Personality and Sexual Orientation: New Data and Meta-analysis. Mark S. Allen & Davina A. Robson. The Journal of Sex Research , Jun 8 2020.
ABSTRACT: This research explored associations between personality and sexual orientation. In Study 1, we explored whether the Big Five trait dimensions relate to sexual orientation in a nationally representative sample of Australian adults (n = 13,351). Personality differences were observed between those who identified as heterosexual (straight), bisexual, and homosexual (gay/lesbian) on all five measured traits. In Study 2, we conducted an updated systematic review and meta-analysis of personality and sexual orientation. A total of 21 studies (35 independent samples, 262 effect sizes) comprising 377,951 men and women were identified that satisfied inclusion criteria. Results showed that bisexual individuals reported higher levels of openness than homosexual individuals, who in turn, reported higher levels of openness than heterosexual individuals. Bisexual individuals also report lower levels of conscientiousness than both heterosexual and homosexual individuals. Sex moderation effects showed that homosexual men scored higher than heterosexual men on neuroticism, agreeableness and conscientiousness, whereas homosexual women scored lower than heterosexual women on extraversion, agreeableness, and conscientiousness. There was also evidence that personality differences between sexual orientation categories tend to decline with age. These findings align with the gender-shift hypothesis and should be of interest to theorists working in personality science and sexual identity development.
General Discussion
This research sought to determine whether personality differs between heterosexual, bisexual, and homosexual persons. Findings from Study 2 (that integrate Study 1) indicate that bisexual individuals report higher levels of openness than homosexual individuals, who in turn, report higher levels of openness than heterosexual individuals. Bisexual individuals also showed lower levels of conscientiousness than heterosexual and homosexual individuals. Sex moderation effects were observed for neuroticism, extraversion, agreeableness, and conscientiousness. These effects were such that homosexual men scored higher than heterosexual men on neuroticism, agreeableness and conscientiousness, whereas homosexual women scored lower than heterosexual women on extraversion, agreeableness, and conscientiousness. There was also evidence that personality trait differences between sexual orientation categories tend to decline with age. Overall, the findings of this research suggest small-medium effect size differences in trait personality between persons of different sexual orientation.
The finding that homosexual men and women scored higher on openness than heterosexual men and women is consistent with findings reported in a recent meta-analysis (Allen & Walter,
2018a) as well as theoretical propositions that openness should be most important for sexual identity formation (Lippa,
2005; Zoeterman & Wright,
2014). An important finding was that bisexual individuals reported higher levels of openness than both heterosexual individuals (medium effect size) and homosexual individuals (small effect size). This indicates that the association between openness and sexual orientation – if considered on a continuum from completely heterosexual to completely homosexual – is curvilinear in nature. This finding might be anticipated given that bisexual individuals tend to experience more identity confusion and emotional turmoil than homosexual individuals (Balsam & Mohr,
2007) meaning high levels of openness might be a useful attribute to discourage identity foreclosure (making a commitment without full exploration) among those who might otherwise adopt a bisexual orientation (see Zoeterman & Wright,
2014).
The findings of this research differ somewhat from those reported in a recent meta-analysis (Allen & Walter,
2018a) that found limited evidence for the gender-shift hypothesis. The previous meta-analyzes found that heterosexual and homosexual persons did not differ on extraversion, agreeableness or conscientiousness (Allen & Walter,
2018a). In fact, the only sex moderation effect observed was for neuroticism in which homosexual men reported higher levels of neuroticism than heterosexual men and homosexual women reported lower levels of neuroticism than heterosexual women. None of those findings were observed here. Rather, the neuroticism sex moderation effect indicated that while heterosexual men did score lower on neuroticism than homosexual (and bisexual) men, homosexual and heterosexual women did not differ in levels of neuroticism. Important sex moderations were also observed for extraversion (such that homosexual women report higher levels of extraversion than heterosexual women, but levels of extraversion did not differ between homosexual and heterosexual men), agreeableness (such that homosexual women report lower levels of agreeableness than heterosexual women, whereas homosexual men report higher levels of agreeableness than heterosexual men), and conscientiousness (such that homosexual women report lower levels of conscientiousness than heterosexual women, whereas homosexual men report higher levels of conscientiousness than heterosexual men).
