Purpose: The purpose of these two studies was to explore the relationship between video monitoring and quantity of performance in the absence of demand characteristics.

Design/methodology/approach: Data were gathered via two experiments involving business students working on a motor task. Participants were randomly assigned to the monitored and unmonitored conditions. Experiment 1 (n = 75) was inductive while Experiment 2 (n = 139) was partially inductive.

Findings: Experiment 1 showed that monitored participants’ performance was lower than that of unmonitored participants. Further, monitoring reduced outliers, increased interquartile variance, and normalized the distribution. Experiment 2 replicated the effect of monitoring on performance controlling for cognitive ability and emotions, demonstrated that negative emotions interacted with monitoring condition, and suggested that differences in performance were not due to cheating or variation in task-related strategies. We offer a grounded theory of video monitoring proposing that different implicit decision rules are activated when people are monitored as compared to when they are not monitored.

Implications: Future research needs to determine the extent to which our results extend to similar settings in the workplace and to other forms of observation. At this time, we believe organizations should carefully consider the consequences of electronic monitoring. Controlling expectations in the lab or workplace does not necessarily eliminate the independent effect of monitoring. Therefore, researchers must beware misinterpretation of effect sizes and overlooking the role of observation in their data.

Originality/value: These studies demonstrate that video monitoring can create observer effects in the absence of demand characteristics. Our inductive approach revealed the nature of the effects beyond mean differences and served as the basis for developing a testable theory of monitoring that goes beyond what was previously known.