The Question

Research consistently links personality to leadership, but most of that research has worked at the level of broad traits: Extraversion, Conscientiousness, Openness, and so on. Baker and colleagues (2026) argue this approach has been leaving substantial signal on the table. By examining personality at the facet level, the more specific components that sit beneath each broad trait, and by attending to the context in which leaders operate, this meta-analysis provides the most comprehensive and nuanced picture of personality and leadership to date.

The Study

The meta-analysis drew on 203 datasets and 4,815 correlations, examining 21 personality facets across the Big Five framework in relation to two distinct leadership outcomes: leader effectiveness, how well leaders actually perform, and leader emergence, whether individuals come to be perceived and selected as leaders in the first place (Baker et al., 2026). The distinction between these two outcomes turns out to matter considerably.

Previous meta-analyses in this area have typically examined only a subset of facets, aggregated student and working adult samples together, and included self-reported leadership outcomes that inflate apparent validity through common method variance. Baker and colleagues excluded self-reported effectiveness measures and examined student samples separately as a moderator, both of which produce more conservative and credible estimates.

What Predicts Effectiveness

Among the strongest predictors of leader effectiveness were achievement motivation (ρ = .23), order (ρ = .23), and self-discipline (ρ = .23), all facets of Conscientiousness, along with optimism (ρ = .19), a facet of Emotional Stability (Baker et al., 2026). The finding on optimism is particularly noteworthy: it had not been examined in any of the nine previous meta-analyses on this topic, yet it emerged as one of the strongest individual predictors, especially for higher level leaders. Its association with goal persistence and positive framing of setbacks offers a plausible mechanism.

By contrast, some facets showed negligible or negative relationships with effectiveness despite belonging to traits commonly assumed to matter. Ingenuity, a facet of Openness, showed a correlation of essentially zero (ρ = -.01), and curiosity was negatively associated (ρ = -.06). Within Conscientiousness, virtue and rule following showed minimal relationships with effectiveness, in stark contrast to achievement motivation and order (Baker et al., 2026). The broad trait scores obscure these differences entirely.

What Predicts Emergence

The picture for leader emergence is quite different, and the divergence is one of the study’s most practically important findings. Assertiveness was the strongest predictor of emergence among extraversion facets (ρ = .21), consistent with its role in social dominance and the perception of leaderlike qualities (Baker et al., 2026). This aligns with implicit leadership theory: people who interject, take control of conversations, and project confidence fit the mental prototype of a leader.

But several facets that predicted effectiveness showed negligible or negative relationships with emergence. Altruism was associated with effectiveness (ρ = .16) but negatively associated with emergence (ρ = -.12). Order and self-discipline showed some of the strongest correlations with effectiveness but null or negative relationships with emergence. Achievement motivation, a strong effectiveness predictor, had essentially no relationship with emergence (ρ = .01) (Baker et al., 2026).

The implication is direct and uncomfortable: the qualities that cause people to be perceived as leaders and promoted into leadership roles are not the same qualities that make them effective once there. Organisations that allow emergence processes, informal influence, social dominance, and prototype-matching to drive leader selection may be systematically selecting for the wrong characteristics.

Context Matters: Leadership Level

The meta-analysis also examined whether the validity of personality facets differed by leadership level, comparing student leaders, lower level organisational leaders, and higher level leaders. The differences were substantial in some cases.

Conscientiousness facets were considerably more predictive for student leaders (order: ρ = .40) than for lower or higher level organisational leaders (ρ = .09 to .10) (Baker et al., 2026). This likely reflects the genuine importance of these traits for academic performance, which dominates student leadership contexts, rather than organisational leadership per se. Studies that aggregate student and working adult samples, as most previous meta-analyses have done, will therefore produce inflated validity estimates for conscientiousness that do not generalise to employed contexts.

Optimism was significantly more predictive for higher level leaders (ρ = .28) than for lower level leaders (ρ = .16), while extraversion showed the reverse pattern, with stronger relationships at lower levels (ρ = .34) than higher levels (ρ = .08) (Baker et al., 2026). This makes conceptual sense: lower level leaders succeed through direct interpersonal influence and relationship management; higher level leaders need to set tone, maintain direction under pressure, and make decisions in conditions of sustained uncertainty.

The Value of Optimally Weighted Composites

When the authors combined facets into regression-weighted composites rather than simply using broad trait scores, predictive validity improved substantially. The openness composite more than doubled the validity compared with the broad openness trait (Multiple R = .19 versus ρ = .08), and the conscientiousness composite was notably stronger than the broad trait (Multiple R = .36 versus ρ = .26) (Baker et al., 2026). This has direct practical implications: organisations assessing personality for leadership selection using only broad trait scores are working with a considerably blunter instrument than necessary.

Why Previous Estimates Were Inflated

The study also provides an important corrective to earlier meta-analytic figures. Judge and colleagues (2002), Hoffman and colleagues (2011), and Ensari and colleagues (2011) all found correlations of .35 to .39 for assertiveness with leadership outcomes. The present study finds .14 for effectiveness and .21 for emergence (Baker et al., 2026). The difference is largely attributable to the inclusion of self-reported leadership outcomes in earlier studies, which inflate estimates through common method variance, and the aggregation of student samples with employed samples. These are not trivial methodological choices: they have shaped a generation of research and practice built on validity estimates that do not accurately reflect the relationship between personality and leadership in working adult populations.

Why It Matters

The study has layered practical implications. At the assessment level, it makes a strong case for moving beyond broad trait scores to facet-level measurement and optimally weighted composites when predicting leader outcomes. The investment in more granular assessment pays off in predictive validity. At the selection level, it highlights the danger of relying on emergence processes to identify effective leaders: the personality profile that predicts who gets chosen as a leader diverges substantially from the profile that predicts who performs well once chosen. And at the development level, traits that receive relatively little attention in leadership programmes, optimism, stress resistance, altruism, emerge here as meaningful contributors to effectiveness, pointing toward underexplored development targets.

Reference

Baker, N., Scott, W., Nye, C. D., Chernyshenko, O. S., Park, H. W., & Omori, C. L. (2026). The many facets of leadership: A meta-analysis of personality facets, leader effectiveness, and emergence. Journal of Applied Psychology. Advance online publication. https://dx.doi.org/10.1037/apl0001378