A meta-analysis of correlations between depression and first person singular pronoun use

•Depression is linked to the use of first person singular pronouns (r = 0.13).•This effect is not moderated by demographic factors, such as gender.•The effect is nominally larger among unpublished studies.•There is little to no evidence of publication bias in this literature.

Depression is a burden. We discuss how theories, identification, assessment, and treatment of depression are at least partially tied to the correlation between first person singular pronoun use and individual differences in depression. We conducted a meta-analysis (k = 21, N = 3758) of these correlations, including numerous unpublished correlations from the file drawer. Our fixed effects analysis revealed a small correlation (r = 0.13, 95% CI = [0.10-0.16]) by modern standards. The correlation was not moderated by gender, nor by whether the effect had been published. These results more firmly establish first person singular pronoun use as a linguistic marker of depression-a marker that appears to be useful across demographic lines.