Theory of Reasoned Action (TRA) - Fishbein & Ajzen (1975)

Theory Publication Information

Citation Information

Why Was the Model Created?

The Theory of Reasoned Action emerged from Fishbein and Ajzen’s attempt to resolve one of the most persistent and frustrating problems in social psychology: the weak relationship between attitudes and actual behavior. Throughout the 1960s and early 1970s, researchers had consistently found that individuals’ attitudes toward objects, behaviors, or institutions were remarkably poor predictors of their actual behavior. People might express favorable attitudes toward environmental protection yet not recycle. They might endorse exercise and healthy eating yet maintain sedentary lifestyles and poor diets. This attitude-behavior gap puzzled researchers and limited the practical applicability of attitude research.

Fishbein and Ajzen’s fundamental insight was that the problem lay not in attitudes themselves but in the level of specificity at which attitudes were measured. Most attitude research assessed general attitudes toward general targets, then attempted to predict specific behaviors. Their key proposition was that behavior is most directly predicted not by attitudes but by behavioral intentions - an individual’s conscious plan or decision to perform a behavior.

The development of TRA was driven by both theoretical and practical considerations. Theoretically, Fishbein and Ajzen sought to construct a parsimonious model identifying the minimal set of variables necessary to predict behavior. They proposed that behavioral intention, the immediate antecedent of behavior, is determined by exactly two factors: (1) the individual’s attitude toward the specific behavior and (2) the individual’s subjective norm regarding that behavior (perceived social pressure to perform or not perform the behavior). This elegant parsimony represented a significant theoretical advance. Rather than invoking numerous psychological and social variables, TRA suggested that behavior stems from these two primary determinants.

TRA was developed as a general social psychology theory, not specifically for technology adoption. It addressed practical needs across public health, organizational, and policy domains - predicting contraceptive use, energy conservation, occupational choices, and other behaviors. The technology adoption application came later when Davis (1989) adapted TRA’s attitude-intention-behavior structure to create the Technology Acceptance Model, substituting Perceived Usefulness and Perceived Ease of Use as technology-specific belief constructs. TRA’s parsimonious structure and focus on behavioral intention made it readily adaptable to technology contexts.

Core Concepts and Definitions

TRA formalizes a small set of psychological constructs, each with precise operationalization:

• Behavioral Intention (BI):The individual’s conscious plan or decision to perform a specific behavior. TRA’s central mediating variable and the direct antecedent of behavior. Operationalized through intention items asking about willingness, plans, and likelihood of performing the behavior.
• Attitude Toward the Behavior (AB):The individual’s overall favorable or unfavorable evaluation of performing the specific behavior. Distinct from general attitudes toward objects, AB targets the behavior itself.
• Subjective Norm (SN):The individual’s perception of social pressure to perform or not perform the behavior. Captures perceptions of what important referent others think the individual should do.
• Behavioral Beliefs (bi):Beliefs about the likely consequences of performing the behavior. Combined with outcome evaluations (ei), they form the cognitive foundation of attitude (AB = Σbi · ei).
• Outcome Evaluations (ei):The individual’s evaluation of each consequence as good or bad, desirable or undesirable.
• Normative Beliefs (nj):Beliefs about whether specific referent individuals or groups think the individual should perform the behavior. Combined with motivation to comply (mj), they form the cognitive foundation of subjective norm (SN = Σnj · mj).
• Motivation to Comply (mj):The individual’s motivation to conform with each referent’s wishes.
• Behavior (B): The observable action performed by the individual. In technology adoption contexts, this is typically actual system use or adoption.
• Principle of Compatibility: A core theoretical requirement that attitude, intention, and behavior be measured at the same level of specificity (action, target, context, and time). Mismatches in specificity degrade predictive validity.

What Does the Model Measure?

