03 · What You Need to Know
A Framework Can Evolve, but You Cannot Rewrite the History of the Study
A theoretical framework is constructed for a particular study
A theory and a theoretical framework are related but not identical.
The theory is the more abstract explanatory structure. A theoretical framework is produced when researchers translate relevant concepts, propositions, and relationships from theory into the specific phenomenon and context being investigated.
Recent methodological guidance on qualitative research explicitly describes a theoretical framework as both a process and a product. Researchers make decisions about how abstract theoretical ideas apply to the phenomenon under investigation, and those decisions need to be justified.
This matters because the framework you construct before entering the field is based partly on what you understand about the phenomenon at that point. Research may subsequently reveal that some of those assumptions were incomplete.
Qualitative research provides the clearest case for legitimate change
Qualitative inquiry is often iterative rather than strictly linear. Data collection and analysis may occur in overlapping cycles, with early observations changing what researchers ask, where they look next, and how they conceptualize the phenomenon.
Current methodological guidance is unusually explicit on this point. Stalmeijer and Varpio note that a theoretical framework developed before data collection in a fully theory-informed inductive qualitative study may need to change as researchers work with participants and develop a more accurate understanding of the phenomenon.
Recent empirical research with qualitative research experts similarly describes theory as capable of entering, leaving, broadening, or becoming more specific at different stages of a study.
Changing a framework in this context can therefore reflect methodological responsiveness rather than methodological failure.
Not every methodology permits the same degree of theoretical flexibility
The legitimacy of change depends partly on what kind of study you are conducting.
An exploratory qualitative study may deliberately permit theory to evolve during data collection and analysis. Grounded theory methodologies are explicitly iterative, with emerging theoretical categories informing subsequent data collection.
A preregistered confirmatory experiment operates under different expectations. If researchers specify theoretical predictions in advance and then replace the framework after seeing the results, the new framework cannot honestly be described as the basis from which the original confirmatory hypotheses were derived.
The framework can still change as part of interpretation or subsequent theory development. What changes is the status of the claim.
Prospective theoretical reasoning
The framework or theoretical prediction was specified before the relevant evidence was examined and genuinely informed the study.
Post hoc theoretical reasoning
A theoretical explanation was developed or selected after researchers encountered the evidence it is intended to explain.
Both can contribute to scholarship. They simply should not be presented as though they were the same process.
A framework may change because the phenomenon was initially misunderstood
Imagine that researchers begin with a framework treating individual students as the principal decision-makers in a learning activity. Early observations reveal that instructors actually structure most of the relevant interactions and strongly determine what students can do.
The framework's original representation of agency may therefore be wrong for the empirical setting.
Maintaining the original framework merely because it appeared in the proposal would preserve procedural consistency at the expense of conceptual accuracy.
Revision can be the more defensible choice when emerging evidence demonstrates that the framework was constructed around an inaccurate understanding of the phenomenon.
A framework may change because an important concept is missing
Data may repeatedly reveal something the initial framework cannot represent.
Suppose a study of faculty adoption of generative AI begins with an individual-level theory emphasizing beliefs and intentions. During interviews, institutional restrictions repeatedly emerge as central to whether faculty members can act on those intentions.
The researchers now face a theoretical decision. They could treat institutional restriction as contextual information, extend the original framework, introduce another theoretical perspective, or replace the framework if the original explanation proves fundamentally inadequate.
The appropriate response depends on how central the missing concept is and what theoretical claim the study ultimately intends to make.
A framework may become narrower as the study becomes more focused
Change does not always mean adding concepts.
An initial framework may be broad because researchers do not yet know which mechanisms matter most. As data collection and analysis proceed, the study may reveal that only one part of the framework is relevant.
Theoretical focus can then narrow.
This can be entirely appropriate in iterative research. Recent qualitative research experts have described theory as sometimes beginning broadly and becoming more specific during the research process.
The question is whether the narrowing follows emerging understanding or merely removes theoretical components that produced inconvenient results.
You may discover that another theory fits better
Sometimes revision is not enough.
An initial theory may emphasize a mechanism that turns out to have little explanatory value, while another theoretical perspective accounts much more convincingly for what participants do or what the empirical patterns show.
Changing theories can then be defensible, particularly in exploratory or abductive research.
Abductive theorizing involves moving between empirical observations and possible explanations, especially when surprising evidence does not fit existing expectations. Research on theorizing in qualitative inquiry has shown how empirical findings can lead researchers to reassess and refine their theoretical framework rather than remaining committed to the original explanation.
This is closely related to the problem of different theories explaining the same phenomenon. A study may reveal only after deeper engagement with the phenomenon which theoretical perspective provides the more useful explanation.
Changing the framework can change the research question
Theory and research questions influence each other.
If the framework changes substantially, the question may need reconsideration because the new theory could foreground different actors, mechanisms, relationships, or levels of analysis.
Recent methodological guidance provides an explicit example in which changing the theoretical framing of a qualitative study also changes the research questions because the researchers come to understand the central activity differently.
This does not mean every theoretical adjustment requires rewriting the research question. Minor refinements may leave it intact.
But if the framework changes what the study is fundamentally asking about, pretending that the original question remains untouched can create conceptual inconsistency.
Changing the research question after the study starts has consequences
Once data collection begins, a revised question may affect more than wording.
Different questions may require different participants, observations, interview prompts, variables, time points, or analytical strategies. Existing data may no longer be sufficient to answer the revised question.
Researchers therefore need to trace the consequences of theoretical change through the study.
Framework changes A different mechanism or theoretical perspective becomes central.
Recheck the question Determine whether the existing research question still represents the inquiry accurately.
Recheck the evidence Determine whether the existing sampling and data collection can address the revised theoretical question.
