Studying Human Resource Strategy: An Approach and an Agenda

Summary Clearly, HRS research is in its infancy. We have only begun to make progress toward a better understanding of OHRS content, while FHRS content remains uncharted. Some research has been done in organizations with formal strategic human resource planning processes (e.g., Rush, 1983; Dyer, 1984), but much remains to be done here, as well. And the more informal approaches to organizational and functional strategic decision making represent virgin territories. All of these areas cannot be tackled in a single stream of research, but because strategy research is difficult, time consuming, and expensive, it is probably wise to combine areas and issues where possible. For example, a set of studies might be designed to describe and develop measures of both the content of OHRS and the process by which the content was formulated. Or, as we have done, research can focus on description and measurement, on the one hand, and on the other, either determinants or effects. Narrowly defined studies should be avoided at this stage, as should ill-defined ones. Hence, the need for theory building (Schendel and Hofer, 1979). The immediate tasks are to explore the designated territories, develop meaningful concepts and constructs, and to generate working propositions, later to be followed by hypotheses (Duncan, 1979). It is descriptive theory that is needed now; formulating prescriptive theory aimed at strategic decision makers must await hypothesis development and testing. Descriptive theory demands descriptive research. For the most part, this means intensive case studies employing either longitudinal or retrospective designs that can capture the dynamic nature of strategy and strategy-making processes. It may also mean action research to gain access to sensitive issues and intricate relationships (Bower, 1982). Surveys have a place, too; they may, however, be better viewed as supplemental to more intensive research than (as is more typical) the other way around. Large sample sizes (except in some of the survey work) are unlikely. Thus, sample organizations should be selected on the basis of their characteristics (variables of interest) and not randomly (Harrigan, 1983). For example, researchers should seek out organizations known to be experiencing change in OSs and those known to have similar OSs but quite different internal and external environments. To assess effects, organizations representing a broad range of performance outcomes (e.g., managing human resources, meeting OS goals, and making money) will have to be located. Inevitably, the relevance vs. rigor conundrum will arise. To the extent that prescriptive theory is the final goal, relevance must receive the nod. This means an emphasis on external rather than internal validity and on theory and research that focuses on those outcomes that practitioners want to influence and on variables that they can either measure (and thus adapt to) or manipulate and control (Thomas and Tyson, 1982). This amounts to a charge to do relevant research which is as much as possible rigorously designed rather than, as so often has been the case in personnel (Dyer, 1980), the other way around.