Suchergebnisse

Predictive modeling methods from the field of machine learning have become a popular tool across various disciplines for exploring and analyzing diverse data. These methods often do not require specific prior knowledge about the functional form of the relationship under study and are able to adapt to complex non-linear and non-additive interrelations between the outcome and its predictors while focusing specifically on prediction performance. This modeling perspective is beginning to be adopted by survey researchers in order to adjust or improve various aspects of data collection and/or survey management. To facilitate this strand of research, this paper (1) provides an introduction to prominent tree-based machine learning methods, (2) reviews and discusses previous and (potential) prospective applications of tree-based supervised learning in survey research, and (3) exemplifies the usage of these techniques in the context of modeling and predicting nonresponse in panel surveys.

Abstract
Reinterview designs are a potential tool to estimate and adjust for mode measurement effects, that is, relative differences in mode-specific measurement error bias. In 2011, a reinterview design was successfully applied to the Dutch Crime Victimization Survey, which led to a redesign of the survey. Reinterview designs may, however, be very costly, especially when face to face is included as a survey mode. The crucial question is whether benefits outweigh costs, that is, whether the potential increase in the accuracy of survey statistics is worth the investment. The answer to this question depends heavily on the purpose of the reinterview, that is, assessment versus adjustment, the size of the measurement effects, and the relative cost of the modes. Reinterview designs also make a number of assumptions that will not hold for every setting. In this article, we perform a cost–benefit analysis for two surveys, the Dutch Health Survey and the Dutch Labour Force Survey, and discuss the utility and validity of reinterviews. We conclude that a reinterview may not be useful due to relatively small measurement differences for the Labour Force Survey, whereas it may be useful for the Health Survey.

Wissenschaftliche Umfragen können keine aussagekräftigen Ergebnisse liefern, wenn ihre Datenqualität durch fehlende oder verfälschte Antworten beeinträchtigt wird. Eine Herausforderung der Sozialforschung besteht darin, solche Fehlerquellen zu erkennen und zu kontrollieren. Der Band präsentiert Erkenntnisse und Methoden zur Behandlung von Unit Nonresponse, Missing Data und verschiedene Arten von Messfehlern im Kontext von Web und Mixed-Mode Panel, Mobile Web und Faceto-Face-Befragungen.