Recheck the analysis Determine whether coding, variables, comparisons, or analytical procedures need modification.
Recheck the claims Separate what was established prospectively from interpretations or propositions developed after seeing the evidence.
You cannot claim that the new framework guided decisions it did not guide
This is perhaps the most important reporting principle.
Suppose you collected interviews using Theory A, discovered unexpected patterns, and later interpreted those patterns using Theory B.
That can be methodologically legitimate.
What you should not write is that “Theory B guided the development of the interview protocol” if Theory B was not selected until after the interviews were completed.
Likewise, if hypotheses were developed under Theory A, do not retrospectively present them as predictions derived from Theory B merely because Theory B explains the results more successfully.
Watch Out
Changing theory after encountering the data is not inherently problematic. Concealing that chronology is. Readers need to know which theoretical ideas shaped the original design and which emerged during analysis or interpretation.
Changing theory after seeing results can become HARKing
In confirmatory research, particular caution is required when theoretical changes occur after results are known.
If researchers observe an unexpected result, develop a new theoretical explanation, and then report the resulting hypothesis as though it had been specified beforehand, they blur the distinction between prediction and post hoc explanation.
Post hoc theorizing itself is not the problem. Unexpected findings often generate valuable theoretical development.
The problem is presenting an explanation generated by the evidence as a successful prediction of that same evidence.
A more transparent approach is to report the original theoretical expectation, describe the unexpected finding, present the alternative theoretical interpretation as post hoc or exploratory, and identify how future research could test it independently.
A revised framework may require new data
Imagine that your original theory focuses on individual attitudes, so your interviews contain detailed questions about beliefs and intentions. Halfway through analysis, you adopt an organizational theory emphasizing leadership, policy, and departmental routines.
Your existing interviews may contain occasional comments about these issues, but that does not mean the dataset provides adequate evidence for the new theoretical questions.
If the design permits iteration, additional interviews, observations, documents, or participants may be needed.
Changing the framework without checking whether the evidence can support it risks producing an interpretation whose theoretical sophistication exceeds the data available.
A revised framework may require changes to coding or analysis
Theory can influence what researchers code, compare, model, or interpret.
If the theoretical framework changes, analytical procedures may need to change as well. In qualitative research, this might involve revisiting earlier transcripts with revised sensitizing concepts or adding new codes while retaining inductively developed categories.
Iterative qualitative analysis explicitly involves revisiting data as new questions and connections emerge. This recursive movement can deepen interpretation when the process is documented rather than concealed.
In quantitative research, changing the theoretical model may require new variables or analyses that were not part of the original confirmatory plan. Such analyses can be informative, but they should be identified appropriately as exploratory or post hoc where relevant.
Ethics approval, preregistration, protocols, and institutional requirements may also be affected
A theoretical change does not automatically require a new ethics application. A change that substantially alters research questions, participant groups, recruitment, procedures, data collected, or risks might.
The requirements depend on the institution and approved protocol, so researchers should check with the relevant ethics or institutional review authority rather than assuming that a theoretical change is administratively inconsequential.
Similarly, if the study is preregistered, registered, governed by a protocol, or part of a funded project, deviations should be handled according to the applicable requirements.
Theoretical flexibility does not override research governance.
Changing the framework can be a finding rather than an embarrassment
If a framework changes because empirical evidence reveals that the original theoretical assumptions were inadequate, the evolution itself can be intellectually informative.
You may discover that the phenomenon operates at a different level than expected, that an assumed mechanism is absent, that a theoretical boundary has been crossed, or that another explanation performs better.
Such findings can contribute to challenging an existing theory or extending its explanatory boundaries.
The original framework was not necessarily a mistake. It may have provided the intellectual contrast that made the limitation visible.
Do not change frameworks merely because the results are disappointing
Suppose your theoretically predicted relationships are not statistically significant. You search the literature, discover another theory compatible with the observed pattern, and replace the original framework in the manuscript.
This makes the finished research story cleaner. It also removes evidence that the original theoretical expectation was not supported.
A more defensible approach preserves the chronology. Report what was predicted, what was observed, and what alternative explanation emerged afterward.
Theory should help researchers learn from evidence, not provide an unlimited supply of emergency exits whenever the p-values become impolite.
The later the change, the more carefully its consequences should be examined
Changing a framework during early proposal development is relatively straightforward. Changing it after preliminary qualitative interviews may be compatible with an iterative design. Changing it after all data have been collected requires careful consideration of whether the existing evidence can support the new framework. Changing it after final results are known creates additional concerns about post hoc explanation.
| When the change occurs |
Typical concern |
What to check |
| Before data collection |
Conceptual alignment |
Revise questions, framework, methods, and approvals before proceeding. |
| During iterative data collection |
Whether emerging evidence justifies theoretical refinement |
Revisit sampling, questions, data collection, and analysis. |
| After data collection but before final analysis |
Whether existing data adequately address the revised framework |
Assess data sufficiency and distinguish original from revised theoretical reasoning. |
| After results are known |
Post hoc theorizing and retrospective prediction |
Preserve original expectations and label new explanations transparently. |
Documentation is what turns flexibility into defensible research
When the framework evolves, keep a record of what changed, when, why, and what consequences followed.
In qualitative research, reflexive notes, analytic memos, decision trails, and revised framework diagrams can document this development. In preregistered or protocol-driven research, formal amendments and deviation statements may serve related purposes.
The purpose is not bureaucratic perfection. It is to make the intellectual development of the study reconstructable.
Recent qualitative research experts have noted that the “story” of how theory changes during a study can be difficult to convey in a finished paper precisely because published research often compresses an iterative process into a linear narrative. That compression should improve readability without creating a false history of how the theoretical reasoning actually developed.
