theoretical framework

A marketing dissertation at a UK business school normally rests on one of nine theories: the theory of planned behaviour, the technology acceptance model or UTAUT, the stimulus–organism–response model, customer-based brand equity, commitment–trust theory, SERVQUAL, the elaboration likelihood model, service-dominant logic or Hofstede’s cultural dimensions. Pick the one whose variables you can actually measure with the sample you can reach, and the framework chapter writes itself.

This guide is for UK undergraduate and MSc students in marketing, digital marketing, consumer behaviour and international business who have a topic and a supervisor but no theory. It sets out what each framework explains, the data it needs, the criticism a marker expects you to know, and which to choose. It is the marketing companion to our comparison of theoretical frameworks for an education dissertation, and the logic of choosing is the same: a theory is a lens, not a decoration.

What does a theoretical framework do in a marketing dissertation?

It supplies the variables. “What influences students to buy second-hand clothes online” has no variables until a theory names them; the theory of planned behaviour turns it into attitude, subjective norm, perceived behavioural control and intention, each measurable with published items. The framework also supplies the hypotheses, because the theory says which variable predicts which, and the discussion, because your findings either fit the theory or push against it. Markers read the framework chapter to check that the constructs in your questionnaire were derived from somewhere rather than invented; a questionnaire with no theory behind it is the commonest reason a marketing dissertation drops a band on methodology.

Scribbr on the three moves of a theoretical framework; the marketing-specific choices below sit on top of them.

Which theory explains why consumers intend to buy?

The theory of planned behaviour, published by Icek Ajzen in 1991 in Organizational Behavior and Human Decision Processes, is the most used framework in UK marketing dissertations: it is fully specified, its constructs come with validated item wording, and it fits any question of the form “why do people intend to do X”. Attitude toward the behaviour, subjective norm and perceived behavioural control predict intention; intention and control predict behaviour.

Data it needs: a cross-sectional survey, typically 150 to 250 respondents, three to five Likert items per construct. The criticism markers expect you to know is that the model explains intention far better than behaviour, and that it has been formally declared exhausted for health behaviour: Sniehotta, Presseau and Araújo-Soares’ 2014 paper in Health Psychology Review is titled “Time to retire the theory of planned behaviour”. Cite it, then justify the model for a purchase-intention study, usually by adding a fourth construct such as perceived value and testing whether it improves the prediction.

Which theory fits apps, e-commerce and anything digital?

The technology acceptance model from Fred Davis’s 1989 paper in MIS Quarterly reduces adoption to two beliefs, perceived usefulness and perceived ease of use, which predict attitude and intention to use. Its 2003 successor, the unified theory of acceptance and use of technology by Venkatesh, Morris, Davis and Davis in the same journal, adds social influence, facilitating conditions and four moderators.

Use TAM for adoption of a specific technology: a retailer’s app, a chatbot, augmented-reality try-on. Use UTAUT when the sample can test moderation, usually 250 or more. Both are so heavily used that the marker will want to see what you added: a context-specific construct such as perceived privacy risk, or an untested population. The standing criticism is parsimony at the cost of explanation; TAM says usefulness matters, not why a given app is perceived as useful.

A marketing student mapping survey constructs to a theory diagram on a whiteboard in a business school seminar room — The framework supplies the constructs; the questionnaire measures them. Draw the model before you write a single item.

Which theory explains atmospherics, packaging and social media cues?

The stimulus–organism–response model comes from Mehrabian and Russell’s 1974 environmental psychology and was restated for consumer research by Jacob Jacoby in 2002 in the Journal of Consumer Psychology. A stimulus (store music, website layout, an influencer’s post, sustainable packaging) produces an internal state (pleasure, arousal, trust), which produces an approach or avoidance response (dwell time, purchase, sharing).

It is the most flexible framework on this list, which is its strength and its danger. It fits experiments and surveys; an online experiment with two conditions and 60 to 100 participants is a legitimate undergraduate study. The criticism is that S-O-R is a skeleton: it says there is a mediator but not what it is, so name the organism variables from a second, more specific literature. A dissertation that says only “S-O-R” has not chosen a framework yet.

Which theory should a branding dissertation use?

