Blog

· By Thesisroom

Updated

Conceptual Framework: What It Is and How to Build One for Your Thesis

What goes in a conceptual framework, eight steps to build one for your thesis, how to draw it as boxes and arrows, an invented example in two versions, and a worksheet.

This guide has a printable version: open the printable conceptual framework worksheet

A hand holds a pen over a grid notebook page filled with sketched boxes, lines and small arrows
Photo by Davide Baraldi on Unsplash

A conceptual framework is the map of the key concepts in your study and the relationships you expect among them, often drawn as a diagram and explained in a few paragraphs. A dissertation guide hosted by Valdosta State University puts it this way: it “includes your study variables and depicts the established or predicted relationship(s) among these variables.” This guide covers that research meaning of the term, the one your committee or examiners will ask about.

What is a conceptual framework in a thesis?

It is your own account of which ideas matter in your study and how they connect. In a methods article in the International Journal of Qualitative Methods, Jabareen (2009) defines it as “a network, or ‘a plane,’ of interlinked concepts that together provide a comprehensive understanding of a phenomenon or phenomena.”

The word “your” matters. In their introduction for new biology education researchers, Luft and colleagues (2022) stress that conceptual frameworks “are constructed by the researchers conducting the study.” You draw on the literature, but the arrangement of concepts is yours, and you have to justify it.

A framework does three jobs in a thesis:

  • It defines your terms: each concept gets one meaning that holds from the first chapter to the last.
  • It shows the links: the reader sees which concepts you expect to connect, and how.
  • It steers your design: Luft and colleagues write that these relationships “should guide the researcher’s methods of approaching the study” and inform what data you collect and how you analyze them.

What does a conceptual framework contain?

A conceptual framework contains concepts (or variables), the relationships between them, the source behind each part, and a diagram with a written explanation. National University’s library guide lists the raw material as “Variables, concepts, theories, and/or parts of other existing frameworks.”

The guides we read group the parts differently. These are the parts examiners can test, whatever your department calls them:

  • Concepts or variables: the ideas your research question names or implies. In a quantitative study they are often measurable variables.
  • Relationships: the expected links between concepts, such as one raising another, or one changing the strength of a link.
  • The basis for each part: a citation for every concept and link. National University’s guide says that when you create your framework, “you’ll need to cite each facet.”
  • Boundaries: what you chose to leave out. Luft and colleagues note that frameworks “help readers understand why certain components of a system were not explored in the study.”
  • The diagram and its narrative: National University’s guide says the framework can be “a graphic or a narrative,” but “should always be explained and cited.”

How is a conceptual framework different from a theoretical framework?

The short answer: a theoretical framework is often an established theory you borrow, and a conceptual framework is the structure you build for your own study. Luft and colleagues put the contrast plainly: conceptual frameworks are built by the researchers, which “is different from theoretical frameworks that are often taken from established literature.”

National University’s guide warns that the two terms “are often and erroneously used interchangeably.” Use the word your supervisor and handbook use, and define it once. For the theory side, with worked cases, see our guide on what counts as a theoretical framework, with examples.

How do you build a conceptual framework? Eight steps

Build it from your research question outward, and test it against your methods before you call it finished. The steps below follow the advice in the university guides and articles linked above. Your own handbook may ask for a different order.

  1. Write your problem and research question in one sentence each. The framework exists to answer that question, so every concept must trace back to it. If the problem is still vague, start with how to find and test a research gap.
  2. List every concept the question names or implies. National University’s guide advises you to “pull out variables, concepts, theories, and existing frameworks explained in the relevant literature.” Write them as a plain list first, with no diagram.
  3. Define each concept in one sentence, with a citation. Use the definition you will measure or explore, not a dictionary meaning. If two sources define a concept differently, choose one and say why.
  4. Decide how each pair of concepts relates, and write down the basis. Westcliff University’s Doctoral Writing Center puts it bluntly: “do not make connections out of thin air but have a rationale behind the type of relationship you expect the variables to have.”
  5. Cut what your study will not examine. Keep the concepts your data can reach. Note each concept you cut and one line on why, because an examiner may ask.
  6. Draw the diagram. Use boxes for concepts and lines for relationships, with a legend. The next section shows how, in text you can copy.
  7. Write the narrative that walks through the diagram. Explain each concept, then each link, with its citation, in the same order as the diagram. Westcliff’s guide notes that frameworks are often shown as “a visual graphic with a short textual explanation.”
  8. Test it against your research questions, methods and analysis. Every concept should have a place where you collect data on it, and every link should match a question you answer. Once the thesis has several chapters, Thesisroom’s Argument Map lists each defined term and research question as a quoted passage and shows where a definition drifts between chapters.

If a concept passes step 8 but never appears in your findings, the framework has come loose from the analysis. Our guide on when your theoretical framework does not connect to your analysis covers the repair moves, and they apply to conceptual frameworks too.

