Insufficient Effort Survey Responding: An Under‐Appreciated Problem in Work and Organisational Health Psychology Research

Abstract

Insufficient effort responding (IER) is problematic in that it can add a systematic source of variance for variables with average responses that depart from the scale midpoints. We present a rationale for why IER is of particular importance to Work and Organisational Health Psychology (WOHP) researchers. We also demonstrate its biasing effects using several variables of interest to WOHP researchers (perceived work ability, negative affectivity, perceived disability, work–safety tension, accident/injury frequencies, and experienced and instigated incivility) in two datasets. As expected, IER was significantly correlated with the focal study variables. We also found some evidence that hypothesised bivariate correlations between these variables were inflated when IER respondents were included. Corroborating IER's potential confounding role, we further found significant declines in the magnitude of the hypothesised bivariate correlations after partialling out IER. In addition, we found evidence for biasing (under‐estimation) effects for predictors not contaminated by IER in multiple regression models where some predictors and the outcome were both contaminated by IER. We call for WOHP researchers to routinely discourage IER from occurring in their surveys, screen for IER prior to analyzing survey data, and establish a standard practice for handling IER cases.