Your supervisor has said “use the theory of planned behaviour” and you have three weeks to turn a theory you half remember into a framework, a questionnaire and testable hypotheses. This guide walks through the constructs, how to turn them into items, which hypotheses to write and where students lose marks. If you would rather build it with a structure beside you, start your dissertation plan with Tesify, where every word stays 100% yours.
What is the theory of planned behaviour?
The theory of planned behaviour (TPB) explains and predicts a specific behaviour from the person’s intention to perform it. Icek Ajzen developed it in 1985 as an extension of the theory of reasoned action that Fishbein and Ajzen proposed in 1980. The key addition was perceived behavioural control. The standard source to cite is Ajzen (1991), ‘The theory of planned behavior’, Organizational Behavior and Human Decision Processes, 50(2), pp. 179–211.
Because it predicts a named behaviour, it suits dissertations on consumer choices (buying a product, switching supplier), health behaviours (exercise, screening, healthy eating), study behaviours and workplace behaviours. If your topic is only about attitudes or feelings with no behaviour attached, another framework is probably a better fit; our comparison of theoretical frameworks for a marketing dissertation will help you choose.
What are the constructs, and how are they defined?
| Construct | Meaning | What it predicts |
|---|---|---|
| Attitude toward the behaviour | The person’s evaluation of performing the behaviour, shaped by beliefs about outcomes and how they value them | Intention |
| Subjective norm | The perceived judgement of significant others about the behaviour | Intention |
| Perceived behavioural control | The perceived ease or difficulty of performing the behaviour | Intention and, directly, behaviour |
| Behavioural intention | Readiness to perform the behaviour, treated as its immediate antecedent | Behaviour |
The distinguishing feature is perceived behavioural control, which is expected to influence behaviour both directly and through intention. In effect, a favourable intention produces the behaviour only when the person feels able to carry it out.
How do you build the framework for your dissertation?
- Name one specific behaviour. “Buying sustainable clothing” is too broad. “Buying at least one item of second-hand clothing from an online marketplace in the next month” names the action, the target, the context and the time.
- Define each construct for that behaviour. Attitude becomes “my evaluation of buying second-hand clothing online”, and so on.
- Draw the model. Three predictors point to intention, intention points to behaviour, and perceived behavioural control also points to behaviour. Our guide to drawing a conceptual framework shows how to label the arrows with hypotheses.
- Decide what you will extend. Many dissertations add one or two variables, such as past behaviour or perceived risk. Justify each addition from the literature rather than adding variables for the sake of it.
Which hypotheses should you write?
Write one hypothesis per arrow, in a form you can test. For the second-hand clothing example (illustrative):
- H1: A more positive attitude toward buying second-hand clothing online is associated with stronger purchase intention.
- H2: A stronger subjective norm is associated with stronger purchase intention.
- H3: Higher perceived behavioural control is associated with stronger purchase intention.
- H4: Purchase intention is associated with reported purchasing behaviour at follow-up.
- H5: Perceived behavioural control is associated with reported purchasing behaviour independently of intention.
If you have only one time point, drop H4 and H5 and say clearly that you are predicting intention, not behaviour. That limitation is common, and markers accept it when you name it.
How do you turn the constructs into questionnaire items?
TPB constructs are usually measured with several items each on a seven-point scale, worded to match the behaviour exactly. The consistency of the wording matters: every item should name the same action, target, context and time. Ajzen’s own website provides guidance on constructing a TPB questionnaire, and you should read it before you write items. Typical item patterns, illustrated for the clothing example:
- Attitude: “For me, buying second-hand clothing online in the next month would be: bad – good / unpleasant – pleasant / unwise – wise.”
- Subjective norm: “Most people who are important to me think I should buy second-hand clothing online in the next month.”
- Perceived behavioural control: “I am confident that I could buy second-hand clothing online in the next month if I wanted to.”
- Intention: “I intend to buy second-hand clothing online in the next month.”
Pilot the questionnaire with a handful of people from your target group, then check the internal consistency of each construct in your own data. Our guide to acceptable Cronbach’s alpha values explains how to report it, and our marketing operationalisation table shows how to lay out constructs, items and sources.
How do you analyse the data?
Most undergraduate and master’s projects use correlation and multiple regression: intention regressed on attitude, subjective norm and perceived behavioural control, with the standardised coefficients showing which predictor matters most. If you have follow-up behaviour, test it as a second step. Work out your sample size before you recruit; our guide to sample size for an undergraduate dissertation explains the reasoning, and you should report how you checked the assumptions of your test.
How do you pilot the questionnaire before the main survey?
Ask five to ten people from your target group to complete the questionnaire while you watch for confusion. Note any item they reread, any word they ask about and any question they skip. Then ask each person two things: what did you think the behaviour was, and which item felt hardest to answer? If their description of the behaviour differs from yours, your behaviour statement is too vague and your items will not be consistent. Revise the wording, record what you changed and why, and report the pilot in your methodology. A short pilot protects you from discovering a flawed item after two hundred people have answered it.