A more formal route for qualitative, multidisciplinary topics

Jabareen (2009) sets out a procedure he calls conceptual framework analysis, based on grounded theory. It is meant for phenomena that draw on several disciplines. His eight phases are:

  1. Mapping the selected data sources
  2. Extensive reading and categorizing of the selected data
  3. Identifying and naming concepts
  4. Deconstructing and categorizing the concepts
  5. Integrating concepts
  6. Synthesis, resynthesis, and making it all make sense
  7. Validating the conceptual framework
  8. Rethinking the conceptual framework

Two phases are useful even if you do not follow the full method. In phase 4 he builds a table with four columns: the concept’s name, a description, its role, and its references. In phase 7 he asks whether the framework makes sense “not only to the researcher but also to other scholars and practitioners.”

How do you draw a conceptual framework diagram?

Draw each concept as a box and each relationship as a line, then add a legend that says what your lines mean. No guide we read sets one standard notation, so the legend is what makes your diagram readable.

A workable set of conventions, which you define in your own legend:

  • Box: one concept or variable, named exactly as in your definitions.
  • Arrow: an expected direction of influence, from one concept to another. In a quantitative study, an arrow is a claim you will test.
  • Line with no arrowhead: an expected association with no direction claimed.
  • Arrow pointing at another arrow: a moderator, a concept that changes the strength of a link. Westcliff’s guide defines a moderator as a variable that “can impact the effect that the independent variable has on the dependent variable.”
  • Box in the middle of a chain: a mediator, a concept that explains how the link works. Westcliff’s guide says it helps “understand the connection between the independent and dependent variable.”
  • A label on each arrow: “raises”, “lowers” or “changes the strength of”, so the reader does not have to guess.

Be careful with arrows in qualitative work. National University’s guide, written for its own doctoral students, advises that a qualitative framework should not include arrow lines, “which could imply causal or correlational linkages.” Rules differ between universities, so check your handbook.

You do not need design software to draft it. A relationship table holds the same information as the picture, and you can paste it into your proposal while you refine the drawing:

FromLinkToBasis in the literatureResearch question it serves
Concept AraisesConcept B[the study that links A to B]RQ1
Concept Cchanges the strength ofthe A to B link[the study that shows this]RQ2

What does a conceptual framework look like? A worked example

The example below is invented to show the method. It is not a real study, and the bracketed placeholders mark where a real thesis would cite its sources.

A Master’s student in public health asks: “How do training, workload and supervisor support relate to community health workers’ use of a mobile reporting app in two rural districts?” After steps 1 to 5, six concepts survive:

  • Training received: the app training a worker has completed.
  • Confidence with the app: how sure the worker feels using it, from a short survey scale.
  • Perceived workload: how stretched the worker feels in a typical week.
  • Supervisor support: how often a supervisor checks in about reporting.
  • Network coverage: whether the worker’s area has a reliable mobile signal.
  • App use: the outcome, measured as reports submitted on time each month.

Version 1: quantitative, with arrows

The student expects training to raise confidence, and confidence to raise app use. Confidence is therefore a mediator. Network coverage may weaken or strengthen the step from confidence to use, so it is a moderator.

FromLinkToBasis in the literatureResearch question it serves
Training receivedraisesConfidence with the app[cite a study linking training to confidence]RQ1
Confidence with the appraisesApp use[cite a study linking confidence to use]RQ1
Perceived workloadlowersApp use[cite a study on workload and reporting]RQ2
Supervisor supportraisesApp use[cite a study on supervision and reporting]RQ2
Network coveragechanges the strength ofthe confidence to app use link[cite a study on connectivity and app use]RQ3

The same diagram as a list you can sketch from:

  • Training received → Confidence with the app → App use
  • Perceived workload → App use (lowers)
  • Supervisor support → App use (raises)
  • Network coverage → the arrow from Confidence with the app to App use (changes its strength)

Each row is a hypothesis the student will test. A row with no citation in the basis column is a warning sign: either find the source or drop the arrow.

Version 2: qualitative, with no arrows

Now suppose the student interviews a dozen workers in each district and asks: “How do community health workers experience reporting through a mobile app?” The same concepts appear, but they are grouped as areas to explore, not links to test.

Concept groupConcepts inside itWhat the interviews ask aboutSource to cite
PreparationTraining received, confidence with the appHow workers learned the app and how sure they feel[cite]
Working conditionsPerceived workload, network coverageWhen and where reporting gets hard[cite]
RelationshipsSupervisor support, help from other workersWho they turn to when reporting goes wrong[cite]
At the centerThe experience of reporting through the appWhat reporting means in their working day[cite]

In the drawing, the four groups sit as boxes around the central one, joined by plain lines or simply placed side by side. The student expects the groups to change during analysis and says so in the narrative.

