Are Better Leaders Than They Think: Gender Differences in the Self-Assessment of Leadership Skills in the Maritime Industry

Female officers (, σ = 0.363) tended overall to be rated slightly higher by their co-workers than their male counterparts (, σ = 0.506). This effect was of a medium size ((t ₍₄₅₎ = −1.493, , 95 % CI of , d = 0.474, one-tailed p = 0.072) but the effect was not statistically significant (i.e., the type I error level was higher than the commonly accepted 0.05 level).

When it came to the officers’ self-evaluation relative to their co-worker’s evaluation (calculated as Self-rating—Co-worker’s rating), female officers (, σ = 2.55) tended to underestimate their own overall leadership skills compared to their male counterparts (, σ = 2.96). The difference between the women and men officers was of medium size in the predicted directions (, 95 %CI of , t ₍₄₅₎ = 1.993, d = 0.595, one-tailed p = 0.026). Female officers scored themselves, on average, lower than their co-workers, while male officers marginally scored themselves higher than the rating by their co-workers. These findings are in accordance with predictions of hypothesis 1.

The co-workers rated female (, σ = 1.03) or male (, σ = 1.67) officers with a statistically non-differentiable level of negative leadership skills. The effect size was small, to the extent of being negligible (t ₍₄₅₎ = 0.30,, 95 % CI of , d = 0.097, one-tailed p = 0.383). Thus, on the whole, female and male officers were on average rated by their co-workers to have the same level of negative leadership skills.

However, when it came to the officers’ self-evaluation relative to their co-worker’s evaluation (calculated as Self-rating—Co-worker’s rating), female officers (, σ = 1.195) tended to greatly overestimate their own level of negative leadership skills, compared to their male counterparts (, σ = 0.930). The difference between female and male officers was of medium size in the predicted directions (, 95 %CI of , t ₍₄₅₎ = −2.957, d = 0.595, one-tailed p < 0.003). Female officers scored themselves, on average, higher on negative leadership skills than their co-workers, while male officers marginally scored themselves lower than the rating by their co-workers. These findings are in accordance with predictions of hypothesis 2a.

There was a statistically significant difference on positive leadership skills between female (, σ = 0.475) and male (, σ = 0.504) officers. The effect size was (t ₍₄₅₎ = −1.937,, 95 % CI of , d = 0.593, one-sided p < 0.03). Female officers were, on average, rated higher by their co-workers on positive leadership skills than the male officers.

When it came to the officers’ self-evaluation relative to their co-worker’s evaluation (calculated as Self-rating—Co-worker’s rating), female officers (, σ = 2.253) tended to underestimate their own level of positive leadership skills compared to their male counterparts (, σ = 2.36). The difference between female and male officers was small but in the predicted direction (t ₍₄₅₎ = 1.104,, one-sided 95 % CI of , d = 0.33, one-sided p = 0.138). These findings are in the same direction as the predictions of hypothesis 2b, but the observed effect is too small for us to reject the null hypothesis.

After observing that the differences between the leader’s self-rating and the co-workers rating is dependent upon gender and the valence (the negative or positive value attributed to a leadership skill) of the leadership skills in question, we evaluated the possibility of an interaction effect (as would be expected based upon hypothesis 1 to 2a and 2b). A repeated measures ANOVA, with gender as a within-group factor, was calculated. The results showed that even in our small sample, there are indications of an interaction effect (F_{(1, 45)} = 3.973, p = 0.052, η _p ² = 0.081). The interaction effect is shown in Fig. 1.

As can be seen in Fig. 1