An education PhD or EdD does not need the right theorist; it needs a theory that does work. UK education theses mostly build on five frameworks — Bourdieu, Vygotsky and sociocultural theory, Bronfenbrenner, cultural-historical activity theory and Bernstein — and the examiners’ test is the same for each: does the framework generate the questions, shape the analysis and explain the findings?
Below, each framework is set out as a doctoral candidate has to use it: what it explains, what it cannot, the canonical source to cite, and the kind of thesis it fits. Then the chapter itself: where the framework sits, what it must contain, and how the viva tests it. Research degree regulations differ by institution; check your own code of practice for where the framework chapter is expected.
What does a theoretical framework do in an education thesis?
Three jobs, and the viva checks all three. It defines the concepts the thesis uses, so that words like learning, capital or participation mean one thing throughout. It generates the research questions, so that the questions are derivable from the theory rather than bolted on. And it structures the analysis, so that findings are read through the framework and the discussion says what the framework could and could not explain. A framework that appears in chapter two and vanishes until the conclusion is the commonest structural failure examiners name, and it is visible from the contents page.
The choice is also disciplinary signalling: each framework carries a tradition with its own journals, methods and examiners. The table sets out the five that UK education theses most often build on.
| Framework | Canonical sources | What it explains well | What it cannot do | Typical doctoral use |
|---|---|---|---|---|
| Bourdieu: habitus, field, capital | Bourdieu and Passeron, Reproduction in Education, Society and Culture (1970, English 1977); Bourdieu, The Forms of Capital (1986) | Why educational inequality persists; how family-acquired cultural capital converts into school success; the rules of a field | Individual agency and change; classroom-level learning processes | Widening participation, access to elite universities, teacher careers, parental choice |
| Vygotsky and sociocultural theory | Vygotsky’s work on the zone of proximal development; scaffolding as developed by Wood, Bruner and Ross | How learning happens through mediated social interaction; the gap between unaided and assisted performance | Structural inequality; institutional politics | Classroom interaction, early years, dialogic pedagogy, teacher–pupil talk |
| Bronfenbrenner: ecological systems | Bronfenbrenner, The Ecology of Human Development (1979); the later bioecological model with Ceci | How a child’s development sits inside nested micro, meso, exo, macro and chrono systems | Mechanisms inside any one system; it maps, it does not explain | Inclusion, SEND, safeguarding, school–family–community studies |
| Cultural-historical activity theory (CHAT) | Engeström, Learning by Expanding (1987), building on Vygotsky and Leontiev | How an activity system — subject, object, tools, rules, community, division of labour — produces contradictions and change | Individual meaning-making; it is heavy for small studies | Curriculum change, professional learning, technology adoption in schools and colleges |
| Bernstein: classification, framing, pedagogic discourse | Bernstein, Class, Codes and Control vols 3 and 4 (1975, 1990); Pedagogy, Symbolic Control and Identity (1996, 2000) | How knowledge is selected, organised and transmitted; power and control in curriculum and pedagogy | Learner experience; affect; it demands fluency in a dense vocabulary | Curriculum policy, knowledge in vocational and higher education, assessment |
When does Bourdieu fit, and when is he a mistake?
Bourdieu fits when the thesis is about who gets what from education and why the pattern persists. Reproduction in Education, Society and Culture, written with Passeron, argues that schooling rewards behaviours — accent, dress, deportment, ease — that privileged children acquire at home, so that the system reproduces inequality while appearing to reward merit. The Forms of Capital (1986) supplies the working concepts: cultural capital as competences and qualifications acquired through early, imperceptible family learning; habitus as the lasting, acquired schemes of perception, thought and action that operate below awareness; field as the arena in which people manoeuvre for resources under rules particular to it.
A doctoral thesis using Bourdieu must operationalise those concepts. The examiner will ask how you recognised habitus in an interview transcript, which capitals you coded for, and what the field was. He is a mistake when the thesis is really about how a child learns to read: Bourdieu explains position, not process, and a classroom-interaction study that borrows his vocabulary will be told in the viva that the theory did no work.
What does a Vygotskian framework commit you to?
