Introduction

Certain personality traits aid in individuals’ “getting along” and “getting ahead” in their social environments, and the relationship for personality traits with ideologies and behaviors is moderated by the geographical context in which individuals reside. For decades, psychologists have explored how individuals and their situations interact, influencing our actions and beliefs. Given that socio-cultural components permeate every aspect of human life, it is necessary to delve into the nuanced relationship between personality traits and geographical context. This idea is captured by Person-Environment Fit Theory1,2,3,4,5, which suggests a self-regulatory mechanism that strives to align an individual’s personality with their socio-cultural surroundings.

We have shown previously that the strength and direction of relationships for pro-environmental behaviors with five broad personality traits (i.e., five factors discussed by Goldberg6 and 27 interstitial facets assessed with the SAPA personality inventory (SPI)7 vary across different regions in the USA. That is, the relationship between higher-order personality traits and pro-environmental behaviors is moderated by the average levels of pro-environmental behaviors exhibited by people in a given area8. Specifically, Extraversion and Openness were found to be more positively correlated with pro-environmental behaviors in ZIP (Zone Improvement Plan) Code Tabulation Areas (ZCTAs) with higher average levels of pro-environmental behaviors, while Extraversion and Neuroticism were more negatively associated with pro-environmental behaviors in ZCTAs with lower average pro-environmental behaviors. Similar variations have been observed in the greater London area, where the relationship between personality traits and wellbeing varies across units of geographical space defined by postal codes9. That is, neighborhoods with greater population density and ethnic diversity demonstrated the strongest positive associations between Openness and life satisfaction, while neighborhoods with overall lower levels of life satisfaction had stronger associations for Agreeableness and Conscientiousness with life satisfaction. Geographically varying associations as a function of socio-cultural context (e.g., prominence of conservative ideology) have been observed across the US for the association between Conservatism and longevity10, with a significant association between Conservatism and longevity found in more conservative places but little to no association found in less conservative place. Similarly, the associations between Agreeableness, Conscientiousness, and religiosity vary depending on the religious context, with these traits being more positively associated with religiosity in highly religious regions11.

Considering predictive accuracy for behaviors is optimized through using a combination of psychological characteristics, situational context, and time12, the aforementioned literature collectively highlight how socio-cultural context influences the relationship between psychological traits and ideologies, as well as behaviors. Deeper comprehension of the nature of variation in associations between personality traits and ideologies, like Conservatism, can lead to greater understanding of the unique ways personality is enacted within geographical contexts to enable individuals to best “get along” and “get ahead”13 in their social groups.

Person-environment fit theory

Person-Environment Fit theory1,2,3,4,5, Socioanalytic theory13,14,15, and Cultural Cognition16,17,18,19,20,21 theory address why personality characteristics may have variable effects on ideologies and behaviors. Person-Environment Fit theory posits that humans have complex transactional relationships with the environments (i.e., social, physical, cultural) they inhabit and are constantly at work in creating and maintaining “fit” in those environments1,2,3,4,5. Core tenets of Person-Environment Fit theory include Interaction and Compatibility. Interaction refers to the greater predictive accuracy and understanding of behavior achieved when accounting for both persons and environments together, rather than in isolation. Compatibility refers to how the most optimal life outcomes occur when individual characteristics align with environmental characteristics, such as values.

Person-Environment fit can be thought of as a “self-regulatory mechanism” constantly at work to optimize positive psychological outcomes and overall wellbeing2,4.To maintain Person-Environment Fit, humans either find environments that fit their personality (Attraction), modify their environments or construct new ones to fit who they are and what they believe (Construction), or adapt to their environments by adjusting their ideologies and behaviors (Conformation). When attraction and construction are not viable options, individuals tend to conform their thoughts, ideologies, motivations, and behaviors to meet the demands of their environments and preserve social ties2,3. Individual psychological characteristics and environmental characteristics are in constant interplay and are thus fundamentally entangled3,22. This dynamic interplay significantly influences a variety of life outcomes, including the development of psychopathologies2,3,22.

Socioanalytic theory

Socioanalytic theory describes a framework for understanding personality focused on elucidating why and how people differ in their motivations and abilities to garner acceptance in their social environments, acquire and maintain resources and status, as well as to discover and generate sense of purpose and meaning13,14,15. The theory draws on components of role theory23 and evolutionary psychology24 and posits that there are significant individual differences in how individuals strive to best “get along” and “get ahead” in their environments. Socioanalytic theory complements work on Person-Environment Fit theory, such that personality is suggested to be in constant interplay with socio-cultural and physical contexts, which ultimately produces behavior25. Through this framework, it is important to consider personality from a perspective of both identity and reputation. That is, individuals function best within their environmental contexts when they are able to form and maintain both group and individual identity as well as positive reputation in their social groups26,27. These processes lead individuals to adapt to their environments in ways that are most functional to “get along” and “get ahead” as best as they can. Moreover, it is logical that personality traits would not universally predict certain ideologies and behaviors in the exact same way across different social, physical, and cultural environments because how these traits function in individuals best “getting along” and “getting ahead” is highly dependent on characteristics of the socio-cultural environment.

