Explanatory Research

The term explanatory research implies that the research in question is intended to explain, rather than simply to describe, the phenomena studied. This type of research has had a contested history in qualitative inquiry, and divergent views of the appropriateness of such goals in qualitative research are currently held. This entry summarizes the current state of this debate and describes some of the most important qualitative methods for such explanation.

Traditionally, the research denoted by the term explanatory research has been quantitative in nature and has typically tested prior hypotheses by measuring relationships between variables; the data are analyzed using statistical techniques. In the narrowest sense, this term is sometimes used synonymously with experimental research, with the implication that only experiments are capable of answering causal questions. More broadly, it can also include other types of quantitative research grouped under terms such as causal modeling and structural equation modeling, which attempt to identify causal relationships through the analysis of correlations between variables.

However, the terms causal and explanatory have also been applied to various types of qualitative research, although these uses have been controversial both within and outside of the qualitative research community. Such uses were more common in the earlier history of qualitative research, but with the inception of [Page 324]the “paradigm wars” during the latter part of the 20th century, the very idea of causation became problematic in qualitative research. The prevalence of the view that only quantitative methods can be used to investigate causality led many writers to avoid making explicit causal claims in their work, whereas other qualitative scholars have argued that the entire concept of causality is illegitimate or inappropriate in qualitative research.

The debates about what research counts as “explanatory” have taken on major political dimensions since 2000, as advocates of what they call “science-based research,” which privileges the use of randomized control trials (RCTs) as the “gold standard” for causal explanation, gained control of federal funding for educational research in the Bush administration. (A similar development occurred earlier in Great Britain.) Although there has been widespread criticism of this position, it has many adherents and its influence has been felt beyond educational research.

Despite this, the use of terms such as influence, impact, affect, and contribute to, is common in qualitative research reports, and such terms imply causality in some sense. In addition, a growing number of researchers (both qualitative and quantitative) now argue that, in some circumstances, quantitative approaches are not necessarily the best (or only) ways of reaching explanatory conclusions and that qualitative methods can be used to systematically develop and test causal explanations.

There are several important criticisms of randomized experimental designs as the preeminent research strategy for explanatory purposes. First, in many situations, and for some issues, it is difficult or impossible to rigorously implement such designs, and many purported RCTs have in fact been so flawed that their causal conclusions are questionable. Second, many other types of research (ranging from quasi-experiments, to causal modeling, to qualitative approaches) can establish causal conclusions, not with certainty (no method can do this) but beyond reasonable doubt.

...