How does a conceptual framework work in qualitative and quantitative research?

In the guides we read, a quantitative framework maps variables and the relationships you will test. A qualitative one maps concepts you will explore, and it may change as you analyze. The sources disagree on the terms, so state your usage.

  • Quantitative: the Valdosta guide describes the framework as including “your study variables” and depicting their “established or predicted relationship(s).” In that kind of study, each arrow can become a hypothesis.
  • Qualitative: Jabareen (2009) argues a framework should be built “not on variable or factors but on concepts alone.” He writes that conceptual frameworks “do not enable us to predict an outcome” and offer understanding instead.
  • The naming dispute: Jabareen suggests the term “model” when a framework uses variables or factors. Other guides, including Valdosta’s and Westcliff’s, use “conceptual framework” for variable maps too. Use the word your department uses, and define it once.
  • Revision: describing one published framework, Luft and colleagues note that it “may be modified as additional data are collected and analyzed by other researchers,” and call this “not uncommon.” In a qualitative thesis, show any change to your framework and the reason for it.

Where does the conceptual framework go in a thesis?

Your handbook decides, and rules differ between universities. As one example, the Valdosta guide places a brief summary in Chapter 1 and the full framework in Chapter 2, noting it “May appear in chapter 3.”

The same guide says it offers “guidelines only” and that you “must consult with your dissertation chair and committee members” on the elements and their order. Our PhD thesis structure, chapter by chapter shows where each chapter sits and what examiners check in it.

Wherever it sits, the framework should come after the literature that justifies it and before the methods it shapes. National University’s guide sums up the order as “problem statement drives framework drives RQs drives methodology.”

What are the common weaknesses in a conceptual framework?

These are the weaknesses to look for in your own framework, ideally before your supervisor reads it:

  • Boxes with no definitions: a concept named in the diagram but never defined in the text.
  • Arrows with no basis: a relationship you expect but cannot cite. Find the source or remove the arrow.
  • Too many boxes: concepts your data cannot reach. Every box needs a place in your methods.
  • Causal arrows in a qualitative study: arrows that promise a test your design does not run.
  • A borrowed diagram presented as yours: a published model copied with no adaptation. If you adopt an existing framework, which the Valdosta guide allows, cite it and say what you changed.
  • A framework that disappears: concepts defined in Chapter 2 that never return in the findings or discussion.
  • A framework that blinds you: the Valdosta guide compares a framework to “a telescope or microscope” that “both enhances what you can see and also restricts your breadth of vision.” Say in your limitations what yours may have kept out of view.

You may also have to defend the framework aloud. Our four-part script for explaining a framework in a viva shows how to turn these parts into a spoken answer.

Printable conceptual framework worksheet

Fill in one row per concept and one row per relationship, then tick the checks before you draw the final diagram.

Research question: ______________________________________________

ConceptOne-sentence definitionSourceHow it is measured or explored
FromLink (raises, lowers, changes the strength of, associated with)ToBasis in the literatureResearch question it serves

Concepts I left out, and why: ______________________________________________

  • Every concept traces back to my research question
  • Every concept has one definition and a citation
  • Every relationship has a source in the basis column
  • Every concept has a place in my methods where I collect data on it
  • My diagram has a legend that says what a line and an arrow mean
  • I used no causal arrows if my study is qualitative
  • I checked my handbook for where the framework goes and what it is called
  • My supervisor has seen the diagram and the narrative

Frequently asked questions

What is a conceptual framework in simple words?It is a map of the main ideas in your study and how you expect them to connect, often a diagram plus a short explanation. It shows your reader which concepts you will examine, what each one means, and why you think they are linked, with a source for each part.
What are the elements of a conceptual framework?The guides we read group the parts differently. They include concepts or variables, the relationships between them, a source for each part, and a diagram with a written explanation. Stating what you left out, and why, is also worth adding, because an examiner may ask about it.
Can I use an existing conceptual framework?Yes, if your handbook allows it. The Valdosta State guide says you may adopt an existing conceptual framework or develop your own modified version based on the literature review. Cite the original, and explain what you changed so the framework fits your question.
Is a diagram required?Not always. National University’s guide calls the graphic optional but says the framework should always be explained and cited. Your committee may expect one, so check your handbook. Even a simple table of concepts and relationships helps your reader follow your logic.
Does a qualitative study need a conceptual framework?It can have one, but it looks different. It groups the concepts you will explore instead of drawing arrows to test, and it can change during analysis. National University’s guide, for one, advises leaving out arrows that imply causal links in a qualitative framework.
How do I explain my conceptual framework in a viva or defense?Name each concept and its definition, say how you expect them to connect and why, then show where each one appears in your methods and findings.

Sources

How this guide was made: Thesisroom drafted this guide with the help of AI writing tools. Every fact, figure and source in this post was checked against its original source before publishing.