Customer-based brand equity is Kevin Lane Keller’s 1993 model from the Journal of Marketing: brand equity is the differential effect of brand knowledge on consumer response, and brand knowledge is brand awareness plus brand image (the associations held in memory). David Aaker’s 1991 book Managing Brand Equity gives the parallel five-asset model of awareness, associations, perceived quality, loyalty and other proprietary assets. Use Keller for consumer-side questions and Aaker when the study is about a firm’s brand management.

Data it needs: a survey of consumers of one or two named brands measuring awareness, associations, perceived quality and loyalty with scales from the applied literature. The concepts are intuitive, which is why brand equity dissertations score well on framework and poorly on findings: the model is descriptive, so ask something the data can answer, such as whether one association predicts loyalty more than another, rather than simply mapping the brand.

Which theory explains loyalty and relationships?

The commitment–trust theory of relationship marketing, from Morgan and Hunt’s 1994 paper in the Journal of Marketing, holds that commitment and trust mediate between a firm’s behaviour (shared values, communication, opportunistic behaviour) and outcomes such as propensity to leave and cooperation. Use it for loyalty schemes, subscription retention, B2B supplier relationships or the trust problem in influencer marketing.

Data it needs: a survey of existing customers, or a B2B sample reached through a placement, measuring trust and commitment with established scales, typically 120 to 200 respondents. The marker will look for a fresh context rather than a replication. The criticism is that the theory was built on channel relationships between firms and travels imperfectly to consumer contexts where switching is cheap.

Which theory fits a service quality dissertation?

SERVQUAL, from Parasuraman, Zeithaml and Berry’s 1988 paper in the Journal of Retailing, measures the gap between expected and perceived service on five dimensions: tangibles, reliability, responsiveness, assurance and empathy. It comes with a 22-item instrument and fits hospitality, banking, higher education, healthcare and retail.

The criticism must be engaged with: the expectations-minus-perceptions difference score is statistically awkward, and Cronin and Taylor’s 1992 SERVPERF alternative argues that perceptions alone predict satisfaction better. SERVQUAL without SERVPERF signals a shallow literature review. Check the reliability of each dimension in your own sample and report it, using the thresholds in our guide to acceptable Cronbach’s alpha values, because the empathy and tangibles dimensions routinely underperform.

Which theory explains whether advertising persuades?

The elaboration likelihood model by Petty and Cacioppo (1986) proposes two routes to persuasion: a central route, where the audience processes the argument, and a peripheral route, where cues such as source attractiveness, number of likes or celebrity endorsement do the work. Motivation and ability decide which route operates. It is the natural framework for influencer marketing, sponsored-content disclosure, green claims or comparative advertising.

Data it needs: an online experiment manipulating argument quality and a peripheral cue in a 2 × 2 design, 80 to 120 participants, with a manipulation check. Pair it with Ohanian’s 1990 source-credibility scale when the cue is an endorser, and be ready for the question whether social media users ever process centrally at all.

Two versions of a social media advertisement laid side by side on a desk as experimental stimuli with a printed questionnaire — Persuasion frameworks want an experiment: two stimuli, one manipulated cue, a manipulation check and a response measure.

Which theory suits a conceptual or qualitative marketing dissertation?

Service-dominant logic, set out by Vargo and Lusch in 2004 in the Journal of Marketing, reframes all exchange as service, with value co-created by the customer and the firm rather than embedded in a product. It is the framework of choice for MSc dissertations on co-creation, platforms, customer communities and sustainability, and it works with interviews, netnography and case studies where the previous theories demand a survey.

Its weakness is that it explains everything and predicts nothing: no variables to operationalise, no hypotheses to test, so a dissertation using it must be explicitly interpretive and say so in the methodology. The foundational premises have been revised repeatedly since 2004; cite the current set.

Which theory should an international marketing dissertation use?

Hofstede’s cultural dimensions, first published in 1980 and extended since, score national cultures on dimensions including power distance, individualism, masculinity, uncertainty avoidance, long-term orientation and indulgence. Two-country comparisons of advertising appeals or brand perceptions use the scores as the explanatory variable.

Use it only if you can sample in both countries. The criticism is severe and must appear in the chapter: McSweeney’s 2002 critique in Human Relations argues that the dimensions were derived from a single company’s employee survey and cannot represent national cultures. The safe design uses the dimensions as a heuristic and measures cultural values at the individual level with a validated scale rather than assigning respondents their country’s score.

