Assessing Group-Level Constructs Under Missing Data Conditions: A Monte Carlo Simulation

Abstract

The authors reviewed recently published research on small groups and teams to understand how within-team nonresponse is reported and handled. They used Monte Carlo simulation to investigate how data-handling choices affect measurement reliability and hypothesis testing under conditions of random and systematic nonresponse. More complete reporting of nonresponse is recommended by the authors, and they propose guidelines for analyzing team-level constructs using data from teams without full response.