In a laboratory study, the presence of individual- or work-group-level electronic performance monitoring (EPM) was manipulated as participants worked on a data-entry task alone, as a member of a noninteracting aggregate, or as a member of a cohesive group. The pattern of results suggested the operation of a social facilitation effect, as highly skilled monitored participants keyed more entries than highly skilled nonmonitored participants. The opposite pattern was detected among low-skilled participants. No signs of social loafing were detected among group-monitored participants. Nonmonitored workers and members of cohesive groups felt the least stressed. The implications of these findings for organizations adopting EPM systems are discussed.