/nchs/ccqder/question-evaluation/cognitive-interviewing.html]
• 34. Application of cognitive interviewing to improve self ... [https://www.sciencedirect.com/science/article/abs/pii/S1551741113001071]
• 35. Role of cognitive testing in the development of the CAHPS ... [https://pubmed.ncbi.nlm.nih.gov/16316437/]
• 36. Intervention to Improve Menstrual Health Among ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9742407/]
• 37. A Systematic Review of Adolescent Girl Program ... [https://www.sciencedirect.com/science/article/pii/S1054139X17308662]
• 38. A Randomized Controlled Trial to Evaluate the ... [https://www.mdpi.com/2673-7051/5/3/33]
• 39. The Influence of Social Desirability on Sexual Behavior ... [https://link.springer.com/article/10.1007/s10508-021-02197-0]
• 40. The theory of planned behavior [https://www.sciencedirect.com/science/article/pii/074959789190020T]
• 41. Social Desirability Bias in Sexual Behavior Reporting - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC4023461/]
• 42. Interviewing adolescent girls about sexual and reproductive ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-021-01318-1]
• 43. comparing validated measures of intention and assessing the ... [https://implementationsciencecomms.biomedcentral.com/articles/10.1186/s43058-020-00050-4]
• 44. Introduction to Structural Equation Modeling (SEM) in R with ... [https://stats.oarc.ucla.edu/r/seminars/rsem/]
• 45. Chapter 11 Structural Equation modelling with lavaan [https://mspeekenbrink.github.io/sdam-r-companion/structural-equation-modelling-with-lavaan.html]
• 46. Introduction to causal diagrams (DAGs) - francisco yirá's blog [https://www.franciscoyira.com/post/2021-07-11-diagramas-causalidad-cap-3-causal-inference-mixtape/]
• 47. A Framework for Understanding Modifications to Measures ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC3768261/]
• 48. Determining behavioral intention and its predictors toward ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11344351/]
• 49. 青少年女性性健康促进量表的构建与信效度研究 [https://ccj.pku.edu.cn/Article/info?aid=215339071]
• 50. 男性孕前健康信念与行为意向对实际行为的预测作用 [https://www.ebiotrade.com/newsf/2025-7/20250710235530629.htm]
• 51. A systematic approach to cross-cultural adaptation of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4154745/]
• 52. translation, cultural adaptation and psychometric test of the ... [https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-020-09963-0]
• 53. 慢性疾病与健康人群身后器官捐献态度比较：埃及社会文化 ... [https://www.ebiotrade.com/newsf/2025-11/20251108034021627.htm]
• 54. 自杀相关心理健康服务倾向问卷的编制及信效度检验 [https://zh.zhhlzzs.com/CN/10.3761/j.issn.0254-1769.2022.19.010]
• 55. Progress on the application of system dynamics model in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12571711/]
• 56. Structural equation modeling (SEM) in L2 writing research [https://www.sciencedirect.com/science/article/pii/S2772766125000230]
• 57. An Integrated Structural Equation Modelling and Machine ... [https://www.mdpi.com/2673-9909/5/3/105]
• 58. Validation of a Questionnaire on Sexual and Reproductive ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10469261/]
• 59. Designing and validation of a reproductive health need ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-022-01342-9]
• 60. Text has enhanced contrast | ACT Rule | WAI [https://www.w3.org/WAI/standards-guidelines/act/rules/09o5cg/]
• 61. Elements must meet minimum color contrast ratio thresholds [https://dequeuniversity.com/rules/axe/4.8/color-contrast]
• 62. Text color must provide high contrast with its background [https://docs.acquia.com/web-governance/text-color-must-provide-high-contrast-its-background]
• 63. Content validity: judging the relevance, ... [https://www.sciencedirect.com/science/article/pii/S0895435625002124]
• 64. Using Theory of Planned Behaviour to Assess the ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12083717/]
• 65. Using the Theory of Planned Behavior to predict information [https://www.sciencedirect.com/science/article/pii/S1386505624004362]
• 66. Item generation and establishing face and content validity ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12410852/]
• 67. Face and content validity of the EMPOWER-UP questionnaire [https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-024-02727-5]
• 68. Confirmatory Factor Analysis (CFA): A Detailed Overview [https://www.statisticssolutions.com/free-resources/directory-of-statistical-analyses/confirmatory-factor-analysis/]
• 69. Exploratory factor analysis on the development and ... [https://www.nature.com/articles/s41598-025-04517-z]
• 70. Confirmatory and Exploratory Factor Analysis for Validating ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4804052/]
• 71. The 4 Types of Reliability | Definitions, Examples, Methods [https://www.scribbr.com/methodology/types-of-reliability/]
• 72. Reliability and Validity of Measurement [https://opentextbc.ca/researchmethods/chapter/reliability-and-validity-of-measurement/]
• 73. Types of Reliability - Research Methods Knowledge Base [https://conjointly.com/kb/types-of-reliability/]
• 74. What is Reliability? (The Four Types to Know) [https://psychwithkeegan.com/2025/09/what-is-reliability-the-four-types-to-know/]
• 75. Validity and Reliability of a Questionnaire on Attitudes ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11428262/]
• 76. Development, Validation, and Reliability of Medication ... [https://turkjnephrol.org/en/development-validation-and-reliability-of-medication-knowledge-assessment-questionnaire-among-hemodialysis-patients-137091]
• 77. Development and validation of a questionnaire to determine ... [https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-021-06147-3]
• 78. The effect of educational intervention based on protection ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11966937/]
• 79. Development and Psychometric Evaluation of a ... [https://brieflands.com/journals/semj/articles/20464]
• 80. Inter-relationships among key reproductive health ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9248058/]
• 81. Acceptability, feasibility and utility of a Mobile health family ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-019-0767-9]
• 82. An integrated theory based-educational intervention to ... [https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-024-01760-x]
• 83. Rewriting women's health: a content analysis of the Trump ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12613099/]
• 84. Sexual Health Education Based on the Theory of Planned ... [https://jurnal.globalhealthsciencegroup.com/index.php/IJGHR/article/view/4408]
• 85. Extending the technology acceptance model and ... [https://www.sciencedirect.com/science/article/pii/S2405844024038544]