By separating effects for bisexual and homosexual individuals, and focusing on higher quality studies (studies of self-reported sexual identity and representative data samples), we have been able to remove much of the excessive heterogeneity observed in the previous meta-analysis, narrowing confidence intervals, and revealing important sex moderations. That sex moderation effects were observed on four of the five dimensions provides some evidence for the gender-shift hypothesis (Lippa,
2005). Compared to men, women tend to score higher on neuroticism and agreeableness, and to a lesser extent, on extraversion and conscientiousness (De Bolle et al.,
2015; Schmitt et al.,
2008). Given these personality differences between men and women, the sex moderation effects observed for neuroticism, agreeableness and conscientiousness (between homosexual and heterosexual individuals) are in the direction predicted by the gender-shift hypothesis. The sex moderation effect for extraversion – showing that homosexual women score higher that heterosexual women on extraversion – is not in the direction predicted by the gender-shift hypothesis given that heterosexual women tend to score higher on extraversion than heterosexual men (albeit with a small effect size). This finding might be explained by cultural factors, as sex differences in extraversion do not always transfer across world regions (Schmitt et al.,
2008).
Another important new finding was that bisexual individuals tend to score lower on conscientiousness than heterosexual and homosexual individuals, and this effect emerged for both men and women. Combined with the sex moderation effect for homosexual and heterosexual individuals (the finding that homosexual women report lower levels of conscientiousness than heterosexual women, whereas homosexual men report higher levels of conscientiousness than heterosexual men), this pattern of results is indicative of a curvilinear effect for conscientiousness (similar to openness) but only among women. For men, this pattern of results is indicative of a linear association (see
Table 4). Because bisexual individuals tend to experience more identity confusion and emotional turmoil than homosexual individuals (Balsam & Mohr,
2007), we can speculate that – similar to high openness – low conscientiousness might be a useful attribute to discourage identity foreclosure (see Zoeterman & Wright,
2014) but only among women. For men, low conscientiousness might be less important to processes involved in identity formation and further research is required to help understand why this might be the case. Further research into these processes could also help to explain why a bisexual orientation is more common among women than among men (Bailey et al.,
2016).
Researchers have invested a considerable amount of time investigating how trait personality changes over the lifespan (Caspi et al.,
2005; Costa et al.,
2019; Damian et al.,
2019; Roberts et al.,
2006). This research indicates that as people become older neuroticism tends to decrease, whereas agreeableness and conscientiousness tend to increase (the maturity principle). The age moderation effects observed in the current study – showing that differences between sexual orientation groups tend to decline with age – might be important to understanding personality trajectories. It was found that the medium effect size difference in neuroticism between heterosexual and homosexual men, the small effect size difference in extraversion between heterosexual and homosexual women, the small-medium effect size difference in openness between heterosexual, bisexual and homosexual women, and the small effect size difference in agreeableness between heterosexual and homosexual women, all decreased in magnitude as the sample age increased. Given these age moderation effects, sexual orientation might be an important consideration in personality trajectories across the lifespan. It might be the case that general personality trajectories (decreases in neuroticism, and increases in agreeableness and conscientiousness) differ in sexual minority samples. Compared to heterosexual men, homosexual men might show a greater decline in neuroticism and less of a decline in agreeableness and conscientiousness. Compared to heterosexual women, homosexual women might show a greater decline in agreeableness and conscientiousness. Longitudinal research is needed to test these possibilities directly.
The current research has a number of potential shortcomings that readers must consider when interpreting findings. First, while the current meta-analysis provides estimates for population-based samples, the samples included were somewhat limited to industrialized nations and how well findings transfer to all world regions remains unknown. Indeed, sex differences in personality do not always transfer across cultures (Schmitt et al.,
2008) and further research in needed to test whether personality differences between people of different sexual orientations can be applied to all world regions. Second, the meta-analysis tested for linear effects between three sexual orientation categories. The pattern of results indicated that some associations might be curvilinear in nature if sexual orientation is considered on a continuum from fully heterosexual to fully homosexual. Researchers have suggested that in-between groups (e.g., primarily heterosexual, mostly homosexual) might also be important and distinct sexual minority categories (Savin-Williams,
2016; Walton et al.,
2016) and future research might consider exploring (more directly) the magnitude of potential curvilinear effects. There is also some evidence that people who identify as asexual (no sexual attraction to persons of either sex) have distinct personality profiles (Bogaert et al.,
2018; Carvalho et al.,
2017) and further research into asexual samples is also important. In addition, other non-heterosexual identity labels are increasingly being used (e.g., pansexual, queer; Goldberg et al.,
2020) that could also be explored in relation to personality.