The Theory of Reasoned Action is a measurement model. Fishbein and Ajzen (1975) operationalize each theoretical construct through multi-item scales that use bipolar adjective pairs (semantic differential) or agree/disagree Likert items. The measured constructs are:

• Behavioral Intention (BI):Direct measure of the user’s plan or willingness to perform the target behavior.
• Attitude Toward the Behavior (AB): Overall favorable/unfavorable evaluation of performing the behavior, typically via semantic-differential items (good-bad, wise-foolish, pleasant-unpleasant, beneficial-harmful).
• Subjective Norm (SN): Perception of social pressure from important referents to perform or not perform the behavior.
• Behavioral Beliefs (bi) × Outcome Evaluations (ei): Belief strength about each consequence multiplied by evaluation of that consequence; summed across salient beliefs to form the cognitive basis of AB.
• Normative Beliefs (nj) × Motivation to Comply (mj): Belief about whether each referent thinks the behavior should be performed multiplied by motivation to comply with that referent; summed across salient referents to form the cognitive basis of SN.
• Behavior (B): Observed action (typically through self-report or system log).

Fishbein and Ajzen (1975) provide construct definitions, item formats, and scoring procedures, including the expectancy-value sums (AB = Σbi · ei; SN = Σnj · mj) and the principle of compatibility (intent, attitude, and behavior must be matched on action, target, context, and time to preserve predictive validity). Subsequent studies report reliability and validity evidence for TRA scales across a wide range of behaviors.

Preceding Models or Theories

TRA synthesized and extended several prior traditions in social psychology and attitude research:

• Fishbein’s Expectancy-Value Model of Attitude (1963): Provided the foundational belief-evaluation product formulation underlying the attitude construct in TRA.
• Classical attitude theory: Long-standing research on the attitude-behavior relationship from Allport, Thurstone, and subsequent scholars that TRA sought to rescue by introducing intention as a proximal mediator.
• Cognitive consistency theories:Including Festinger’s cognitive dissonance theory and Heider’s balance theory, which TRA drew on by formalizing beliefs as the cognitive foundation of attitudes.
• Early attitude-behavior research (Ajzen & Fishbein, 1969-1973):Joint empirical work establishing that attitudes toward specific behaviors predicted behavior better than attitudes toward objects.
• Learning theory approaches to behavior: Behaviorist traditions from which TRA departed by locating behavior determination in cognitive and social psychological variables rather than in reinforcement histories alone.
• Social psychology foundations of attitude (Allport, Katz, Rosenberg):General attitude theory that TRA sharpened by insisting on behavioral specificity and by adding the social-norm pathway.

Describe The Model

The Theory of Reasoned Action specifies a hierarchical causal chain: beliefs determine attitudes and subjective norms; attitudes and subjective norms jointly determine behavioral intention; and behavioral intention is the immediate, proximal cause of behavior. The chain is:

Behavioral Beliefs → Attitude  +  Normative Beliefs → Subjective Norm  →  Behavioral Intention  →  Behavior

What does the model measure?

• Behavioral intention: Willingness, plans, and likelihood of performing a specific technology-use behavior.
• Attitude toward the behavior: Overall evaluation of the behavior on semantic-differential scales (good/bad, beneficial/harmful, wise/foolish).
• Subjective norm: Perceived social pressure from important referent others regarding whether the individual should perform the behavior.
• Behavioral beliefs and outcome evaluations: Modal salient beliefs about consequences and their evaluations.
• Normative beliefs and motivation to comply:Perceptions of each referent’s expectations and motivation to conform with each.
• Actual behavior: Observable or self-reported performance of the technology-use behavior.

Note: TRA assumes behavior is under volitional control. When non-volitional barriers exist (resources, skills, opportunities), Ajzen (1991)’s Theory of Planned Behavior extends TRA by adding Perceived Behavioral Control as a third predictor.

