Do optimism and pessimism have different relationships with personality dimensions? A re-examination

► The current research re-examined Marshall et al.’s (1992) findings. ► Marshall et al. showed that optimism and pessimism have distinct correlations with personality dimensions. ► We found that their results are likely due to scale artefact (item valence). ► After item valence was controlled, we did not replicate Marshall et al.’s results. ► Suggestions for future research are discussed. The classic, well-cited study by Marshall et al. (1992) demonstrated that optimism correlates stronger with extraversion than does pessimism and pessimism correlates stronger with neuroticism than does optimism, and these results lent support to their claim that optimism and pessimism are two separate constructs. However, we argued that their results are likely the outcome of scale artefact caused by item valence (or item favorability). In an empirical study ( N = 1016), we evaluated the correlation of optimism scores and pessimism scores with the most common measure of optimism – Life Orientation Test-Revised (LOT-R). As expected, when item valence effect was not controlled, we replicated the finding by Marshall et al. (1992) that optimism and pessimism show differential correlations with extraversion and neuroticism. After item valence was controlled such pattern of relationships was greatly reduced. Suggestions for future research to resolve the dimensionality debate for optimism–pessimism are discussed.