A third limitation is the focus on higher-level personality dimensions rather than lower-level trait facets. Research has found that women tend to score higher than men on extraversion facets related to warmth and sociability, whereas men tend to score higher than women on extraversion facets related to assertiveness and excitement seeking (Costa et al.,
2001). To more adequately test the gender shift hypothesis, researchers should look to explore lower-level trait facets. A fourth potential limitation relates to measurement differences in sexual orientation with some studies focusing on self-reported sexual identity (categorized) and others exploring relative levels of same-sex sexual attraction. These differences in conceptualization might contribute somewhat to increased measurement error. A final limitation is that included studies used cross-sectional research designs and therefore whether associations between personality and sexual orientation are spurious or causal in nature remains unknown. Prospective research designs can help provide information on how personality development and sexual identity development interconnect over the adult lifespan.
Despite these potential limitations, the current research extends knowledge in this important area and findings for all five personality dimensions reported in the previous meta-analysis (Allen & Walter,
2018a) have been superseded in this work. For neuroticism, our results show that rather than sexual orientation differences for both men and women (Allen & Walter,
2018a), only men tend to differ on scores for neuroticism. For openness, rather than a straightforward difference between heterosexual and homosexual persons (Allen & Walter,
2018a), our results indicate a curvilinear association in which bisexual individuals score highest on openness. For extraversion, agreeableness and conscientiousness, rather than no personality differences across sexual orientation categories for men or women (Allen & Walter,
2018a), our results show sex moderation effects in which sexual orientation groups differ on all these dimensions. Furthermore, rather than these differences being static across the lifespan (i.e., no age moderation effects; Allen & Walter,
2018a), our results indicate that personality differences between persons of different sexual orientation tend to decline with age. In other words, the current research provides more precise estimates in meta-analysis and greater support for the gender-shift hypothesis overall.
In terms of research progression, researchers might consider exploring further how personality relates to processes important in sexual identity formation. The minority stress model of sexual orientation (Meyer,
2003) outlines that sexual minorities are at greater risk of mental health problems (see Allen,
2018; Semlyen et al.,
2016), because of stigma, prejudice, and discrimination that create a stressful social environment. The degree to which sexual minority individuals are susceptible to such stress is thought to be moderated by personality traits (Bailey,
2020) that are particularly important in stress and coping (Carver & Connor-Smith,
2010; Connor-Smith & Flachsbart,
2007). Researchers might also consider exploring further how personality relates to stages of sexual identity development. Stage based models (e.g., Cass,
1979) have received some criticism for not considering individual difference factors that might explain why some individuals do not progress through stages in a linear manner or progress through all stages (Rosario et al.,
2011). These variations in identity formation might be explained, in part, by personality traits and further research into these processes can help increase understanding of sexual identity development.
To conclude, this research has identified small-medium effect size differences in personality traits between heterosexual, bisexual, and homosexual persons. Important differences were found on all five personality dimensions, and effects were generally in line with predictions of the gender-shift hypothesis. Results also showed that personality trait differences between sexual orientation categories tend to decline with age. The findings of this research could have implications for theory and research on personality trajectories over the lifespan, and might also be of interest to health care professionals working with sexual minority (LGB) individuals. Sexual identity development can be a complex and often difficult process (Rosario et al.,
2011) and personality tests might become a useful method to identify individuals who might benefit greatest from inclusion in support programs for sexual identity distress. More generally, personality tests could be useful in helping LGB individuals gain greater self-awareness of their own personality. We recommend further research using prospective designs, in understudied cultural groups and world-regions, and more rigorous tests of potential curvilinear associations, to further understand how personality and sexual orientation interconnect over the lifespan.
Posts