How do you choose between them?

Framework	Best for	Design and sample	Criticism to address
Theory of planned behaviour	Purchase or switching intention	Survey, 150–250, regression	Intention–behaviour gap; calls to retire it
TAM / UTAUT	Adoption of a named technology	Survey, 150–250 (UTAUT 250+), PLS-SEM	Parsimony; over-replication
Stimulus–organism–response	Atmospherics, packaging, social media cues	Experiment, 60–100, or survey	Skeleton; mediator must be named
Customer-based brand equity	Brand perception and loyalty	Survey of brand users, 100–200	Descriptive; needs a predictive question
Commitment–trust theory	Loyalty, retention, B2B relationships	Survey of existing customers, 120–200	Built on inter-firm channels
SERVQUAL	Service quality in a named sector	22-item survey, 100–200	Difference scores; SERVPERF
Elaboration likelihood model	Advertising and influencer persuasion	2 × 2 experiment, 80–120	Central processing online
Service-dominant logic	Co-creation, platforms, communities	Interviews, netnography, case study	No testable propositions
Hofstede’s dimensions	Two-country comparisons	Survey in both countries	National-level scores; McSweeney

Two rules cut the list. If your data will be a questionnaire, the framework needs constructs with published items, which rules out service-dominant logic. If your sample will be under 100, choose an experiment over a survey model, because the power argument in our guide to sample size for an undergraduate dissertation applies with full force to a five-construct regression. The realistic response rate for a student survey is lower than most proposals assume; the figures in our analysis of business dissertation survey response rates should shape the framework decision, not follow it.

Where does the framework go, and how long should it be?

In most UK business schools the framework closes the literature review, 1,000 to 1,500 words of a 10,000-word dissertation, ending with a labelled figure of the model and a numbered list of hypotheses. The methodology then explains how each construct was measured and the discussion returns to each hypothesis. The methodology conventions for the survey designs above, including the ethics and access questions for a firm-based sample, are in our guide to the methodology chapter of a business dissertation. If the framework is still being chosen at proposal stage, the structure in our guide to writing a dissertation proposal shows where the theory statement sits.

Can you combine two frameworks?

Yes, but only in one direction: a general model extended with one construct from a second literature. TPB plus perceived value, TAM plus privacy risk, S-O-R with trust as the organism variable. Combining two full models produces a questionnaire nobody finishes and a regression with more predictors than your sample can support. Name the base theory, name the extension, and justify in one paragraph why the base alone is insufficient.

When the theory is chosen and the chapter is still blank

Tesify can build the theoretical framework section around the theory you have chosen, working from your research question, your constructs and the papers you have read, so the chapter moves from definition to model to hypotheses in the order a marker expects. The choice of theory, the argument and the reading stay yours and are 100% written by you; what you get is structure and momentum on the chapter students leave until last.

Frequently asked questions

What is the most common theoretical framework in marketing dissertations?

The theory of planned behaviour, followed closely by the technology acceptance model. Both are fully specified survey models with published scale items, which is why they dominate undergraduate work and why markers expect a contextual extension rather than a straight replication.

Do I need a theoretical framework for a qualitative marketing dissertation?

Yes, but it works differently: an interpretive study uses a framework such as service-dominant logic as a sensitising lens rather than a source of hypotheses, and the methodology must say so.

Can I use the 4Ps or SWOT as my theoretical framework?

No. The marketing mix and SWOT are planning tools, not theories; they describe decisions rather than explain behaviour. Markers treat a dissertation framed on them as having no framework at all.

How many theories should I discuss?

Review two or three briefly, then commit to one and justify it. A chapter that describes six theories and chooses none is a literature summary, not a framework.

Is SERVQUAL still acceptable in 2026?

Yes, provided you engage with the SERVPERF critique and report the reliability of each dimension in your own data. Using the 22 items without that discussion is the most common weakness in service quality dissertations.

Can I use Hofstede if my sample is only in the UK?

Only as a heuristic. With one country you cannot vary the national scores, so measure cultural values at the individual level with a validated scale and use Hofstede’s dimensions to interpret rather than to test.

Where do the hypotheses come from?