Cultural cognition theory

Cultural Cognition theory16,17,18,19,20,21 describes how cultural dimensions and value systems shape individuals’ orientations within their socio-cultural environments. An important element of Cultural Cognition theory involves how socio-cultural orientations influence perceptions of norms, evidence, risk, as well as social and political issues. The socio-cultural values inherent in individuals’ unique geographic contexts play a crucial role in shaping their opinions on a variety of social and political issues. Cultural Cognition theory’s assertion that individuals form opinions and behave in ways that are functional for their socio-cultural roles strongly complements Hogan’s Socioanalytic theory, which suggests that individual differences in personality serve to aid in individuals’ best “getting along” and “getting ahead” in their environments14,15,28.

Kahan and colleagues18 demonstrated that individuals often engage in selective interpretations or outright rejections of scientific evidence when it does not align with the value system of their socio-cultural context. Furthermore, this cognitive process, involving conformity of ideologies to match the socio-cultural ethos of one’s environment, occurs as an adaptive means of preserving cooperation, cohesion, and identity16,19 in one’s social groups. Adapting to one’s social environment is advantageous, as “people endorse whichever position reinforces their connection to others with whom they share important ties”16, p. 296]. Cultural cognition also strongly influences individuals’ policy views and opinions29. That is, individuals tend to subscribe to policy stances that align most closely with their socio-cultural values. It is suggested that this assimilation of policy views within socio-cultural environments occurs, in part, due to cultural acquisition, which ultimately enables individuals to form and protect their identities within a socio-cultural context29.

Personality traits and conservatism

Previous research has suggested certain personality traits are significantly associated with conservative ideology30,31,32,33; such that higher Conscientiousness and lower Openness relate to higher Conservatism across individuals. The association between high Conservatism and low Openness is thought to be due to the drive to explore new ideas and ideologies that Openness encompasses, which typically results in more liberal orientations30,31. Conscientiousness is typically found to be positively associated with Conservatism because of the drives to seek order and tradition that are inherent to this personality trait30,31. Agreeableness, Neuroticism, and Extraversion have mixed results in terms of their relationship with Conservatism which is partly due to low consistency in the association across contexts30,33.

Additionally, there are a variety of personality facet or subfactor associations with Conservatism proposed by previous researchers34. For instance, the compassion facet of Agreeableness was shown to be positively associated with liberal ideology while the politeness facet of Agreeableness was positively associated with conservative ideology35. The orderliness facet of Conscientiousness was also found to relate to conservative ideology in this study. Personality facets of intellectual curiosity, intellectual sensitivity, and depression have shown negative associations with right-wing ideology36. Overall, while general associations between personality factors and facets have been demonstrated across studies, there is notable variation in the type of observed association when studied in different samples; for instance, higher “systemic threat” in one’s environment can shape the association between Openness and Conservatism to be either strong or negligible dependent on threat level34. Thus, it is important to better understanding how socio-cultural aspects of geographical context shape the relationship between personality indicators and Conservatism.

Geographically varying effects

Geographical psychologists have been studying the spatial organization and clustering of individual-level psychological attributes for over a decade37,38,39,40. There are clear regional differences in the level and expression of psychological attributes, like personality39. More specifically, there are clear differences in average levels of Five Factor personality traits (i.e., Agreeableness, Conscientiousness, Neuroticism, Extraversion, Openness) across regions influenced by various environmental and socio-cultural factors. For instance, Rentfrow and colleagues39 found that higher Openness is more common on the west coast of the USA, while Neuroticism is generally higher in Northeastern states. These differences are thought to be influenced by regional differences in the USA surrounding culture and lifestyle. Lanning and colleagues37 also observed differences in aggregated levels of personality across four regions of the US defined by their migration history and culture. Along with personality traits, there are also stark regional differences in behaviors and attitudes40. To better understand regional variation in psychological characteristics, it is important to study the socio-cultural contextual factors that drive geographical differences in levels of various psychological characteristics and attempt to understand the nuanced interplay between individuals and their geographical environments38. Additionally, previous work has shown that retest reliability of certain measures of implicit psychological features is much higher when regionally aggregated than it is at the individual levels41 suggesting a potential benefit for reliably studying person by environment interaction through a combination of individual level and regional aggregated level data.

