RESULTS
Table 2 provides descriptive statistics and correlations for all variables.
The correlation between the two dependent variables is .51. Consistent with
prior work, the relationship between the HRM practices and perceptual per-
formance measures is generally positive (11 of 14 correlations). The magni-
tude of the correlations is generally small to moderate, however, potentially
raising questions about the substantive importance of HRM practices. Associ-
ations among HRM practices also tend to be positive (19 of 21 correlations).
We focus on the coefficients on the HRM practice variables in our discus-
sion of the results. Inspection of Tables 3 and 4 shows, however, that many
of the control variables are significantly associated with the perceptual perfor-
mance variables. Table 3 (models 1-5) reports results of the regression equa-
tions for perceived organizational performance and Table 4 (models 6-10)
presents the results for perceived market performance. Each of the HRM
practice coefficients reported in the first column of Table 3 is from a separate
regression that contained the control variables and that one HRM practice.
The model 1 results indicate that five of the seven HRM practice coefficients
are positive and significant. Only the internal labor market and staffing selec-
tivity coefficients are insignificant at conventional levels. Models 2 and 3
report results obtained when the HRM practices are included simultaneously
in the same equation. Two equations are reported because the internal labor