Main Strengths

• Parsimonious theoretical structure: Reduces behavioral prediction to two primary variables, attitude and subjective norm.
• Strong empirical support: Decades of research consistently demonstrate strong relationships between attitudes/subjective norms and intentions, and between intentions and behavior, across diverse populations, behaviors, and contexts.
• Explicit causal mechanism: Specifies explicit causal relationships enabling formal tests of mediation.
• Intention as proximal predictor: Explains why attitudes are often weak predictors of behavior by inserting intention as the proximal psychological mechanism.
• Specificity principle: Resolved longstanding problems in attitude-behavior research by insisting that measurement compatibility across action, target, context, and time is required.
• Actionable framework: Directly implies practical interventions on attitudes and norms to move intentions and ultimately behavior.
• Foundation for extensions: Provided the structural template on which TPB, TAM, UTAUT, and numerous other adoption models were built.

Main Weaknesses

• Volitional behavior assumption: Assumes behavior is under conscious volitional control; many technology-adoption behaviors face non-volitional barriers (resources, policy, system compatibility).
• Attitude-behavior gap:Even with TRA’s refinements, substantial attitude-behavior gaps persist.
• Limited attention to implementation barriers: Psychological focus provides less guidance about technical, training, or organizational obstacles.
• Behavioral belief measurement challenges: Identifying and measuring all relevant modal salient beliefs is difficult.
• Normative influences complexity:Real-world conflicting normative influences are simplified in TRA’s framework.
• Unconscious processes neglected: Assumes conscious deliberation produces intentions; habit and automaticity are under-specified.
• Affective dimension limited: Limited attention to emotion, anxiety, or affect independent of cognitive attitude.
• Technology characteristics neglected:Little attention to technology features (usefulness, ease of use, design quality), which prompted TAM’s development.

How does this model differ from older models?

• Introduces intention as proximal mediator: Resolves the weak attitude-behavior correlation by modeling intention between belief and behavior.
• Specificity-matching: Departs from general attitude theories by requiring compatibility across action, target, context, and time.
• Adds the social pathway: Incorporates subjective norms as a co-equal determinant of behavior, capturing social pressure alongside individual evaluation.
• Cognitive turn over behaviorism: Locates behavior determination in cognitive and social psychological variables rather than in reinforcement alone.
• Explicit causal chain: Moves beyond correlational attitude research to a testable sequential model.
• Demographics operate through psychology: Treats demographic characteristics as distal variables whose effects flow through attitudes and norms.

Key Contributions

• Foundational attitude-behavior bridge: Resolved the long-standing attitude-behavior gap by identifying behavioral intention as the proximal cause of behavior.
• Expectancy-value formalization: Crystallized the expectancy-value product as the cognitive basis of attitude and the cognitive basis of subjective norm.
• Principle of Compatibility: Articulated the measurement requirement that attitude, intention, and behavior be specified at matching action, target, context, and time levels.
• Template for technology acceptance theory: Provided the direct structural antecedent for the Technology Acceptance Model (Davis, 1989), the Theory of Planned Behavior (Ajzen, 1991), and the Unified Theory of Acceptance and Use of Technology (Venkatesh et al., 2003).
• Intervention design framework: Translated behavioral prediction into practical guidance for attitude-change campaigns, norm-shaping interventions, and diagnostic assessment of adoption readiness.
• Cross-domain application: Subsequent research has reported TRA applications across health, environmental, consumer, organizational, and technology-adoption behaviors; predictive performance varies by domain and methodology.
• Methodological legacy: Helped codify measurement practices (semantic-differential attitude scales, belief-elicitation procedures, intention operationalization) that are commonly used in adoption research.

Internal Validity

TRA’s internal validity was rigorously tested through numerous empirical studies conducted by Fishbein, Ajzen, and subsequent researchers:

• Laboratory experiments: Controlled experiments explicitly tested the proposed causal relationships. Attitude and subjective norm were experimentally manipulated, and behavioral intentions were measured to verify that the predicted relationships held under controlled conditions.
• Path analytic studies: Structural equation modeling and path analysis confirmed that attitudes and subjective norms predict intentions, which in turn predict behavior, across diverse behaviors.
• Mediational analysis: Research confirmed that behavioral intention functions as a mediator between attitudes/subjective norms and behavior.
• Belief-attitude relationship validation: Studies confirmed that behavioral beliefs weighted by outcome evaluations predict attitudes in the expected manner.
• Normative belief validation: Studies confirmed that normative beliefs weighted by motivation to comply predict subjective norms.
• Temporal precedence: Prospective designs measured attitudes, subjective norms, and intentions at one timepoint, then measured actual behavior at a subsequent timepoint, supporting causal directionality.
• Multiple behavior domains: Internal validity was strengthened by consistent relationships across voting, family planning, health, donation, energy conservation, and technology-use behaviors.
• Cross-population consistency: Consistent relationships were found across ages, educational levels, cultural backgrounds, and socioeconomic statuses.
• Individual-difference robustness: Basic relationships remained robust across personality and value-based individual differences.

