Ridderprint

Table 6.3. Cox proportional hazard models of voluntary turnover at Shell. Shell M1 M2.1 M2.2 M2.3 M3 a B se p B se p B se p B se p B se p Gender (male) 0.06 0.08 .469 0.10 0.08 .228 0.06 0.08 .513 0.07 0.08 .382 0.10 0.08 .240 Age at hiring b -0.03 0.02 .055 -0.05 ** 0.02 .004 -0.03 0.02 .052 -0.04 * 0.02 .034 -0.05 ** 0.02 .004 Performance rating t -3.78 *** 0.27 .000 -3.73 *** 0.28 .000 Potential rating t -0.22 ** 0.07 .003 -0.02 0.07 .817 Potential unassessed t 1.49 *** 0.16 .000 1.44 *** 0.16 .000 Mobility status tc STIA -1.51 *** 0.41 .000 -1.33 ** 0.41 .001 Repatriate 0.67 * 0.32 .037 0.83 ** 0.32 .010 STIA experience t -0.04 0.03 .101 -0.05 0.03 .057 Log likelihood -5248.37 .000 -5159.67 .000 -5199.68 .000 -5233.15 .000 -5104.07 .000 Parameters 38 39 40 41 44 D (χ 2 ) 177.38 *** .000 97.38 *** .000 30.43 *** .000 111.20 *** .000 Note : Other covariates included, but not printed: hiring year, organizational division, base country, skill pool, and work level. a M3 is compared statistically to M2.1. t time-varying. b continuous, in years. c referent = local and no international experience. * p < .05 ** p < .01 *** p < .001