03 · What You Need to Know
How to Develop a Conceptual Framework Step by Step
First, Understand What You Are Actually Building
A conceptual framework is an organized account of the ideas that help you understand and investigate your research problem.
Different methodological traditions define its boundaries somewhat differently. Ravitch and Riggan describe conceptual frameworks as both a process and a product and emphasize their role in connecting the literature review, research design, and methodology. Ravitch and Carl similarly describe conceptual frameworks as guiding and grounding empirical research and as something researchers continue to develop and refine rather than complete once and forget.
That is a useful way to think about the task. Your framework is not merely the figure placed near the end of your literature review. It is part of the reasoning that holds the study together.
Depending on your research tradition, it may incorporate formal theory, findings from previous research, important concepts, contextual knowledge, assumptions, and your reasoned understanding of how these pieces fit together. Theory can therefore inform a conceptual framework without making the two terms synonymous. If that distinction is still unclear, establish the difference between theoretical and conceptual frameworks before proceeding.
Step 1: Start With the Research Problem
Do not begin by opening presentation software and drawing rectangles.
Write down the problem your research is trying to understand. Then write the central research question beside it.
Ask:
- What phenomenon am I trying to understand?
- What exactly is unknown, uncertain, unexplained, or worth investigating?
- Who or what is being studied?
- What relationships, processes, experiences, or conditions are central to the question?
Your framework should emerge in response to that problem. If a concept does not help explain, frame, or investigate the problem, ask why it needs to appear.
Step 2: Identify the Core Concepts in the Question
Break your research question into its important conceptual pieces.
Suppose a dissertation asks:
How are doctoral students' perceived supervisor support and academic self-efficacy associated with their persistence intentions?
Three concepts are immediately visible:
- perceived supervisor support;
- academic self-efficacy; and
- persistence intentions.
That does not mean those three automatically constitute the final framework. They are starting points.
You now need to determine what each concept means, whether the literature treats it consistently, whether it represents a variable or a broader construct, and what evidence supports connecting it to the others.
This is where understanding variables, constructs, and operational definitions prevents conceptual confusion later.
Step 3: Use the Literature to Discover What Else Matters
Your research question gives you a starting structure, but the literature helps you determine whether that structure is too simple, incomplete, or misguided.
As you read, do more than collect findings. Look for patterns in how researchers explain the phenomenon.
Record:
- concepts that repeatedly appear;
- theories researchers use to explain the phenomenon;
- relationships that have empirical support;
- relationships for which evidence is inconsistent;
- important contextual or boundary conditions;
- competing explanations;
- concepts that sound similar but are defined differently; and
- assumptions underlying previous studies.
A conceptual framework should represent an argument informed by this literature, not a vote in which every frequently mentioned variable earns a box.
Step 4: Decide Which Concepts Actually Belong
At this point, researchers often face the opposite problem: too many possible concepts.
You may find dozens of factors associated with your outcome. Including all of them does not make the framework comprehensive. It can make the framework unfocused and impossible to investigate coherently.
For each candidate concept, ask:
- Is it necessary to answer my research question?
- Is there a theoretical reason to include it?
- Is there credible empirical evidence that makes it relevant?
- Can I define it clearly?
- Can my research design actually investigate it?
- What would be lost if I removed it?
A framework should be selective. Its purpose is to make the study's logic clearer, not to display everything you discovered during the literature review.
Step 5: Define Every Concept Before Connecting Them
Before drawing relationships, make sure you know what the concepts mean.
Terms such as “engagement,” “support,” “achievement,” “motivation,” “quality,” and “success” can mean very different things across studies. Two papers may use the same word while measuring substantially different constructs.
Create a working definition for every major concept and trace that definition to the literature or theory informing your study.
Later, if a concept must be measured or observed, you will need to specify how it becomes an operational definition. Do not collapse the conceptual and operational stages too early. First establish what the concept means; then determine how your study will observe it.
Step 6: Specify What Each Relationship Means
This is one of the most important steps and one of the most frequently skipped.
Suppose your framework contains:
Supervisor support → Academic self-efficacy → Persistence
What does the first arrow mean?
Does supervisor support cause higher self-efficacy? Predict it? Associate with it? Contribute to it? Shape students' perceptions of their capabilities? Does self-efficacy mediate a relationship between support and persistence?
Those are different claims.
Do not draw an arrow until you can complete this sentence:
“This connection means that I am proposing that...”
Then ask what evidence justifies the proposition and whether your research design can evaluate it.
Watch Out
Directional arrows often look causal. If your study is cross-sectional, observational, or otherwise unable to establish causal effects, do not let the diagram visually claim more than the research design can support. Use labels, non-directional connections, or explanatory text when necessary.
Step 7: Consider the Unit and Level of Analysis
A concept can make sense individually while creating problems when connected to concepts measured at another level.
Imagine that your framework connects an individual student's self-efficacy to “university research culture.” One is an individual-level construct; the other may refer to a department, faculty, institution, or students' perceptions of that environment.
You need to know which.
