The Emotion Regulation Questionnaire: Psychometric Properties in General Community Samples

Abstract

The Emotion Regulation Questionnaire (ERQ) is a 10-item self-report measure of 2 emotion regulation strategies, cognitive reappraisal and expressive suppression. It is a widely used measure of emotion regulation, but its factor structure has rarely been examined outside of university student samples, and some authors have recently questioned its factorial validity in general community samples. In this study, we examine the psychometric properties of the ERQ (original English version) in 3 Australian general community samples (N = 300, 400, 348). Confirmatory factor analyses in each sample demonstrated that the traditional 2-factor model (comprised of cognitive reappraisal and expressive suppression factors) was replicable and an excellent fit to the data. In all samples, ERQ cognitive reappraisal (α = .89–.90) and expressive suppression (α = .76–.80) scores had acceptable to excellent levels of internal consistency reliability. As expected, cognitive reappraisal scores were significantly negatively correlated with psychological distress and alexithymia, whereas expressive suppression scores were significantly positively correlated with psychological distress and alexithymia. We conclude that, similar to previous findings in student samples, the ERQ has strong psychometric properties in general community samples and can therefore be used confidently regardless of participants’ student status.

Acknowledgments

Ethics approval for this project was granted by the Edith Cowan University Human Research Ethics Committee. The guidelines of this committee were followed. All participants provided informed consent for their data to be used. The data and materials used in this study can be made available to researchers on request by contacting David Preece ([email protected]). The analysis plan used in this study was not preregistered with an independent institutional registry.

Notes

1 In a factor model, an item error term contains the variance in the item score that is not accounted for by the specified substantiative latent factors. A covariance between two item error terms therefore suggests that there is something similar about these items other than them both being markers of their specified latent factor (Gerbing & Anderson, 1984).

2 In all three samples, some additional participants also completed the online survey. However, their data were excluded during quality screening because they failed an attention check question (which asked them to select a specific point on the Likert scale) or completed the survey impossibly quickly (at a rate of < 2 s per question, suggesting inattentive responding). Across Samples A, B, and C, data from 65 participants were excluded.

3 Qualtrics panels recruit from multiple sources, primarily actively managed market research panels (see Qualtrics, 2014). Participants are sent an e-mail inviting them to complete the online survey.

4 We did not collect data on the ethnicity or cultural values of our participants, however, our collected birthplace data were similar to recent Australian census data (Australian Bureau of Statistics, 2017a), whereby most participants reported being born in Australia or the United Kingdom.