Project Details
Projekt
SPP 1292: Survey Methodology
Subject Area
Social and Behavioural Sciences
Term
from 2008 to 2011
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 28074213
Scientific surveys represent an indispensable instrument of social research. At the same time, they have a significant social influence if their results contribute to public opinion formation, to decision making in politics and jurisdiction as well as to administrative acts, if they lead into scientific diagnoses, causal analyses and prognoses or if they lead into economic decisions. Therefore, it is of essential importance, that no wrong, distorted, vague or misleading conclusions are drawn from the results of scientific surveys.
In practice, however, the required survey quality cannot be attested unquestionably, either because the quality is not always achieved or because there is simply no attempt to judge it. In view of all the potential sources of error surveys may suffer from, this is a remarkable situation. The list of potential sources of error is deadly long; just to name some examples: responses, which may be affected by the survey mode, by question wordings, by response scale formats and required memory tasks; or responses, which are simply affected by the motive to avoid socially undesirable responses. Samples biased by self-selection processes and hence sample estimates, which can only fail to hit their targets, represent another example of the possibility that survey results may draw false pictures of reality.
This situation will become the more precarious the more social change and technological progress challenge the present way of conducting survey research: Firstly, the bases of survey cooperation change under the influence of social change. Secondly, technological progress produces options, which are used increasingly in practice of social research, however without needed basic research having taken place beforehand. A prime example is the widespread implementation of "Online Access Panels" by which research institutes react to the technical options of the Internet. Further challenges add, for instance, the trend to be accessible only by mobile phones, that is to say by completely renouncing ground-based telephones and thus the standard access for telephone surveys.
The Priority Programme aims correspondingly at advancing the survey design in such a way that it masters better than before relevant sources of error while meeting the demands of social change and technological progress.
In practice, however, the required survey quality cannot be attested unquestionably, either because the quality is not always achieved or because there is simply no attempt to judge it. In view of all the potential sources of error surveys may suffer from, this is a remarkable situation. The list of potential sources of error is deadly long; just to name some examples: responses, which may be affected by the survey mode, by question wordings, by response scale formats and required memory tasks; or responses, which are simply affected by the motive to avoid socially undesirable responses. Samples biased by self-selection processes and hence sample estimates, which can only fail to hit their targets, represent another example of the possibility that survey results may draw false pictures of reality.
This situation will become the more precarious the more social change and technological progress challenge the present way of conducting survey research: Firstly, the bases of survey cooperation change under the influence of social change. Secondly, technological progress produces options, which are used increasingly in practice of social research, however without needed basic research having taken place beforehand. A prime example is the widespread implementation of "Online Access Panels" by which research institutes react to the technical options of the Internet. Further challenges add, for instance, the trend to be accessible only by mobile phones, that is to say by completely renouncing ground-based telephones and thus the standard access for telephone surveys.
The Priority Programme aims correspondingly at advancing the survey design in such a way that it masters better than before relevant sources of error while meeting the demands of social change and technological progress.
DFG Programme
Priority Programmes
International Connection
Netherlands, Switzerland, USA
Projects
- Access Panel und Mixed-Mode Internet Survey (Applicant Engel, Uwe )
- Asking sensitive questions: possibilities and limits of randomized response and other techniques in different survey modes (Applicants Diekmann, Andreas ; Preisendörfer, Peter ; Voss, Thomas )
- Coordination project (Applicant Engel, Uwe )
- Errors in autobiographical memory and their effects in time-to-event analysis (Applicants Pigeot, Iris ; Schnell, Rainer )
- Experimental Mobile Phone Panel (Applicant Fuchs, Marek )
- Identification of Falsifications in Survey Data (Applicants Menold, Natalja ; Winker, Peter )
- Improving intercultural comparative research by the application of cognitive techniques using the Internet in several countries (Applicant Braun, Michael )
- Increasing nonresponse error by reducing nonresponse rates? Investigating the biasing effect of methods and procedures aimed at increasing response rates in Web-based access panel surveys (Applicant Bandilla, Wolfgang )
- Modelling response propensities in access panel based surveys (Applicants Münnich, Ralf ; Rendtel, Ulrich )
- Optimization of sampling design and inference for establishment surveys (Applicant Rässler, Susanne )
- Respondent Incentives, Interviewer Training, and Survey Participation (Applicant Börsch-Supan, Ph.D., Axel )
- Robust and efficient multiple imputation of complex data sets (Applicants Reinecke, Jost ; Spieß, Martin )
- Stability of Self-Generated Panel Codes (Applicant Schnell, Rainer )
- Survey Nonresponse Analysis based on Social Security Administration Records (Applicant Schnell, Rainer )
- The effects of incentives on data quality in online panels (Applicant Göritz, Anja )
- The factorial survey as a method for measuring attitudes in population surveys (Applicants Hinz, Thomas ; Liebig, Stefan )
- Use of population data bases with double sampling for choice-based sample surveys and time-to-event data (Applicants Pigeot, Iris ; Schnell, Rainer )
Spokesperson
Professor Dr. Uwe Engel
