A dialectic data integration approach for mixed methods survey validation

Abstract

In quantitative studies, surveys are often used as tools to gauge attitudes, knowledge, and behaviors. Often, when these surveys are used in new contexts or with new populations, they require validation procedures such as confirmatory factor analysis or comparison to similar measures. These methods, however, are bounded by the need for large sample sizes which are not always feasible. In this paper, we discuss the use of mixed-methods research for survey validation. We present an example study that incorporates both traditional quantitative data and qualitative data into the validation of a survey targeted at engineering students. First, we present the philosophical underpinnings of quantitative and qualitative validation and discuss connections that allow both traditions to be incorporated into the same survey validation. Second, we discuss how quantitative and qualitative data can be mixed to form a deeper understanding of the participants, their educational context, and how survey results might be interpreted in that context among those participants. This paper contributes to research in engineering education by providing a dialectic data integration approach to support survey validation through the use of mixed methods.