http://hdl.handle.net/10900/53294 2026-05-12T20:51:23Z http://hdl.handle.net/10900/47296 The Impact of multiple imputation for DACSEIS Rässler, Susanne This paper is designed to provide an extensive introduction to the principles of multiple imputation and to give some general recommendations of using multiple imputation techniques in the DACSEIS universes. The definition of an ignorable missingness mechanism is explained, and the concept of the observed-data likelihood is discussed. To introduce the multiple imputation principle a short introduction of Bayesian statistics is provided. A small simulation study is performed comparing different approaches to illuminate the advantages and disadvantages of different imputation techniques. Finally, an overview about recently available multiple imputation software is given and violations of the assumptions made are addressed. 2004-01-01T00:00:00Z http://hdl.handle.net/10900/47281 On the simulation of complex universes in the case of applying the german microcensus Münnich, Ralf; Schürle, Josef The aim of the DACSEIS project is to deliver recommendations on the use of variance estimators under complex survey designs in the presence of non-response. Since mathematical comparisons on the efficiency of variance estimation methods in this field are generally unavailable or lead to irrelevant results, adequate simulation studies have to be carried out that are based on realistic data sets. To be able to carry out a simulation study in the frame of complex designs one has to draw samples from a universe respecting for the true sampling design. However, in many cases, no data or only outdated data are available for the universe which leads to the need of adequately generating a micro data set from the sample. Within this paper, a procedure of generating the universe for the German microcensus, which is a 1 sample of the population living in Germany, will be presented. The procedure allows for an adequate consideration of the individual information on a limited data set and can therefore be used as a basis for the simulations on variance estimation methods on the German microcensus data. 2003-01-01T00:00:00Z http://hdl.handle.net/10900/47246 A comparison of the five Labour Force Surveys of the DACSEIS project from a sampling theory point of view Quatember, Andreas Labour market data are important for the assessment of the working of the national social and economic policies and as an indicator for social trouble spots. The European Union therefore pays very much attention in the harmonisation of the national Labour Force Surveys to be able to have comparable data of high quality. In this paper these surveys, that are included in the DACSEIS project (IST-2000-26057) are compared from a sampling theory point of view to show the similiarities as well as the differences of these surveys from this aspect. 2002-01-01T00:00:00Z http://hdl.handle.net/10900/47236 A method of weighting adjustment for survey data subject to nonignorable nonresponse Zhang, Li-Chun Weighting adjustment is a standard quasi-randomization approach for survey data subject to nonresponse Little (1986). The existing methods are typically based on the assumption that nonresponse is independent of the survey variable conditional to the auxiliary variables used to form the adjustment cells. In this paper we consider nonignorable nonresponse which is independent of certain auxiliary information conditional to the variable of interest. We estimate the size of the sample adjustment cells using a method of moment conditional to the sample. The method relies on only the nonresponse mechanism, and is independent of the sample design. In variance estimation, we evaluate the nonresponse effect on estimation and design, analogously to the concept of design effect. By comparing the nonresponse effects under a nonignorable model against those under an ignorable one, we obtain a means of measuring the effect of nonignorability. We motivate and illustrate our approach for estimation of household composition. 2002-01-01T00:00:00Z