Before finalizing relationships, clarify what your study is actually studying. A framework that silently moves between individual, group, classroom, organizational, and societal levels can encourage conclusions the data cannot support.
Step 8: Decide Whether Theory Belongs in the Framework
A conceptual framework can be informed by formal theory, but do not insert a theory merely because you think a dissertation needs a famous name attached to the model.
Ask what the theory contributes. Does it explain why particular concepts matter? Does it propose relationships your study examines? Does it help interpret the phenomenon?
If it does, integrate the relevant theoretical reasoning explicitly. If it does not, do not force the theory into the framework after the fact.
Theories should help explain your framework rather than function as decorative labels.
Step 9: Draft the Framework in Words Before Drawing It
Try writing one short paragraph that explains the framework without a figure.
For example:
“This study conceptualizes doctoral persistence intentions as related to students' perceived supervisor support and academic self-efficacy. The framework proposes that students who perceive stronger academic support may report stronger beliefs in their ability to perform doctoral research tasks, while both support and self-efficacy may be associated with intentions to persist. The proposed relationships are grounded in the theoretical and empirical literature reviewed in this study.”
This is hypothetical wording, not a claim about actual relationships.
If you cannot explain your framework clearly in prose, drawing it usually will not solve the underlying conceptual problem.
Step 10: Draw the Simplest Diagram That Accurately Represents the Logic
Now the boxes and arrows become useful.
Each major concept can be represented visually, and relationships can be shown using consistent lines, arrows, grouping, or labels. Keep the representation as simple as the research problem allows.
A useful conceptual framework diagram should let a reader answer:
- What are the central concepts?
- Which concepts are related?
- What direction or type of relationship is proposed, if any?
- Which elements are outcomes, explanatory concepts, contextual factors, or other relevant components?
- What does the study actually intend to investigate?
Do not add visual complexity merely to make the framework look doctoral-level. Complexity should come from the phenomenon when necessary, not from extra shapes.
Step 11: Test the Framework Against the Research Design
A conceptual framework can look convincing while being impossible to investigate with the proposed methods.
Take each important concept and relationship and ask what evidence would be required to address it.
If your framework contains a construct you cannot observe directly, determine what indicators or evidence would represent it. This is particularly important for concepts such as trust, motivation, identity, self-efficacy, organizational culture, or perceived support. Understanding how researchers connect unobservable constructs to indicators and proxies helps prevent a framework from becoming disconnected from data collection.
Then compare the entire framework with your research questions, sampling, instruments or data sources, and planned analysis.
Ravitch and Riggan explicitly present conceptual frameworks as mechanisms for aligning literature review, research design, and methodology, while Ravitch and Carl describe frameworks as guiding and grounding empirical research.
Step 12: Treat the Framework as Something You Can Refine
Your first conceptual framework does not have to be your final one.
Framework development is often iterative. Reading may expose a missing concept. A pilot interview may reveal that two concepts you treated as separate are difficult to distinguish in the context of your study. Your committee may identify an unsupported relationship. Refining the research question may make part of the framework irrelevant.
Ravitch and Carl explicitly describe conceptual frameworks as dynamic and discuss their continued development and refinement across research.
Revision is therefore not evidence that the original framework failed. It can be evidence that your understanding of the study has improved.
04 · A Practical Example
Building a Conceptual Framework for a Dissertation
Hypothetical Example
A Dissertation Student Studying Doctoral Persistence
Imagine a dissertation student, Maya, wants to investigate why some doctoral students remain committed to completing their dissertations while others seriously consider leaving. Her early idea is that “support affects persistence.”
That statement is too vague for a conceptual framework. What kind of support? Support from whom? What does persistence mean? Is she studying actual completion, current enrollment, intention to continue, or something else?
1. Clarify the question Maya narrows the study to the relationship among perceived supervisor support, academic self-efficacy, and doctoral students' intentions to persist.
2. Define the concepts She reviews the literature to determine how supervisor support, academic self-efficacy, and persistence intentions have been conceptualized and distinguishes them from related concepts.
3. Examine theory and evidence She identifies theoretical arguments and empirical findings relevant to why support and beliefs about one's capabilities might be related to persistence. She also records contradictory findings and contextual qualifications rather than collecting only evidence that supports her initial idea.
4. State the proposed relationships Before drawing arrows, Maya writes what each relationship is intended to mean and avoids causal wording that her proposed design cannot justify.
5. Draw the framework She creates a simple model containing only the concepts and relationships central to her research question, rather than adding every factor associated with doctoral completion in the literature.
6. Test the framework against the study She checks whether her research questions, participant selection, measures, and planned analysis can actually address each important part of the framework.
7. Revise After discussing the model with her committee and revisiting the literature, she removes one poorly justified relationship and clarifies the meaning of another.
The important change is not that Maya now has a polished diagram. She has moved from the vague statement “support affects persistence” to a defensible account of what she intends to study and why those particular concepts and relationships belong together.
A dissertation does not become stronger because its framework contains more variables. It becomes stronger when the framework makes the study's reasoning easier to inspect and defend.