03 · What You Need to Know
Understanding the Paradigm Problem in Mixed-Methods Research
Where Did the Compatibility Problem Come From?
The debate is partly a legacy of the so-called paradigm wars.
Quantitative research was often associated with positivism or postpositivism. Qualitative research was associated with interpretivism or constructivism. If methods were inseparable from those paradigms, combining quantitative and qualitative research seemed to require combining philosophies that disagreed about reality, knowledge, objectivity, values, and researcher involvement.
This became known broadly as an incompatibility problem.
Mixed-methods scholarship challenged the assumption in several ways. One response was to argue that methods are not inherently tied to one paradigm. Another was to use pragmatism as a philosophical basis for methodological pluralism. A third was to retain multiple paradigms deliberately and bring them into dialogue. These debates remain relevant because they concern not merely which techniques can coexist but how findings generated under different assumptions can be interpreted together.
Methods Themselves Are Not Automatically Incompatible
An important distinction comes first: combining methods is not identical to combining paradigms.
An interview does not permanently belong to interpretivism. Statistical analysis does not permanently belong to positivism. As discussed in the broader question of whether the same research method can be used under different paradigms, methods acquire philosophical significance through the assumptions and purposes governing their use.
Mixed-methods research was originally conceptualized in influential work as combining quantitative and qualitative methods without assuming that either type was inherently linked to one particular inquiry paradigm.
Consequently, a study using a questionnaire and interviews does not automatically contain two incompatible paradigms.
Compatibility Becomes More Important When Assumptions Conflict
Suppose researchers measure "student engagement" using a standardized scale and then conduct interviews about what engagement means to students.
The quantitative analysis may initially treat engagement as a sufficiently stable construct that individuals can be assigned comparable scores.
The qualitative analysis may reveal that students use "engagement" to describe fundamentally different experiences across disciplines, learning environments, and cultural contexts.
Now the issue is no longer simply methodological.
If the final interpretation assumes both that engagement has one stable meaning captured by the scale and that engagement has no stable meaning across participants, the study may contain a contradiction requiring explanation.
The researchers could revise the quantitative interpretation, distinguish different levels of the construct, treat the contradiction as an important finding, or justify another conceptual relationship. What they should not do is ignore the tension.
Ontology Can Create the Deepest Compatibility Problems
Ontological assumptions concern what exists and the nature of reality.
Suppose one component assumes that a social construct exists as a stable phenomenon independently of people's interpretations. Another assumes that the phenomenon is constituted entirely through those interpretations.
Those claims may be difficult to maintain simultaneously if they refer to exactly the same thing at exactly the same level.
But apparent ontological conflict can sometimes be resolved by greater precision.
A critical realist study, for example, might argue that institutional structures exist and exert causal influence while also recognizing that people's interpretations of those structures affect how they respond. The existence of structures and the social construction of meanings need not be mutually exclusive claims.
The question is therefore not whether two paradigms use different vocabulary. It is whether their substantive claims about reality contradict one another in the context of the study.
Epistemological Differences May Be Easier to Combine
Epistemology concerns what can be known and how knowledge is justified.
A postpositivist component may estimate an empirical relationship while treating measurement as fallible. An interpretive component may develop contextual understanding of how participants make sense of the same phenomenon.
These forms of knowledge are different, but difference does not automatically imply contradiction.
One may answer "How much?" while another answers "What does this mean to participants?" or "How does this process unfold?"
If the questions and claims are complementary, the components may coexist without requiring identical epistemologies.
This distinction is one reason compatibility should be evaluated at the level of actual assumptions and claims rather than through paradigm labels alone.
Compatibility Is Not the Same as Sameness
Two components do not need to make identical assumptions to be compatible.
Consider a multidisciplinary research team. A statistician and an ethnographer may approach evidence differently while still contributing coherent knowledge to the same problem.
The same principle can apply philosophically.
Philosophical uniformity
All components operate from substantially the same paradigm and assumptions.
Paradigmatic compatibility
Different assumptions can coexist without producing unresolved contradictions that undermine the study's integration or conclusions.
Dialectical plurality
Differences or contradictions are intentionally preserved and examined because their interaction contributes to understanding.
The third possibility is especially important. Compatibility need not always mean harmonizing every perspective.
What Is the Incompatibility Thesis?
The incompatibility thesis broadly holds that certain philosophical paradigms contain sufficiently different assumptions that methods or findings grounded in them cannot simply be combined into one coherent inquiry.
Historically, the argument was particularly associated with contrasts between positivist and constructivist or interpretivist traditions.
The strongest version creates a problem for mixed methods: if quantitative and qualitative methods are inherently tied to incompatible paradigms, then meaningful integration becomes philosophically impossible.
Mixed-methods scholars have challenged that conclusion from several directions. Some reject the assumption that methods are paradigm-bound. Others accept paradigmatic differences but argue that dialogue across them can be productive. Still others adopt an overarching philosophy that accommodates multiple forms of evidence. The debate has therefore moved beyond a simple choice between total compatibility and total incompatibility.