From the framework. Each path in the model becomes one hypothesis, worded so that a specific statistical test can support or fail to support it. If a hypothesis has no path in your model, either the model is incomplete or the hypothesis does not belong.

What is the difference between a theoretical and a conceptual framework?

The theoretical framework is the established theory you adopt; the conceptual framework is your own model of how the constructs relate in your specific study, usually the theory plus your extension, drawn as a figure. UK business schools expect both, with the conceptual model at the end of the literature review.

For most UK undergraduate and master’s education dissertations the right theoretical framework is one of eight, and the choice is decided by your research question, not by which theorist you liked most in year two. The table below sets them side by side on the only criteria that matter at this level: what the theory explains, what data it needs, how much reading it demands, and where a marker will catch you using it badly. A ranked shortlist and a single recommendation follow.

Framework	Key text	Explains	Data it wants	Reading load	Where students get caught
Sociocultural theory (Vygotsky)	Vygotsky (1978) Mind in Society	How learning happens through interaction and more capable others	Classroom observation, talk, peer and adult support	Low	Using “scaffolding” as if Vygotsky coined it (Wood, Bruner and Ross did, in 1976)
Cultural capital, habitus and field (Bourdieu)	Bourdieu (1986) “The forms of capital”	Why class advantage reproduces through schooling	Interviews, life histories, policy and admissions documents	High	Treating capital as a quantity, dropping habitus and field
Self-efficacy (Bandura)	Bandura (1977, Psychological Review)	Why belief in one’s capability predicts effort and persistence	Validated scales, pre/post measures	Low	Measuring confidence with a home-made scale
Bioecological systems (Bronfenbrenner)	Bronfenbrenner (1979); Bronfenbrenner and Morris (2006)	How a child’s development sits inside nested contexts	Multi-source data: child, family, school, policy	Medium	Citing the 1979 concentric-circles model and calling it the current theory
Self-determination theory (Deci and Ryan)	Ryan and Deci (2000, American Psychologist)	Motivation through autonomy, competence and relatedness	Questionnaires, interviews about choice and pressure	Low	Reducing it to “intrinsic good, extrinsic bad”
Communities of practice (Lave and Wenger)	Lave and Wenger (1991) Situated Learning	How newcomers learn by participating in a practice	Placement, mentoring and induction data	Medium	Using it for a classroom of pupils rather than a professional community
Growth mindset (Dweck)	Dweck (2006) Mindset; Sisk et al. (2018)	Whether beliefs about intelligence affect achievement	Mindset scales, attainment	Low	Ignoring the 2018 meta-analyses that found weak effects
Purposes of education (Biesta)	Biesta (2009)	What schooling is for: qualification, socialisation, subjectification	Policy documents, curricula, teacher interviews	Medium	Using it to describe rather than to evaluate

What a theoretical framework is being asked to do

Sheffield Hallam University’s research-project guidance describes the theoretical framework as your “toolbox”: “a set of theories, concepts, ideas and hypotheses that you will use to build a solution to the research problem or gap you have identified.” If a theory appears in your framework and never reappears in your findings or discussion, it was decoration, and markers at every UK education department will say so in the feedback.

The practical test before you commit to any row of the table: write your research question, then write one sentence beginning “This framework lets me see…”. If you cannot finish the sentence with something specific to your data, move down the list. The mechanics of turning a framework into a defensible design are set out in our guide to writing the methodology chapter of an education dissertation; this article is about which lens to pick.

Degree Doctor on the difference between a theoretical and a conceptual framework, with worked examples.

The ranked shortlist

1. Sociocultural theory: the best default for a classroom study

If your data are classroom talk, peer collaboration, adult support or the way a task is set up, Vygotsky’s account of learning as first social and then internal, with the zone of proximal development as the space between what a learner can do alone and with help, will carry a whole undergraduate project.

Who it suits: primary and early-years projects, studies of group work, teaching-assistant support, and any dissertation with observation at its centre.

Where it falls short: it says little about why some learners get more help than others, so a project about inequality will find it thin. And be precise with the vocabulary: “scaffolding” was introduced by Wood, Bruner and Ross in 1976, building on Vygotsky, and attributing it to Vygotsky himself is one of the most common errors in education reference lists.

