This course focuses on the testing of causal hypotheses in a cross cultural context. First the basic reasoning and the connected statistical analyses will be discussed for experimental, quasi experimental and non-experimental designs. After that the course will concentrate on non-experimental research. In that part the transformation of verbal theories in testable propositions will be given a lot of attention as an essential part of non-experimental research. After that estimation, testing and correction of causal hypotheses will be discussed.

After this general introduction of the SEM approach we will introduce a general model for structural equations which is a combination of an econometric simultaneous equation model and a factor analysis model. This general model allows the specification, estimation and testing of all known linear structural equation models used.   After that different special cases of this general model will be discussed and illustrated. Models for cross-sectional data, longitudinal data, latent growth, multiple groups and multilevel data will be discussed.

In large scale cross-national survey research testing of causal hypotheses is confronted with two major issues. The first is that one can only compare results across cultures if the measurement instruments across the countries are comparable. This problem will get a lot of attention because it is essential in cross national research. The second issue is how one can evaluate the equality or differences in effects across countries given equality of measurement instruments. This can be done in different ways: with interaction terms in pooled data, with equality constraints in multiple group analysis or with the use of random effects or multilevel models. All three possibilities will be discussed and demonstrated in this course.

The data from the European Social Survey, going on in more than 20 countries, provides sufficient information to illustrate the possibilities and problems of causal analysis in cross cultural context. So this data set will be used to illustrate how one can cope in a proper way with these issues.

The course will be a mixture of lectures and exercises where the ESS data will be used. The idea is that participants work in groups on a substantive problem and doing so get familiar with the approaches presented in this course. The substantive topics from the ESS which will be used in the exercises will be social capital theory and theories about migration in Europe .

In the conference after the course the students are supposed to present the results of their analyses while experts will be invited to introduce alternative procedures for SEM approach to tackle the same or similar problems. In this way the students will get a quite complete picture of the different possibilities although the emphasis will be on the SEM approach during the course.

*

One of the papers on Social Capital from the website. More is better.