Statistical considerations in the interpretation of research on occupational stress management interventions

Considerable research effort has been conducted into evaluating occupationa l stress management interventions (SMIs) that target individual workers. We consider several factors central to the interpretation of such research. S pecifically, we survey 27 published SMI reports taking into account (1) the amount and detail of descriptive information provided in respect to sample characteristics and intervention procedures; (2) the level of statistical power evident in study designs; (3) whether studies are reductionist, in th at they target specific workers experiencing elevated levels of strain, or preventive in that they are available to allcomers, regardless of initial s train levels; (4) whether statistically significant change where it occurs is meaningful in real-world (or clinical) terms, and reliable in measuremen t terms. We also outline recommendations for descriptive data and statistic al considerations to be included in planning and interpreting SMI research. Our conclusion is that stress management researchers should attend to thos e factors to a far greater extent. At present the quality of reporting and research design is such that it is difficult to form an impression of what type of SMI is appropriate to whom, and in what circumstances.