The Dark Five - statistical anomaly or persistent phenomena: An exploratory investigation into the factor structure of the SD4 in multiple samples

Recent analysis of the dark tetrad has tentatively proposed that a more appropriate representation of dark traits, consisting of 22 items loading onto a 5-factor solution could resolve acknowledged methodological issues. Replication was needed to determine if this new model (D5-SD4) is a persistent phenomenon or a statistical artefact related to the original data. This study compared model fit and measurement invariance of original SD4 and D5-SD4 in 12 samples from 11 different countries ( N = 7045). The D5-SD4 exhibited better fit whether 28 items were used for the SD4, or only the shared 22. Measurement invariance and alignment optimization showed that D5-SD4 could be used to compare results between groups, which was not the case for the original SD4. The results demonstrate the reliability of the D5-SD4 structure in comparison to the SD4. Suggestions for the future improvements of the core dark traits model are given. • There have been persistent issues with measurement of the Dark Tetrad since inception. • Previous research identified a better fitting model for the SD4, called the Dark 5. • This paper explored the replicability of the five-factor model in 12 samples. • The Dark 5 demonstrated better fit than the SD4 across all samples. • Similar results were found for the shard 22 items and all 28 items.