Pragmatism Offers One Response
Pragmatism became influential partly because it redirects attention from rigid paradigm-method ownership toward inquiry, purposes, actions, and consequences.
A pragmatic researcher can ask what quantitative and qualitative evidence each contributes to addressing the research problem without assuming that the methods arrive carrying permanently incompatible philosophies.
This is one reason pragmatism became closely associated with mixed methods.
Yet mixed-methods research does not have to use pragmatism. Nor does writing "this study is pragmatic" automatically solve genuine conceptual contradictions between components.
Pragmatism provides a philosophical position. It is not a compatibility reset button.
A Dialectical Approach Gives a Different Answer
Dialectical mixed-methods approaches do not necessarily try to eliminate paradigmatic differences.
Instead, researchers may intentionally bring different perspectives into dialogue. A postpositivist analysis and a constructivist interpretation can be allowed to ask different questions, generate different understandings, and even challenge one another.
Greene and related mixed-methods scholarship treat such engagement as potentially productive because paradox, contradiction, and difference can lead researchers to reframe questions and develop new understanding.
This approach requires philosophical transparency. If perspectives disagree, researchers need to explain the disagreement and what is learned from it rather than quietly merging conclusions as though the underlying assumptions were identical.
Transformative Approaches Provide Another Basis for Integration
A transformative paradigm may organize quantitative and qualitative evidence around shared commitments to social justice, power, cultural responsiveness, marginalized perspectives, and human rights.
In this case, compatibility may be established through the broader philosophical framework governing both components.
A quantitative analysis could document disparities while qualitative inquiry investigates how those disparities are experienced and produced. Both components serve the same transformative purpose even though their evidence and analytical procedures differ.
Mertens explicitly argues that paradigms should be understood through assumptions about ethics, reality, epistemology, and systematic inquiry rather than simply equated with quantitative, qualitative, or mixed methods.
Critical Realism Can Accommodate Different Forms of Evidence
Critical realism provides another route.
Quantitative evidence can identify empirical patterns. Qualitative evidence can illuminate experiences, processes, and contexts. Documentary evidence can reveal institutional structures.
These different forms of evidence can be integrated within one realist explanatory framework concerned with the mechanisms capable of generating observed outcomes.
The components need not share identical methods because they contribute evidence about different dimensions of a stratified reality.
What Happens When Findings Conflict?
Contradictory findings are often treated as a problem to be fixed. In mixed methods, they can be analytically valuable.
Greene, Caracelli, and Graham's influential framework identified several purposes for combining methods, including triangulation, complementarity, development, initiation, and expansion. Initiation specifically uses paradoxes and contradictions between methods to generate new perspectives or reframe questions.
Suppose survey results show high student satisfaction while interviews reveal substantial frustration.
The appropriate response is not necessarily to decide which dataset is correct.
Perhaps the survey scale measures satisfaction with course outcomes while interviews reveal dissatisfaction with workload. Perhaps students interpret the response categories differently. Perhaps social desirability affects one method more than another.
The contradiction may expose assumptions that neither component reveals independently.
Compatibility Should Be Evaluated at Several Levels
| Level |
Compatibility question |
| Ontology |
Do the components make contradictory claims about what exists or what the phenomenon fundamentally is? |
| Epistemology |
Can the different forms of knowledge be related without treating one as invalid by definition? |
| Axiology |
Do the components assume substantially different roles for values, and can those differences be justified? |
| Methodology |
Do the inquiry strategies make sense together given the overall research purpose? |
| Evidence |
What does each form of evidence represent, and are those interpretations compatible? |
| Integration |
Can findings be connected, compared, or contrasted without making contradictory assumptions invisible? |
| Inference |
Does the final conclusion claim more philosophical unity than the study actually supports? |
This approach is more useful than asking whether two paradigm names are officially compatible.
Not Every Mixed-Methods Study Has a Major Compatibility Problem
Sometimes the philosophical issue is modest.
A pragmatic study may use a survey to estimate prevalence and interviews to understand implementation barriers. Both methods operate under one coherent pragmatic framework.
A critical realist study may use statistics and interviews as different sources of evidence about patterns and mechanisms.
In such cases, researchers do not need to manufacture a paradigm conflict simply because two data types are present.
The philosophical burden should be proportionate to the actual problem.
Compatibility Matters Most at the Point of Integration
Separate components can sometimes coexist easily until researchers attempt to draw one overall conclusion.
This is where hidden contradictions become visible.
If one strand treats a variable as a stable representation of reality while another concludes that the variable has no coherent meaning across participants, what does the integrated conclusion say?
If one component seeks a context-independent causal effect while another argues that the phenomenon cannot be separated meaningfully from context, how will those claims be related?
Integration requires researchers to decide whether findings converge, complement one another, address different levels, expose contradiction, or require revision of the original conceptualization.
Paradigmatic compatibility therefore matters because integration is not merely the physical placement of two results sections beside each other.
Watch Out
Do not assume that using two methods creates a philosophical problem, but do not assume that calling the study "mixed methods" removes one either. Examine the assumptions that become consequential when evidence and conclusions are integrated.