External Validity

TRA achieved substantial external validity through diverse research approaches:

• Prospective field studies: Researchers measured attitudes and subjective norms before behavior occurred and tracked actual behavior months later, demonstrating real-world predictive validity.
• Real behavior measurement: Studies measured actual blood donation, actual voting, and actual technology adoption, not only intentions or hypothetical choices.
• Diverse behavioral contexts: External validity was demonstrated across health, environmental, organizational, consumer, social, and technology-adoption domains.
• Cross-cultural research: Studies in different countries and cultural contexts found attitudes and subjective norms predicted intentions across cultures, though relative weights of the two pathways varied.
• Longitudinal studies: Extended studies tracking behavior over months and years found intentions measured early predicted behavior long afterward.
• Meta-analytic evidence: Sheppard, Hartwick, and Warshaw’s (1988)meta-analysis and subsequent reviews provided robust aggregate evidence across the literature.
• Technology adoption applications: Studies found intentions to use information technology predicted actual adoption, proficiency development, and sustained usage in organizational settings.
• Real-world implementation contexts: Employee intentions to use new information systems predicted actual system usage in field studies of organizational deployments.

Relevance to Technology Adoption

TRA is directly relevant to technology adoption because it identifies the psychological and social pathways through which individuals decide to use or not use a technology. Unlike later models that focus on technology features, TRA locates the decision in the individual’s evaluation of the behavior and perceived social pressure.

Barriers to Technology Adoption Identified by TRA

• Unfavorable attitudes toward adoption: Negative evaluations of adoption consequences, beliefs that adoption will increase complexity, reduce autonomy, or require unacceptable effort.
• Negative subjective norms: Perception that respected colleagues, supervisors, or referent groups do not support adoption, even when the individual personally holds favorable attitudes.
• Attitude-norm conflict: Internal tension when attitudes favor adoption but perceived norms oppose it (or vice versa), producing weaker intentions.
• Weak or missing behavioral beliefs: Lack of positive beliefs about how adoption addresses real needs.
• Erroneous behavioral beliefs: False beliefs overestimating difficulty or underestimating benefits.
• Misperceptions of normative beliefs: Assuming colleagues or leaders oppose adoption when support actually exists.
• Non-volitional constraints: Resource, policy, and compatibility barriers that block translation of strong intentions into behavior (a known limitation of TRA, later addressed in TPB and UTAUT).
• Competing intentions: Limited cognitive and motivational resources consumed by other work demands.
• Temporal barriers: Insufficient time for attitude and norm formation before implementation.

Leadership Actions TRA Prescribes to Reduce Barriers

• Educate on consequences and benefits: Address behavioral beliefs with accurate information about productivity, capabilities, and career outcomes.
• Provide evidence, not assertion: Use case studies, data, and peer testimonials to build positive behavioral beliefs.
• Highlight valued outcomes: Connect adoption to outcomes each individual already values.
• Create positive subjective norms: Leaders visibly adopt, peer champions are empowered, and early-adopter success is made visible.
• Correct misperceptions about norms: Communicate broad support, highlight positive colleague experiences, and dispel myths about resistance.
• Demonstrate organizational commitment: Allocate resources to training, support, and infrastructure.
• Use multiple communication channels: Email, meetings, training, peer mentoring, and leadership communication together reinforce attitudes and norms.
• Align with organizational values: Frame adoption as consistent with innovation, efficiency, service, or continuous improvement.
• Engage opinion leaders: Identify and enlist influential colleagues whose endorsement shifts norms.
• Set clear expectations: Communicate that adoption is expected and valued.
• Remove non-volitional barriers: Ensure training, technical support, system compatibility, and implementation time are sufficient for intentions to translate into behavior.
• Monitor intention development:Assess employees’ intentions early to diagnose whether attitude or norm interventions are taking hold.
• Sustain reinforcement: Maintain communication, visibility of leadership support, and recognition of adopters throughout implementation.

