Dynamics in the Development of Organizational Stress

点击翻译按钮获取中文摘要

The project is intended to succeed an already ongoing project, and it is supposed to be part of a series of projects that will continue this stream of research. The already ongoing DFG-project (DO 763/9-1) deals with the meta-analysis of panel studies of organizational stress. In general, the series of projects will continue developing the methodological foundations of meta-analyses of panel studies as well as providing successively more detailed meta-analyses of primary studies. The present proposal has four major aims. First, it should enable us to continue the meta-analysis of existing panels studies on organizational stress. The hired doctoral student (of the previously granted 2 year project) should finalize her doctoral thesis and then continue pub-lishing results as a post doc.Second, subsequent meta-analyses should be extended to include data obtained in diary studies. This enables us investigating the impact of time intervals between measurement occa-sion on effect sizes of organizational stressors across a much wider range of time intervals than those available so far. Among other things, this will yield evidence of optimal time inter-vals, which would be important for the design of future panels studies.Third, the proposal should enable us conducting repeated measurement studies, which could be carried out by doctoral students of an already granted graduate school. In these stud-ies we will investigate if studies using different time intervals could serve a substitutes for each other. This addresses, for example, the question if results of a study using a 1-week in-terval reflect the same processes as studies using a 1-year interval. From a substantive point of view we will focus on self-esteem threating stressors such as the assignment of illegitimate tasks.Forth, we already developed a new type of meta-analysis that explicitly accounts for dif-ferent time intervals of primary studies (CoTiMA). This should be completed and pro-grammed in a user-friendly fashion. The mathematical foundations are well-know for long, but the translation into statistical programs that allow for testing structural equation models has just been done recently and is only possible with using a particular package (OpenMx in R). Our first CoTiMA, for instance, still required a couple of thousand lines of code. By means of programming a user-friendlier interface we intend to make CoTiMA usable for many researchers and facilitate spreading it in the community.