ESRA logo
Tuesday 14th July      Wednesday 15th July      Thursday 16th July      Friday 17th July     

Thursday 16th July, 16:00 - 17:30 Room: L-102

Mixed methods designs combining survey data and qualitative data 2

Convenor Professor Mark Trappmann (Institute for Employment Research (IAB) )
Coordinator 1Dr Andreas Hirseland (Institute for Employment Research (IAB))

Session Details

In social science research there is a long tradition of research combining survey data with qualitative data. There can be various reasons why this integration of approaches provides advantages compared to a single method approach and how it is implemented in a research design. Greene et al. (1989) propose a typology consisting of five types of mixed-method designs. Triangulation involves investigating the same aspect of the same phenomenon. If research methods bias results, there is a chance of detecting this bias by using different methods independently. In contrast, complementarity involves investigating different aspects of the same phenomenon by different methods. Results from one method are used to elaborate, enhance, or illustrate results from the other. Development designs sequentially use one method to develop or support the other method. Examples include using qualitative interviews in questionnaire design or residual diagnostics or to use quantitative survey data for theoretical sampling in qualitative research. Initiation designs aim at uncovering paradox or contradictions to initiate new insights. Finally, expansion designs extend the scope of a study by mixing methods. One typical example of this last approach is the combination of quantitative evaluation of programme outcomes with qualitative studies of programme implementation. We encourage submissions dealing with designs of mixed-methods studies combining survey research and qualitative research. Presentations should focus on methodological issues of research designs or analysis of such data.

Paper Details

1. Mixing methods for quality assessment and harmonisation of survey questions
Dr Vlasta Zucha (Statistics Austria)

Three major surveys conducted by Statistics Austria include questions on housing which refer to similar housing characteristics, but vary according to the legal frameworks, context, aim and interest of the particular survey. A mixed method design was used as basis for adaptation and harmonisation of survey questions: The existing survey data of the Microcensus, the Household Budget Survey and EU-SILC was analysed; different documents including legal regulations were examined; expert interviews and respondent debriefing have been carried out. The results of this approach are finally used for the adaptation of questions as well as for data analysis and reporting.

2. Methodological lessons from using alternative methods to measure social processes through surveys
Professor Wim Hardyns (Ghent University, Free University of Brussels, Antwerp University)
Professor Lieven Pauwels (Ghent University)

The role of community (dis)organizational processes is a major issue in contemporary criminology. As a consequence, researchers have been increasingly eager to measure community-level social mechanisms such as social cohesion and disorder. However community inhabitants are predominantly used to measure (dis)organizational processes, this approach requires large numbers of respondents to generate reliable and valid measures. In this article, the use of key informants, observation checklists and administrative data are discussed as alternative methods of measuring community processes. Our findings suggest that these methods can provide reliable and valid measures of specific dimensions of social cohesion and disorder.

3. Use of Norwegian Municipal Public Health Profiles and Data Bank: a mixed methods study
Ms Heidi Lyshol (Norwegian Institute of Public Health)

The Norwegian Institute of Public Health Public conducted a study of how Norway’s 428 municipalities used the Municipal Data Bank and the Municipal Public Health Profiles.
Semi-structured qualitative interviews were conducted; informants also filled in questionnaires and were observed while using the MDB.
Data from these sources were used in a mixed methods study.

The informants appreciated the MPHPs, while MDB results were ambiguous. The indicators were used for planning, in collaborations, and to evaluate interventions.

Further indicators and training are needed, and to involve municipal leaders in public health work and gain support for their public health