2. Bourdieu: the strongest lens for inequality, and the hardest to use well

Cultural capital, habitus and field are the framework UK education research reaches for on class, access and attainment gaps, and Diane Reay’s work on habitus in British schooling shows the concepts applied to exactly the settings undergraduates study. Bourdieu explains why schooling rewards dispositions some families already have, which makes it powerful for interview and life-history data and for reading admissions or curriculum policy.

Who it suits: sociology-of-education projects, widening participation, parental engagement, first-generation students.

Where it falls short: reading load. The three concepts only work together, and a dissertation that uses “cultural capital” as a stand-alone tally of museum visits and books at home, with no habitus and no field, has borrowed a word rather than a theory.

3. Self-efficacy: the cleanest quantitative framework available

Bandura’s 1977 account of self-efficacy, the belief in one’s capability to organise and execute the actions a task requires, has produced validated instruments for almost every education population, including the Teachers’ Sense of Efficacy Scale developed by Tschannen-Moran and Woolfolk Hoy in 2001. If your design is a questionnaire, a pre/post intervention or a comparison between groups, this is the framework whose measurement problems have already been solved for you.

Who it suits: trainee-teacher and NQT studies, pupil confidence in a subject, intervention evaluations.

Where it falls short: it is a theory about belief, not about outcomes, and students routinely over-claim that raised efficacy means raised attainment. It also tempts home-made scales. If you write your own efficacy items instead of using a published instrument, you inherit the validity problem the framework was supposed to remove; our comparison of which validated scales an undergraduate may actually use covers the licensing side.

A student comparing two theory textbooks side by side with a research question written on a notepad between them — The test is the sentence “This framework lets me see…”. If it cannot be finished with something specific to your data, choose another row.

4. Bronfenbrenner: excellent for context, dangerous if you cite the wrong version

For a project on a child’s experience across home, school and policy, the bioecological model gives you a vocabulary for nested influence that nothing else matches. The trap is which version you cite. The 1979 book gave the familiar concentric circles; by the 2006 chapter with Morris the theory had become the process–person–context–time model, with proximal processes at its centre. Tudge and colleagues reviewed 25 papers published since 2001 that claimed to use the theory and found that all but four relied on outmoded versions, producing what they called conceptual confusion. A marker who knows that paper will check your reference list.

Who it suits: early childhood, SEND, transitions, studies with data from more than one setting.

Where it falls short: a single-site interview study cannot populate the outer systems with evidence, and a diagram of circles with your school in the middle is description, not analysis.

5. Self-determination theory: motivation with a proper evidence base

Ryan and Deci’s theory of motivation as supported or thwarted by autonomy, competence and relatedness gives a project on engagement, choice or classroom climate three concepts that map directly onto interview questions and onto published scales.

Who it suits: pupil engagement, homework, reward systems, post-16 course choice.

Where it falls short: the popular reduction to “intrinsic good, extrinsic bad” misreads the theory, which describes a continuum of regulation. If your discussion chapter uses only two categories, you have not used the framework.

6. Communities of practice: for placement, mentoring and induction

Lave and Wenger’s idea that newcomers learn through legitimate peripheral participation in a community of practitioners fits studies of trainee teachers on placement, mentoring, staffroom culture and early-career induction almost perfectly.

Who it suits: initial teacher education, teaching assistants, professional learning.

Where it falls short: it was built to explain apprenticeship-style learning among adults, and stretching it over a class of eight-year-olds usually fails. Keep it for professional communities.

7. Growth mindset: usable, but only with the meta-analyses in the framework

Dweck’s distinction between fixed and growth beliefs about intelligence is the framework most undergraduates arrive with, because most schools have run some version of it. It can still anchor a good dissertation, provided the framework itself reports the evidence honestly. Sisk and colleagues published two meta-analyses in Psychological Science in 2018, one of 273 studies with more than 365,000 participants on the mindset–achievement relationship and one of 43 intervention studies with more than 57,000 participants, and found weak effects in both. A framework section that presents growth mindset as settled will be marked down; one that sets Dweck against Sisk and asks what a small study can add is a framework doing its job.

Who it suits: school-based evaluations of an existing mindset initiative, studies of pupil beliefs.

Where it falls short: a project designed to show that mindset “works” has an answer before it has data.

