AbstractThis paper reports two studies that utilized full structural equation modeling to explore the antecedents of proactivity at work. It was motivated by observations from a recent replication of a leading proactivity model, Parker et al. (J Appl Psychol 91(3):636–652, 2006. https://doi.org/10.1037/0021-9010.91.3.636) in which the replication model failed to confirm multiple previously reported variable relationships and manifested significant evidence of omitted variable bias (correlation between exogenous predictor variables and the disturbance term of the model’s dependent variable). Study 1 revisited the antecedents of proactivity in light of the unexpected replication model results. Utilizing the same diverse dataset (N = 521), it analyzed the relationships among all of the original model variables, together with a new contextual information availability variable, and constructed a new model (the “Modified Model”) consisting of the path relationships that were observed as statistically significant. The Modified Model differs from the original model in multiple respects: (i) it removes four of eight original variable path relationships, which could not be replicated as statistically significant, (ii) it removes a posited proactivity antecedent, flexible role orientation, whose path relationship with its predictor, co-worker trust, was observed as significant, but negative, (iii) it adds a direct path relationship between proactive personality and proactivity at work, (iv) it respecifies co-worker trust from an exogenous to an endogenous proactivity antecedent of proactivity at work, and (v) it specifies CI availability as new exogenous antecedent of co-worker trust. The Modified Model achieved good model fit and high explained variance (R2 = 0.64). In addition, the Modified Model did not manifest the evidence of omitted variable bias that was observed for the original model utilizing the same dataset. Study 2 was undertaken to cross-verify both of the prior studies using a fresh dataset (N = 479), and both prior studies were confirmed. Re-replication of the Parker et al. (J Appl Psychol 91(3):636–652, 2006. https://doi.org/10.1037/0021-9010.91.3.636) model again manifested evidence of omitted variable bias and provided evidence that corresponded with the results of the initial replication—significant differences from the observations reported by the original study. On the other hand, replication of the Modified Model again manifested no evidence of omitted variable bias, and resulted in observations that confirmed the Study1 observations used to construct the Modified Model. Taken together, Studies 1 and 2 provide evidence of the overall validity of the Modified Model of the antecedents of proactivity at work.
Read full abstract