When should you extend the model?
The basic model often explains only part of the variation in intention, so many dissertations add a variable. Common extensions include past behaviour (people who did something before tend to do it again), perceived risk or trust in online contexts, and moral or personal norms for ethical consumption. Add a variable only if you can answer three questions: what does the literature say about its link with intention, how will you measure it with a validated or well-justified scale, and which hypothesis will you add for it? An extension that you cannot defend in your literature review makes the framework look inflated rather than rigorous.
How should you report the results? (illustrative layout)
A clear layout helps the marker see each hypothesis and its verdict at a glance. Report descriptive statistics and reliability first, then the correlation matrix, then the regression. The table below is a template with invented labels and no real numbers.
| Predictor of intention | What to report | Hypothesis |
|---|---|---|
| Attitude | Standardised coefficient, confidence interval and significance | H1 |
| Subjective norm | Standardised coefficient, confidence interval and significance | H2 |
| Perceived behavioural control | Standardised coefficient, confidence interval and significance | H3 |
| Model summary | Variance explained and the overall test of the model | All |
Under the table, write one sentence per hypothesis that says whether it was supported, quote the coefficient, and avoid causal wording. In your discussion, rank the predictors, say which surprised you and link each finding back to the literature you reviewed.
Does it work outside consumer research? (illustrative)
Yes. A psychology or health student might choose the behaviour “attending a health check offered by my university in the next term”. The same four constructs apply: attitude toward attending, the perceived view of friends and family, the perceived ease of fitting it into the timetable, and the intention to go. The hypotheses, items and regression would follow the same pattern, but the discussion would focus on practical barriers, such as timing and awareness, rather than on brand choice. The choice of behaviour is yours, but it must be specific, observable and within the control of your participants.
What are the limitations markers expect you to discuss?
Critics note that the theory neglects emotions and individual needs that affect behaviour regardless of attitudes, that most TPB research is correlational, and that the causal direction may run both ways. Add the intention-behaviour gap: people often do not do what they intend. A strong discussion names two or three of these, says how they apply to your design and suggests how future work could address them, for example with a longitudinal design or an experiment.
Before you commit to the theory, run a quick feasibility check. Can you name the behaviour in one sentence? Can you reach at least the number of respondents your analysis needs? Can you finish data collection before your deadline, including any follow-up? If any answer is no, simplify the design now, for example by predicting intention only, rather than discovering the problem in week eight.
What mistakes cost students marks?
- A vague behaviour. Without a named action, target, context and time, your items cannot be consistent.
- Mixing attitude toward the product with attitude toward the behaviour. Measure the evaluation of performing the behaviour.
- Adding variables without justification. Every extension needs a literature-based reason.
- Claiming to predict behaviour from one survey. If you only measured intention, say so.
- Treating the theory as decoration. Return to it in the discussion and say what each result means for each construct.
What would a complete plan look like? (illustrative)
A compact plan for the clothing example: a behaviour statement; a framework with five hypotheses; a questionnaire with three to four items per construct, a pilot of about ten respondents, and a short demographic section; a sample size justified from your analysis; ethics approval and an information sheet; regression analysis with assumption checks; and a discussion that links each coefficient to a construct and a practical implication for a second-hand marketplace. Each part maps to a chapter, which is exactly how a marker will read it.
Tesify helps you plan the chapters around this structure, so the behaviour statement, framework, items and analysis stay consistent. Over 9,000 students have used it across more than 15,000 chapters, and the writing stays 100% yours. Build your theory of planned behaviour dissertation plan with Tesify.
Objection-handling FAQ
Is the theory of planned behaviour too old to use in 2026?
No. It remains widely used in consumer and health research. Acknowledge its criticisms and consider whether an extended model suits your question better.
Does the theory of planned behaviour work for a purely qualitative dissertation?
It can structure interview questions around attitudes, norms and control, but it is mostly used in quantitative designs. Check what your supervisor expects.
How many items should each construct have?
Several items per construct on a seven-point scale is typical, but follow the guidance you cite and report reliability from your own data.
Can I measure behaviour as well as intention?
Yes, with a follow-up questionnaire or an objective record. With one time point, state that you predict intention only.
Can I use ready-made TPB items from a published study?
You can adapt wording from published studies if you cite them and check permissions, but items must name your own behaviour, target, context and time.
Does using Tesify count as academic misconduct?
Tesify supports planning and structure, and the writing stays 100% yours. Check your university’s policy on AI tools and disclose your use as it requires.
What about my data and privacy?
Do not upload participant data to any tool unless your ethics approval allows it. Check the privacy information on the Tesify site before sharing anything personal.
What does it cost to get started?
Check the Tesify app for current terms before you decide. This guide stands on its own whether or not you use the product.
Which software do I need for the regression?
Any package you can justify and report, such as SPSS, R or jamovi. Our comparison of those three explains how to choose.