To a view of learning as social and mediated. Vygotsky’s zone of proximal development — the space between what a learner can do unsupported and what they cannot do even with support — was introduced in the last years of his life before his death in 1934, and it locates learning in interaction rather than in the individual. Scaffolding, developed by Wood, Bruner and Ross in applying the concept, gives the thesis its analytic unit: the moment at which a more capable other adjusts support.
The commitment is methodological. A Vygotskian thesis needs data on interaction — recorded talk, observed joint activity, artefacts produced together — and an analysis that can show mediation happening. A survey of teacher attitudes cannot carry it. The framework fits early years, dialogic pedagogy, language learning and any study whose question begins with how learning happens between people; it does not fit questions about policy, structure or inequality.
Why do so many EdD theses use Bronfenbrenner, and what is the risk?
Because it maps the practitioner’s world. Bronfenbrenner’s The Ecology of Human Development (1979) places the developing child inside nested systems: the microsystem of family, school and peers; the mesosystem of connections between them; the exosystem of settings that affect the child indirectly, such as a parent’s workplace; the macrosystem of culture and socio-economic context; and the chronosystem of change over time. Later, with Stephen Ceci, he recast it as the bioecological model, giving the developing person a more active role in response to criticism that the original underplayed individual differences.
For an EdD candidate studying inclusion or school–community relations the model organises everything they already know into a diagram, and that is the risk: a diagram is not an explanation. Examiners increasingly ask which version of the model the thesis uses — the 1979 systems or the later bioecological form — and what mechanism inside a system the thesis actually tests. A Bronfenbrenner thesis that stops at mapping will be asked in the viva what the theory predicted, and the candidate needs an answer.
When is activity theory worth its weight?
When the object of study is a system under strain. Cultural-historical activity theory runs from Vygotsky through Leontiev to Engeström, whose Learning by Expanding (1987) set out the activity system that doctoral theses now draw as a triangle: subject, object, mediating tools, rules, community and division of labour. The analytic payload is contradiction. When a new assessment policy (a rule) conflicts with a department’s tools or its division of labour, the tensions surface as disturbances in practice, and expansive learning — which Engeström describes as beginning with individuals questioning accepted practice and expanding into a collective movement — is how the system remakes itself.
CHAT fits doctoral studies of curriculum change, professional learning, technology adoption and partnership working, where the unit of analysis is an institution rather than a learner. It is heavy for small studies, and it demands that you model at least one whole activity system with data on every node, which is a methods commitment to make before ethics approval, not after.
What does Bernstein let you say that the others do not?
How knowledge itself is organised and controlled. Bernstein’s classification and framing, from Class, Codes and Control volume 3 (1975), separate power relations between contexts — how strongly subjects, spaces and roles are insulated from one another — from control within contexts, meaning who decides the selection, sequence and pacing of what is taught. Volume 4 (1990) developed the pedagogic device, the mechanism by which knowledge produced elsewhere is recontextualised into curriculum; Pedagogy, Symbolic Control and Identity (1996, second edition 2000) extended it to identity. Behind them sit the elaborated and restricted codes of his earliest work on language and class.
A Bernsteinian thesis can analyse a curriculum document, an assessment regime or a lesson and say precisely where power and control lie. That precision is bought with vocabulary: examiners from the tradition expect the concepts used exactly. It fits curriculum policy, knowledge in vocational and higher education, and assessment; it fits badly where the question is about how learners feel.
Where does the framework chapter go, and what must it contain?
In most UK education theses it is either the second half of the literature review or a separate chapter three, after the review and before the methodology, and the choice depends on whether the framework emerged from the literature or was chosen before it. Either way the chapter must contain four things: the concepts, defined from the canonical source and not from a textbook summary; the reasons this framework was chosen over its nearest rival, argued rather than asserted; the research questions, shown to follow from the concepts; and the analytic categories that will be used in the findings chapters, with their origin in the framework made explicit. Our guide to building a literature review chapter covers the synthesis that precedes it, and the introduction chapter is where the framework is first named.
