Mixing Survey and Qualitative Data 1 |
|
Convenor | Professor Nina Baur (Technische Universität Berlin ) |
Coordinator 1 | Dr Leila Akremi (Technische Universität Berlin) |
Coordinator 2 | Ms Melanie Wenzel (Technische Universität Berlin) |
Attitude scales are a popular method of empirical prejudice research yet often involve considerable unobserved heterogeneity. Furthermore, in the light of current critiques of the prejudice-as-attitude perspective, such scales appear to lack content validity: If prejudice expressions often take an ambivalent and fragmented form and need to be interpreted within their discursive context, standardized measurements provide only a poor indication of such rhetorical practice. Mixed methods approaches offer a productive alternative for empirical prejudice research. This is exemplified by a study on current anti-Americanism in Germany combining survey data and qualitative interviews in a sequential explanatory design.
Presenting the research design of our project “Using maps for orientation and interaction in space”, I will discuss the advantages and challenges in using a complementarity mixed methods approach (Greene et al. 1989). Data collection was organized as a social field experiment combining surveys and ethnographic methods.
Advantages are: having different kinds of data, thus being able to compensate shortcomings of one method, and therefore controlling for compounding factors and ability to verify our results. Challenges we are dealing with are: maintaining an overview, processing lots of data, and deciding at what points we should combine different data and how.
The presentation will focus on three methodological aspects that arise when mixing survey data and qualitative interviews: (1) Purposive sampling based on secondary analyses and the sample of a quantitative survey should be followed by theoretical sampling based on the qualitative findings to fully exploit the potential of qualitative analyses. (2) Information from survey data can be used for sampling interviewees but have to be verified during field work. (3) Findings of survey data and qualitative interviews construct the phenomenon in different ways so that a joined interpretation implies decisions how to compare the results.