The Theory of Planned Behavior: Predicting Intentions, Missing the Gap

What Is the Theory of Planned Behavior?

The Theory of Planned Behavior (TPB) explains intentions through attitudes toward a behavior, subjective norms, and perceived behavioral control. Intentions and perceived control help predict behavior. Icek Ajzen’s 1991 account extends the Theory of Reasoned Action to behaviors that are not entirely under voluntary control.

The Three Predictors of Intention

Attitudes Toward the Behavior

A person’s favorable or unfavorable evaluation of performing the behavior. In the model, this relates to behavioral beliefs about likely outcomes and how the person evaluates those outcomes. Someone might favor exercise because they expect more energy, while also disliking its time demands.

Subjective Norms

Perceived social pressure to perform or avoid the behavior. This relates to normative beliefs about what relevant people expect and the person’s motivation to comply. For example, perceived encouragement from friends might support an intention to exercise.

Perceived Behavioral Control

The person’s perception of how easy or difficult the behavior would be. Control beliefs concern factors that could help or impede action. Perceived control can differ from actual access to time, skills, money, or opportunities; its value as a proxy depends on its accuracy.

Example: Measuring Attitudes and Beliefs

First specify the behavior. In a hypothetical commuting study, the target could be cycling from home to work twice a week during the next month. Ask about that action consistently, rather than mixing a general liking for cycling with intentions about a particular commute.

A direct attitude measure could ask participants to rate making those trips on scales such as bad–good and unpleasant–pleasant. Belief elicitation is different: ask people to describe expected advantages and disadvantages, then assess how likely they think each outcome is and how they evaluate it. Someone might expect to save money but also arrive tired. The overall attitude rating and these underlying behavioral beliefs answer related, distinct questions.

Likewise, ask whose approval matters and what circumstances would help or prevent the trips. Ajzen’s questionnaire guide recommends defining the behavior and population, eliciting beliefs, and testing the measures. These example prompts are an illustration, not a validated questionnaire.

How Well Does the TPB Predict?

Armitage and Conner’s 2001 meta-analysis synthesized 185 studies and reported that the model accounted for 39% of variation in intentions and 27% in behavior overall. For behavior, the corresponding estimates were 31% with self-report measures and 21% with objective or observed measures. Subjective norms were generally a weaker intention predictor, with measurement and construct coverage identified as concerns.

McEachan and colleagues’ 2011 review identified 237 prospective tests from 206 articles on health behavior. It reported 44.3% of variation in intentions and 19.3% in behavior. Results differed by behavior, follow-up duration, and measurement. Past behavior added predictive information where it could be included.

These figures describe how much variation the model accounted for in the studies. They are not percentages of people successfully changed by an intervention. They do not establish a universal ranking of predictors or show that changing the strongest correlate will cause the largest behavioral improvement.

The Intention–Behavior Gap

A positive intention does not guarantee action. Webb and Sheeran’s 2006 meta-analysis examined 47 experimental tests that changed intentions. The average standardized change was larger for intention (d = 0.66) than for behavior (d = 0.36). This supports distinguishing motivation from subsequent action, while the size of the relationship depends on the intervention and context.

The gap concerns whether intentions translate into behavior. Someone who intends to exercise may still lack transport or childcare when the opportunity arises.

Limitations and Criticisms

Ajzen’s original account explicitly recognizes actual resources and opportunities. It also states that predictions depend on matching measures to the behavior and on intentions and control remaining sufficiently stable before action. The practical problem is therefore more specific: measuring perceived control does not necessarily capture the actual resources or opportunities available.

The model can organize questions about a particular behavior: what outcomes people expect, whose expectations matter, and what makes action difficult. Its predictive evidence does not by itself validate a particular intervention package. A design intended to change these beliefs or conditions needs its own outcome evaluation.

Likewise, additional prediction from past behavior does not by itself identify the mechanism as habit. It may also reflect stable circumstances or other unmeasured factors.

What Explains the Intention–Behavior Gap?

Carrying out a plan can be disrupted by competing priorities, a changed schedule, a missing resource, or a task that is harder than expected. These are possibilities to investigate in the particular case. The label “intention–behavior gap” does not identify which one caused the difficulty.

The practical implication is to examine the behavior itself. If the same outcome can be achieved through an activity that fits the person and setting better, that may be more useful than repeatedly trying to strengthen the original intention.

Applications

In an exercise project, TPB questions can distinguish how people evaluate exercise, what they think others expect, and how feasible it seems. The same kinds of questions can be asked about a screening appointment, condom use, or an environmental action. The answers and actual constraints must be investigated for each behavior.

For technology adoption, avoid treating similarly named constructs as identical. The Technology Acceptance Model and TPB are separate models; usefulness and ease of use should not simply be relabeled as TPB measures.

TPB vs. Other Models

• Theory of Reasoned Action: The predecessor emphasizes attitudes and subjective norms in forming intentions. TPB adds perceived behavioral control.
• COM-B: Organizes an assessment around capability, opportunity, and motivation and connects to the broader Behaviour Change Wheel.
• Health Belief Model: Focuses on beliefs about a health threat and the benefits and barriers of responding.
• Social Cognitive Theory: Considers reciprocal relationships among personal factors, behavior, and environment, including self-efficacy.

These distinctions help select questions.

Frequently Asked Questions

What is the Theory of Planned Behavior?

TPB relates intentions to attitudes, subjective norms, and perceived behavioral control. Intentions and perceived control help predict behavior.

How does TPB differ from the Theory of Reasoned Action?

TPB adds perceived behavioral control to address actions that are not fully under voluntary control.

What is the intention–behavior gap?

Forming an intention does not guarantee acting on it. The gap concerns that failure to translate intention into behavior.

Does TPB predict actual behavior?

It accounts for some variation in behavior in the reviewed studies, with results depending on the behavior and measurement. Prediction is useful, but it does not establish which intervention will improve the outcome.

How does TPB compare with COM-B?

TPB provides a model of beliefs, intentions, perceived control, and behavior. COM-B organizes capability, opportunity, and motivation within a broader design system. Neither description replaces evidence for the specific change a team proposes.

Why it matters

People can sincerely intend to act and still fail to do so. TPB helps ask structured questions about those intentions and perceived control. A useful intervention must also address what happens when the person tries to act.