It is also important to consider Person-Culture Match, which refers to how individuals benefit from reduced stress by matching their ideologies and values to that of their socio-cultural context42. Person-Culture Match is associated with greater self-esteem and subjective wellbeing across dozens of nations42 and is thought to influence ideological segregation as well as migration43. That is, mismatch in ideology between persons and their socio-cultural contexts is linked to greater residential mobility and migration to communities matching the ideologies of the migrating persons, which also reflects the Attraction tenet of Person-Environment Fit theory. This research provides support for our hypothesis for differential relationships between personality traits and Conservatism across places because as certain personality characteristics function to enable individuals’ best “fitting in” as well as “getting along” and “getting ahead” in their environments, how these characteristics influence an individual’s conservative ideology and aid in Person-Culture Match should naturally depend on whether the person is living in a conservative environment. All in all, more research is needed to disentangle how personality traits manifest in their contribution to ideological views within various socio-cultural contexts and to understand the nature of how these traits influence individuals’ matching of their socio-cultural values to those of their environment.

Study overview and hypotheses

In the present study, we examined how geographical context influences the complex associations between broad higher-order factors of personality (i.e., Agreeableness, Conscientiousness, Neuroticism, Extraversion, Openness) as well as lower-level interstitial personality trait facets with Conservatism across the USA. In other words, we examined the extent to which variation across geographical contexts in personality trait relationships with Conservatism are influenced by how conservative people are on average in a given place. We anticipated Context Specific Personality Associations similar to our previous findings8. We define Context Specific Personality Associations (CSPAs) as instances when the strength and direction of associations for personality traits with ideologies or behaviors depend upon the geographical context.

We propose that the nature of associations for personality traits with ideologies and behaviors function in such a way that enables individuals to best “fit in” with their socio-cultural environment. We expect that this variation in personality trait associations with ideologies and behaviors across fine units of geographical space occur because higher levels of certain personality traits ultimately function to help individuals “get along” and “get ahead” in their respective socio-cultural environments (Socioanalytic Theory)13,14,15 and because individuals tend to develop opinions and behaviors that are adaptive and functional for their social roles within a socio-cultural context (Cultural Cognition Theory)16,17,18,19,20,21. We also expect this variation is partially due to the Conformation element of the Person-Environment Fit “self-regulatory mechanism”2,4 and that the regional distribution in strength and direction of personality trait associations with ideologies and behaviors may also be influenced by how the Attraction and Construction elements effect where people live and the socio-cultural environments within those places. We propose that the “getting along” and “getting ahead” dimensions of personality contribute meaningfully to contextualized spatial personality associations (CSPAs). Specifically, we hypothesize that regional levels of Conservatism will significantly moderate the relationships between Conservatism and the five major personality traits: Agreeableness, Conscientiousness, Neuroticism, Extraversion, and Openness. That is, we expect the associations for SPI-5 personality factors and lower-level facets with Conservatism to vary systematically across geographic regions as a function of local Conservatism levels. Some traits may exhibit Polarized CSPAs, wherein the direction of the association between the trait and Conservatism shifts from positive to negative depending on the local level of Conservatism. Others may demonstrate Amplified CSPAs, characterized by variations in the strength of the association, but not the direction, across different local levels of Conservatism. We also analyzed how local levels of Conservatism moderate the associations between interstitial personality facets and individual levels of Conservatism to further disentangle the effects.

Results

Individuals’ geographical location of residence moderates the relationship between personality traits and Conservatism. That is, individual-level personality traits and average local levels of Conservatism interact in their relationship with individual-level Conservatism. More specifically, individuals with higher levels of Agreeableness, Conscientiousness, and Extraversion as well as lower levels of Neuroticism exhibited levels of Conservatism more reflective of the ideological ethos of their geographical location than their counterparts.

We identified the ZCTA-specific standardized regression coefficients (slopes) representing the unique association between each personality trait and Conservatism within each ZCTA and calculated the correlations (referred to as “adjusted r” in this paper) between these ZCTA-specific slopes with the random intercepts (i.e., average levels of Conservatism within ZCTAs). The hierarchical linear models employed for calculating these correlation coefficients inherently accounted for (1) any within cluster dependency, (2) the fixed effect of the personality indicator on the outcome, and (3) sample size44. These adjusted correlation coefficients can be interpreted as the extent to which average levels of Conservatism within ZCTAs moderate the relationship for each SPI-5 and SPI-27 personality factor with Conservatism.

