SMABS 2004 Jena University
SMABS 2004 Home Organization About Jena Sponsors Links Imprint / Contact SMABS Home

European Association of Methodology

Department of methodology and evaluation research

Jena University

Contributions: Abstract

Evaluating an alternative teaching method with missing covariates

Susanne Rässler Oleg Müller
University of Erlangen-Nürnberg

In the context of the research project "Neue Statistik", which is funded by the Federal Ministry of Education and Research with about 2 Mio Euro, a complete multimedia course is created for the standard lectures and exercises in statistics as they are typically taught in Germany in the undergraduate courses in Business administration and other studies. The project is done in cooperation of 14 institutes at 10 German universities under the coordination of CeDiS (Center for Digital Systems) at the Free University of Berlin. All areas of learning and teaching shall finally be covered by "Neue Statistik", for more information see and

Our task is to find out whether this new teaching method shows positive impact on the students' grades. Therefore, in a pseudo experiment, students were asked to voluntarily take a special course being taught the alternative way. The lecture was accompanied by interactive computer exercises, and finally a small group of students did their examination directly at the PC. Since the selection process was mainly influenced by the students themselves, we assume a selection bias. Thus, control groups are created using propensity score matching. However, a lot of information and covariates to be used for the matching procedure were derived from two questionnaires filled out by the students and prone to item nonresponse. Basing inference only on the complete cases discards more than 60% of the data records. Therefore, a multiple imputation algorithm (data augmentation with NORM) is applied before any further analysis is performed and the results are compared with the results derived from the complete cases only.