When testing the study’s hypotheses, we conducted separate regression analyses for the
self-reported loneliness predictor variable and the co-worker rated loneliness predictor variable.
Employees’ education level, age, organizational tenure, and the number of months supervised by
their current supervisor were included as controls in the analyses. We also controlled for the
following four employee personality factors: extraversion, agreeableness, trait positive
affectivity, and trait negative affectivity in the event that they are antecedents to loneliness, as
well as to confirm that loneliness differs from these personality traits, and predicts performance
outcomes above and beyond these traits. The results of the regression analyses we have
conducted supported our hypothesized model where greater employee loneliness led to poorer
task, team role and relational performance as mediated by lowered affective commitment and to
a lesser extent increased surface acting.