Research DEFECT I - Design effects in complex surveys

Goal of the project was the empirical estimation of factual standard errors in a complex survey of the general population in Germany. We tried to separate the effects of  inverviewers, field organizations [house effects], sampling points [PSUs] and data collection mode by an interpenetrated sample design of five national surveys with 160 PSUs in a fear of crime survey. More details on the design can be found in Schnell/Kreuter (2002, in German) and Schnell/Kreuter (2005) in English.


Presentations and publications

  • Frauke Kreuter (1999): Messung von Kriminalitätsfurcht; 35. kriminologisches Kolloquium der südwestdeutschen und schweizerischen kriminologischen Institute, Konstanz, 3.7.1999.
  • Rainer Schnell/Frauke Kreuter (2000): Empirical Estimation of Design Effects in Complex Samples. Paper presented at the Fifth International Conference on Logic and Methodology. Cologne, Germany, October 3-6.
  • Rainer Schnell/Frauke Kreuter (2000): Unit-Nonresponse as a function of interviewer calling strategies. Paper presented at the Fifth International Conference on Logic and Methodology. Cologne, Germany, October 3-6.
  • Elisabeth Coutts/Rainer Schnell (2000):  "Not very likely… I’d say 70%": Learning effects in estimation of subjective probabilities.  Paper presented at the Fifth International Conference on Logic and Methodology. Cologne, Germany, October 3-6.
  • Frauke Kreuter (2000): "Uncertainty in capturing uncertainty: Toward measurement of fear of crime. Part I: Measuring subjective probability"; Annual Conference of the International Society of Quality of Life Studies; Girona/Spain 21.7.2000.
  • Rainer Schnell/Frauke Kreuter (2000): Das DEFECT-Projekt: Sampling-Errors und Nonsampling-Errors in komplexen Bevölkerungsstichproben; in: ZUMA-Nachrichten, 47, S. 89-101 (download pdf). 
  • Rainer Schnell (2001): Vergleich einer bundesweiten Quotenstichprobe mit bundesweiten Zufallsstichproben anhand interpenetrierender Stichproben; Sektion Methoden der DGS: Frühjahrstagung in Konstanz, 30./31.3.2001.
  • Frauke Kreuter (2001): Trennung von Interviewer-Effekten und Sampling-Point-Effekten mit Hilfe von interpenetrierenden Stichproben am Beispiel von "fear of crime"-Indikatoren; Sektion Methoden der DGS: Frühjahrstagung in Konstanz, 30./31.3.2001.
  • Rainer Schnell/Frauke Kreuter (2001): Neue Software-Werkzeuge zur Dokumentation der Fragebogenentwicklung (download pdf).
  • Frauke Kreuter (2002): Kriminalitätsfurcht: Messung und methodische Probleme,Opladen, Leske + Budrich.
  • Rainer Schnell/Frauke Kreuter (2002): Separating Interviewer- and Sampling-Point-Effects; in: American Statistical Association/American Association of Public Opinion Research (eds.): Proceedings of the Joint Conference, 2002, Florida, pp. 3132-3133.
  • Rainer Schnell/Frauke Kreuter: (2005) Separating Interviewer and Sampling Point Effects: Journal of Official Statistics, Vol. 21, No. 3, S. 389-410.

Team
Principal investigator: Rainer Schnell
Project coordination: Frauke Kreuter
Project staff: Elisabeth Coutts/Daniela Thume

Details
CATI-Fieldwork
Academic Data, Essen

F2F-Fieldwork
Förster&Thelen, Bochum
Infas, Bonn
IPSOS, Mölln

Funding
German Research Foundation (DFG)

Data
Available on request to the principal investigator

Questionnaire
The DEFECT questionnaire is availabe via the download page.