Presenteeism is problematic since it relates to lower health and productivity. Prior research examined many work and attitudinal variables relating to presenteeism at the individual level. Here, we conceptualize presenteeism as multilevel phenomenon also shaped by the overall attendance behavior (absenteeism and presenteeism) at the work unit. We surveyed employees at a manufacturing plant on presenteeism, health-related lost productive time (HLPT) and absenteeism (N = 911, 22 units) and collected preceding (past 12–7 and 6 months) objective absence data aggregating it at unit level. Considering the individual-level antecedents only higher physical demands predicted higher absence duration. Presenteeism related positively to physical demands, a burdensome social environment, and organizational identification and negatively to ease of replacement, and core self-evaluations. These relationships were similar for HLPT as outcome. Regarding unit-level factors, preceding unit-level absence frequency (but not duration) negatively related to presenteeism. The negative relationship between core self-evaluations and individual presenteeism decreased under a stronger presenteeism context supporting the hypothesized cross-level effect of unit-level presenteeism context strength. Moreover, individual and unit-level presenteeism correlated, as expected, more strongly with health complaints than absenteeism. Our study demonstrates the value of a contextual, multilevel approach for understanding antecedents and consequences of attendance behavior.

The present study provides evidence for the important role of job crafting and self-undermining behaviors at work, two new concepts that were recently integrated into the well-known job demands-resources (JD-R) theory (Bakker and Demerouti, 2017). We investigate how these behaviors are associated with work engagement, emotional exhaustion, and work ability as a long-term indicator of employee’s well-being. Furthermore, we examine the moderating role of personal resources in the stress-strain process by comparing groups of employees representing the five types of job satisfaction defined by Bruggemann (1974). Data was collected in a cross-sectional study within a German DAX company’s manufacturing plant from 1145 blue- and white-collar workers. Results of structural equation modeling provided, as expected, support for an indirect effect of job demands and job resources on emotional exhaustion and work engagement through job crafting and self-undermining. Work ability, on the other hand, was mainly affected by emotional exhaustion, but not by work engagement. Most important, we found significant differences between path coefficients across the five types of job satisfaction indicating that these types represent important constellations of personal resources and job demands that should be considered both for analyzing stress at work and for offering tailored stress interventions in organizations.