“…First, the self-report data may be susceptible to common method variance (Lindell & Whitney,200 1 ). However, we used the split-sample technique to minimize common source bias (Ostroff et al , 2002 ;Podsakoff et al, 2012;Rousseau, 1985;Wilderom et al, 2012), with one dataset being used to aggregate group-level measures, and a separate dataset used to calculate the individual-level variables and examine the hypotheses. Moreover, given the study's primary focus on inte raction effects, concerns over method are attenuated as it is un likely that the moderator effects could be produced by CMV (Lai et al, 2013 ).…”