Where a discipline has no methodology chapter the framework carries more weight; the parallel case in law is set out in our piece on how a law PhD thesis is structured. Education always has a methods chapter, so the framework chapter should stop at the analytic categories and let methods explain how they were applied.
Can an education thesis combine two frameworks?
Yes, if each does a distinct job and the chapter says which. Bourdieu with Bronfenbrenner is common in widening-participation studies: the ecological model maps the settings, the Bourdieusian concepts explain what happens inside them. Vygotsky with CHAT is a natural pairing because they share a lineage. What examiners reject is eclecticism — five theorists cited, none used in the analysis — and incompatible pairings whose assumptions contradict each other without acknowledgement.
How does the viva test the framework?
With three questions, usually early. Why this framework and not its rival? Show me where the framework shaped a finding. What could the framework not explain? The third is the one candidates fail to prepare, and it is the one that shows doctoral judgement: a candidate who can say that Bourdieu explained the pattern of access but not the experience of the students who defied it, and that a different lens would be needed for that, has demonstrated the original contribution examiners are looking for. For EdD candidates the same questions come with a practice edge — what does the framework let a school leader do differently — and the professional doctorate is examined on that answer.
Holding the framework steady across 80,000 words
The characteristic failure is drift: habitus defined one way in chapter three and used another in chapter six, a mesosystem that quietly becomes an exosystem, framing used as an ordinary word by chapter seven. The framework is a set of definitions that four years of writing must not erode. Draft your education thesis in Tesify: keep the concept definitions, the canonical sources and every chapter in one workspace, so that a concept defined in year one is used identically in the findings written in year three. The theory choice and the argument remain yours; the tool holds the definitions steady. There is a free plan; start by writing the five concept definitions your thesis depends on.
Frequently asked questions
What is the most common theoretical framework in education PhDs?
In UK education theses, Bourdieu’s concepts of habitus, field and capital are the most frequently used for questions about inequality and access, and Vygotskian sociocultural theory for questions about classroom learning. Bronfenbrenner is especially common in EdD theses by practitioners.
Do I need a theoretical framework for an EdD?
Yes. The EdD thesis differs from the PhD in length and scope, not in the requirement for theory; Bristol’s programme, the first taught doctorate outside North America when it began in 1991, and the Institute of Education’s EdD for experienced professionals both examine the use of theory in the thesis.
What is the difference between a theoretical and a conceptual framework?
A theoretical framework is an existing theory, cited from its canonical source, applied to your question. A conceptual framework is the model you build for your own study, which may combine concepts from one or more theories with your own. Education theses usually have both, and the chapter should say which is which.
Which Bourdieu texts should an education thesis cite?
Reproduction in Education, Society and Culture, written with Passeron and published in 1970 with an English edition in 1977, for the argument about schooling and inequality, and The Forms of Capital from 1986 for the definitions of cultural, social and economic capital. Cite the primary texts, not summaries.
Is Bronfenbrenner a theory or a model?
Both, depending on the version. The 1979 ecological systems theory describes nested micro, meso, exo, macro and chrono systems; the later bioecological model developed with Ceci gives the developing person an active role. State which version the thesis uses and what it predicts, not only what it maps.
When should an education thesis use activity theory?
When the unit of analysis is an activity system under strain — a department implementing a new curriculum, a college adopting a technology — and the thesis has data on subject, object, tools, rules, community and division of labour. Engeström’s Learning by Expanding from 1987 is the source to cite.
Can I use Vygotsky without observing classrooms?
Rarely. A Vygotskian framework claims that learning happens through mediated interaction, so the thesis needs data on interaction: recorded talk, observed joint activity or jointly produced artefacts. Attitude surveys and interviews alone cannot show mediation.
How long should the theoretical framework chapter be?
Long enough to define the concepts from source, argue the choice against a rival, derive the research questions and name the analytic categories. In most education theses that is a substantial chapter or half of the literature review; a framework that fits in two pages is usually being cited rather than used.
What do examiners ask about the theoretical framework in the viva?
Why this framework rather than its rival, where in the findings the framework shaped the analysis, and what the framework could not explain. Prepare a specific answer to the third; it is the question that demonstrates doctoral judgement.