8. Biesta: the framework for policy and curriculum dissertations

Gert Biesta’s account of the three functions of education, qualification, socialisation and subjectification, set out in his 2009 paper on good education in an age of measurement, gives a documentary dissertation a way to evaluate rather than summarise.

Who it suits: policy analysis, curriculum studies, philosophy-of-education projects with no participants.

Where it falls short: it is an evaluative framework, so a purely descriptive project will not use it.

The recommendation

Choose sociocultural theory if your data are classroom interaction, self-efficacy if your data are numbers, Bourdieu if your question is about inequality and you have the reading time, and Biesta if you are analysing documents. Reach for Bronfenbrenner, self-determination theory, communities of practice or growth mindset only when the question demands them, and in every case cite the primary text and the current version of the theory. If two rows seem to fit, the research question is still too broad: narrow it first with the four moves in our guide to narrowing a dissertation topic, and the framework will usually choose itself.

A hand-drawn diagram connecting a research question to a single named theory and to the data it needs — One question, one lens, one kind of data. A framework section that names three theories without connecting them to the data is a reading list.

How to write the framework section once you have chosen

State the theory in your own words in one paragraph, citing the primary source, and name the two or three concepts you will actually use.
Show it has been used on a question like yours with two or three UK studies, so the marker sees that the choice is conventional in the field rather than personal.
Say what the framework lets you see and what it hides. Every theory has blind spots; naming yours is what distinguishes a framework from a summary.
Connect each concept to a data source or an interview question. This is the sentence that turns theory into method.
Return to it in the discussion. The findings chapter reports; the discussion reads the findings through the lens. If the theory does not reappear, delete it from the framework.

The section normally runs to 800–1,500 words on a 10,000-word dissertation and sits either at the end of the literature review or as a short chapter of its own; your handbook decides. It is also the section with the highest similarity risk, because theory is paraphrased from the same handful of textbooks by every student in the cohort. Write it from the primary source rather than from a summary, and if you want to check the section before your university runs Turnitin, Tesify’s plagiarism checker shows matched passages sentence by sentence so you can rewrite them with the citation in place.

Where the framework goes in the dissertation

Three placements are common in UK education departments and none is wrong: a final section of the literature review, a stand-alone chapter between the review and the methodology, or a short section opening the methodology. What matters is that the theory is introduced before the design decisions that depend on it, and that it matches what you promised in your dissertation proposal.

When the lens is chosen and the blank framework section is what stands between you and the methodology, Tesify can draft the framework section from your own research question, your chosen theory and your reading list, so the argument stays yours and the section gets written this week rather than the week before submission. The reasoning is 100% written by you; what you get is a structured start.

Frequently asked questions

Do I need a theoretical framework for an undergraduate education dissertation?

Almost always, and most UK education handbooks name it explicitly. Even a small practitioner-enquiry project is expected to say which account of learning or schooling it is working with, because that account shapes what you looked for and how you read what you found.

Can I use more than one theory?

You can, but two is the practical limit at this level and you must explain how they fit together. Three named theories in a 10,000-word dissertation is nearly always a reading list rather than a framework.

What is the difference between a theoretical and a conceptual framework?

A theoretical framework uses an existing named theory as the lens. A conceptual framework is the map of concepts and relationships you built for your own study, which may draw on one or more theories. Many education departments use the terms loosely, so check which your handbook means.

Is Bloom’s taxonomy a theoretical framework?

Not in the sense markers mean. It is a classification of learning objectives, useful for designing tasks or analysing questions, but it does not explain why learning happens. Pair it with a theory if you want to use it.

Which framework is easiest to apply?

Sociocultural theory for qualitative classroom data and self-efficacy for quantitative data. Both are short to read, have a clear primary source and come with well-established vocabulary and instruments.

Should I cite the original theorist or a textbook?

The original, always, with a textbook or a recent UK study alongside it to show current use.

Can I still use growth mindset after the 2018 meta-analyses?

Yes, if the framework section reports them. Sisk and colleagues found weak effects across 273 and 43 studies respectively, and a dissertation that sets that evidence against Dweck’s claims and asks what a small local study can add is a stronger project than one that ignores it.

Where does the theoretical framework go in an education dissertation?

At the end of the literature review, as a short chapter of its own, or opening the methodology. All three are accepted in UK education departments; the only rule is that it appears before the design decisions that depend on it.

Tag: theoretical framework