On the Evaluation of Causal Models.
- Specht, David A
- Social Science Research
In this paper we review neglected issues of simultaneous statistical inference and statistical power in survey research applications of the general linear model, and we find that classical hypothesis testing as it is currently applied, is inadequate for the purposes of social research. The intelligent use of statistical inference demands control over the overall level of Type I error and knowledge of the magnitude of effects one is likely to detect. We suggest techniques that can be used to routinely incGrporate considerations of simultaneous inference and power into the statistical analysis of survey data. Several examples of applications of these techniques are presented.