Hypothesis 2. Our findings did not provide support for Hypothesis 3, because the indirect effect of turnover
intention on the relationships between LMX and turnover behavior at Time 2 and Time 3 was not significant. The results show,
however, that the indirect effect of turnover intention on the links between AC and turnover behavior at Time 2 (β = −.30,
p b .001), and between AC and turnover behavior at Time 3 (β = −.22, p b .001), was significant. Bootstrapping results also
indicate that a 99% bias-corrected confidence interval did not contain zero (−.51 b − N −.08) for turnover behavior at Time
2 (−.39 b − N −.05), or for turnover behavior at Time 3, thus supporting Hypothesis 4.
5. Discussion
Previous