Effects of personality and gender on self–other agreement in ratings of leadership

Abstract

We explore the role of leader personality (i.e., the Big 5 traits: Conscientiousness, Agreeableness, Openness, Extraversion, and Neuroticism) and gender in self–other (dis)agreement (SOA) in ratings of leadership. We contend that certain aspects of the leader's persona may be more or less related to self‐ or other‐ratings of the leader's behaviour if those aspects are (1) more or less observable by others, (2) more or less related to internal thoughts versus external behaviours, (3) more or less prone to self‐enhancement or self‐denigrating biases, or (4) more or less socially desirable. We utilize statistical methodologies that capture fully the effects of multiple independent variables on the congruence between two dependent variables (Edwards, , Organizational Behavior and Human Decision Processes, 64, 307), which previously have not been applied to this area of research. Our results support hypotheses predicting less SOA as leader Conscientiousness increases and greater SOA as Agreeableness and Neuroticism increase. Additionally, we found gender to be an important factor in SOA; female leaders exhibited greater SOA than did their male counterparts. We discuss the implications of these findings, limitations, and future research directions. Practitioner points Popular practices such as 360‐degree feedback may reveal discrepancies between a person's self‐ratings and other's ratings. Although often attributed to a lack of self‐awareness, these discrepancies also may be explained by factors such as the personality and gender of the focal individual.

Health-Oriented Leadership and Mental Health From Supervisor and Employee Perspectives: A Multilevel and Multisource Approach

“…Furthermore, it is important to understand when and under which conditions the assessments between supervisors and employees correspond to each other and under which conditions supervisor assessments affect the health of employees. Previous research has shown that self-other agreement might be dependent from contextual work characteristics ( Ostroff et al, 2004 ), cultural aspects ( Cullen et al, 2015 ) as well as personal characteristics of the supervisors, such as age, gender or personality ( Bergner et al, 2016 ; McKee et al, 2018 ). Also, other leadership styles, such as authentic leadership, could moderate this relationship, which can be roughly characterized by a high degree of authenticity, self-confidence and self-regulation, transparency, and honesty of supervisors ( Gardner et al, 2011 ).…”

“…Meta-analytic findings estimated uncorrected correlations between supervisors and their subordinates on performance ratings at a level of 0.22 (supervisor-subordinate; Heidemeier and Moser, 2009 ). Furthermore, it has been assumed that correlations of different sources should be larger for observable patterns compared to non-observable constructs, as observable patterns show less ambiguity in ratings ( Dai et al, 2007 ; Heidemeier and Moser, 2009 ; McKee et al, 2018 ). To date, there is scarce evidence of how multisource assessments of HoL are related to each other, specifically spoken whether the self-rating of supervisors in HoL corresponds to how they are seen by their teams.…”

Gender Bias Still Plagues the Workplace: Looking at Derailment Risk and Performance With Self–Other Ratings

“…The two articles that included gender as a main variable of interest, that is, works by Patiar and Mia (2008) and Vecchio and Anderson (2009), both found that men tended to overestimate their effectiveness as leaders relative to how others rated them, whereas women tended to underestimate their effectiveness. A more recent study (McKee, Lee, Atwater, & Antonakis, 2018) also showed this overrating trend for men. Within the 10 workplace articles that included gender, the percentages of females in the samples ranged from 9% to 72% and averaged 37% female.…”

“…We found that most prior SOA research (from the past decade) does not account for gender as either a main variable of interest, a control variable, or by using a single-gender sample, though much of the research did acknowledge the proportions of men and women in study samples. Three articles using workplace samples that addressed gender as a main variable of interest found that men tend to overestimate how others see them, whereas women tend to underestimate (McKee et al, 2018; Patiar & Mia, 2008; Vecchio & Anderson, 2009). However, these studies did not look at the respective outcomes of over- and underrating for men and women leaders separately.…”

“…It is also reasonable to consider that the gender (mis)match between raters and their leaders may affect perceptions and ratings. According to McKee et al (2018), factors such as gender and personality may affect SOA. In addition, the quality of feedback shared between sources may be worthwhile to investigate.…”