Measurement error in research on the human resources and firm performance relationship: Further evidence and analysis

Our earlier article in Personnel Psychology demonstrated how generalizabili ty theory could be used to obtain improved reliability estimates in the hum an resource (HR) and firm performance literature and that correcting for un reliability using these estimates had important implications for the magnit ude of the HR and firm performance relationship. In their comment, Huselid and Becker both raise criticisms specific to our study and broad issues for the field to consider. In our present article, we argue, using empirical e vidence whenever possible, that the issues and criticisms raised by Huselid and Becker do not change our original conclusions. We also provide new evi dence on how the reliability of HR-related measures may differ at different levels of analysis. Finally, we build on Huselid and Becker's helpful disc ussion of broad research design and strategy issues in the HR and firm perf ormance literature in an effort to help researchers make better informed ch oices regarding their own research designs and strategies in the area.