chapter 27
....................................................................................................................................................MODELING
HRM AND
PERFORMANCE
LINKAGES
....................................................................................................................................barry gerhart
27.1 Introduction
.........................................................................................................................................................................................
My focus here is on methodology in the literature on human resources
(HR) management and performance.^1 Whenever a theoretical model of HR and
performance is tested and estimated using empirical data, a binary decision
regarding whether the model is supported (yes or no) is typically made. If support
is found, it is either because the model is correct or the method is wrong (i.e. a false
positive, or Type I, error of inference). If support is not found, it is either because
the model is wrong or the method is wrong (i.e. a false negative, or Type II, error of
inference).
I thank Ingo Weller for his helpful comments on an earlier version.
(^1) ‘Performance’ often refers toWnancial or operating performance in the literature. The models
discussed here apply equally well to performance as deWned by stakeholders other than shareholders
(e.g. Brewster 1999 ; Paauwe 2004 ). Obviously, diVerent stakeholders’ (shareholders, employees,
customers, public) interests are not identical, but there is often a substantial shared interest in the
ongoing viability of the organization.