Following Models or Theories

TRA served as the direct structural antecedent for:

• Theory of Planned Behavior (Ajzen, 1991): Added Perceived Behavioral Control as a third determinant of intention, addressing TRA’s volitional-control limitation.
• Technology Acceptance Model (Davis, 1989): Adapted TRA’s attitude-intention-behavior chain to technology contexts, substituting Perceived Usefulness and Perceived Ease of Use as the cognitive antecedents of attitude.
• Technology Acceptance Model 2 (Venkatesh & Davis, 2000): Extended TAM with subjective-norm pathways and cognitive instrumental processes.
• Unified Theory of Acceptance and Use of Technology (Venkatesh et al., 2003): Integrated eight prior models, including TRA, TPB, and TAM, into a unified predictive framework.
• Technology Acceptance Model 3 (Venkatesh & Bala, 2008): Further integration of TAM with antecedents of perceived ease of use.
• Reasoned Action Approach (Fishbein & Ajzen, 2010): A later refinement by the original authors that integrates TRA and TPB into a single integrated framework.
• Health behavior models:The Health Belief Model and Integrated Behavioral Model adapted TRA’s structure for public-health adoption contexts.
• Consumer behavior models:Purchase-intention and brand-choice models in marketing literature built on TRA’s intention-behavior chain.

References

1. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211. https://doi.org/10.1016/0749-5978(91)90020-T↩↩↩
2. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-340. https://doi.org/10.2307/249008↩↩
3. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley Publishing Company. ISBN: 978-0-201-02089-2
4. Sheppard, B. H., Hartwick, J., & Warshaw, P. R. (1988). The theory of reasoned action: A meta-analysis of past research with recommendations for modifications and future research. Journal of Consumer Research, 15(3), 325-343.↩
5. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425-478. https://doi.org/10.2307/30036540↩↩

Further Reading

1. Ajzen, I. (1988). Attitudes, personality, and behavior. Dorsey Press.
2. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Prentice-Hall.
3. Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40(4), 471-499.
4. Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35(8), 982-1003.
5. Fishbein, M. (2008). A reasoned action approach to health behavior change. In R. J. DiClemente, R. A. Crosby, & M. C. Kegler (Eds.), Emerging theories in health promotion practice and research (2nd ed., pp. 97-121). Jossey-Bass.
6. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall. ISBN: 978-0-13-815614-5
7. Rogers, E. M. (2003). Diffusion of innovations (5th ed.). Free Press.

Series Navigation

This article is part of a comprehensive bibliography examining foundational and contemporary models of technology adoption. The series progresses through theoretical foundations, early models, and contemporary frameworks:

• Foundational Psychological Theories:Theory of Reasoned Action (Fishbein & Ajzen, 1975) - Current Article; Social Cognitive Theory (Bandura, 1986); Diffusion of Innovations (Rogers, 1962/2003).
• Early Technology Adoption Models:Technology Acceptance Model (Davis, 1989); Theory of Planned Behavior (Ajzen, 1991); Task-Technology Fit (Goodhue & Thompson, 1995).
• Contemporary Integrated Models:Unified Theory of Acceptance and Use of Technology (Venkatesh et al., 2003); Technology Acceptance Model 3 (Venkatesh & Bala, 2008); UTAUT2 (Venkatesh et al., 2012).
• Emerging and Specialized Models:Technology Readiness Index 2.0 (Parasuraman & Colby, 2015); Value-Based Adoption Model (Kim et al., 2007); TRAM (Lin et al., 2007).

← Back to Complete Bibliography