Integrating data analyses in mixed methods research

Although the epistemological arguments of the ''paradigm wars'' sharpened our thinking about issues related to mixed methodology, their lingering legacy has been to slow the progress of integration of methods. All mixed methods studies, by definition, attempt some form of integration, but the paradigm debates have made many researchers nervous about integrating the various strands of their research before they reach the point of drawing conclusions. There has, indeed, been a degree of illogic in the way some researchers have dealt with the issue of integration of data sources and analyses, where it has been considered epistemologically unacceptable to combine these, and yet desirable to corroborate or integrate conclusions drawn from data generated through diverse perspectives. In any case, as noted many years ago by Miles and Huberman (1994), these arguments are largely unproductive. From a pragmatic perspective, the primary issue is to determine what data and analyses are needed to meet the goals of the research and answer the questions at hand. Alternatively, a realist perspective encourages one to seek both process- and variable-oriented data to both detect regularities and understand the mechanisms by which they occur (Maxwell, 2008)