With this methodology, we evaluated the variation in the association between personality traits and Conservatism and to what extent it depended on average levels of Conservatism in individuals’ respective ZCTAs. We found that the relationships between individual-level personality variables and individual-level Conservatism (slopes) within ZCTAs were moderately to strongly correlated with average levels of Conservatism across corresponding ZCTAs. That is, across ZCTAs, individuals with higher levels of Agreeableness (adjusted r = 0.45), Conscientiousness (0.51), Extraversion (0.15), and lower levels of Neuroticism (-0.22) had higher levels of Conservatism in ZCTAs that also had higher average levels of Conservatism. Moreover, in ZCTAs with lower average levels of Conservatism, the strength of associations for Conscientiousness and Neuroticism with Conservatism were weaker, but increasingly stronger in ZCTAs with higher levels of Conservatism (Fig. 1b and c), suggesting an Amplified Context Specific Personality Association (CSPA). On the other hand, we see a Polarized CSPA for Agreeableness and Extraversion; such that, in ZCTAs with lower average levels of Conservatism, the association for Agreeableness and Extraversion with Conservatism was increasingly negative, but in ZCTAs with higher levels of Conservatism, the association was increasingly positive (Fig. 1a and d). The CSPA for Openness was very weak (Fig. 1e). All in all, prominence of Conservatism in a geographical location notably moderated the relationship for individual-level Agreeableness, Conscientiousness, Extraversion, and Neuroticism with individual-level Conservatism.

Fig. 1
figure 1

Scatter plots of within-ZCTA slopes as a function of ZCTA-level conservatism, demonstrating the interaction between personality traits and conservative social context in relation to conservatism with example cities of some of the highest and lowest ZCTA-specific slopes (nZCTAs = 8,708). The scatter plots display ZCTA-Specific Slopes (y-axis; The unique relationship between each SPI-5 higher-order factor and Conservatism within each ZCTA) in association with the average level of Conservatism within each ZCTA (x-axis). The strength of the associations for Conscientiousness and Neuroticism with Conservatism were weaker in ZCTAs with lower average Conservatism but increasingly stronger in ZCTAs with higher levels of Conservatism suggesting an Amplified Context Specific Personality Association (CSPA). In ZCTAs with lower average levels of Conservatism, the association for Agreeableness and Extraversion with Conservatism was increasingly negative, but in ZCTAs with higher levels of Conservatism, the association is increasingly positive, suggesting a polarized Context Specific Personality Association (CSPA). The CSPA for Openness was very weak.

Full size image

When repeating our analysis with the SPI-27 lower-order personality factors (excluding Conservatism because this was our outcome of interest), we found that local levels of Conservatism most strongly moderated the relationship between individual-level Conservatism and SPI-27 personality traits of Authoritarianism (adjusted r = 0.57), Honesty (0.47), Wellbeing (0.47), Compassion (0.42), and Industry (0.36) (see Table 1). The multilevel models for Easy Goingness, Order, Emotional Expressiveness, and Intellect failed to converge, so we do not present results for these indicators. Authoritarianism, Wellbeing, Industry, Anxiety, Art Appreciation, and Conformity had Amplified CSPAs, similar to that of Conscientiousness, and Neuroticism. Polarized CSPAs, similar to that of Agreeableness and Extraversion, were observed for Honesty, Compassion, Self Control, Trust, Impulsivity, Sociability, Perfectionism, Emotional Stability, Charisma, and Sensation Seeking. The CSPAs were very weak for Irritability, Humor, Adaptability, Attention Seeking, Introspection, and Creativity.

Table 1 Mean slope for the relationship between SPI- 5 and SPI-27 factors with conservatism within ZCTAs, standard Deviations, minimum slopes, and maximum slopes, as well as the adjusted correlations between the random intercepts and slopes (i.e., the interaction effect).
Full size table

Density curves representing the range of SPI-5 higher order factors and Conservatism within each ZCTA show large variation in Conservatism across ZCTAs and moderate variation in average levels of SPI-5 personality traits across ZCTAs (Fig. 2). Density curves for the range of slopes, representing the unique association between each SPI-5 personality trait and Conservatism across ZCTAs show notable range in the strength and direction of association between personality traits and Conservatism (Fig. 3).

Fig. 2
figure 2

Density curves for the ZCTA-specific average levels of SPI-5 higher-order factors and Conservatism (nZCTAs = 8,708). Density curves representing the range of SPI-5 higher order factor averages within ZCTAs and Conservatism averages within each ZCTA show large variation in Conservatism across ZCTAs and moderate variation in average levels of SPI-5 personality traits across ZCTAs.

Full size image
Fig. 3
figure 3

Density curves for the range of unique slopes across ZCTAs (i.e., The ZCTA-specific relationships for each SPI-5 higher-order factor with Conservatism; nZCTAs = 8,708). The density curves show the range of unique associations (ZCTA-specific slopes) between each SPI-5 personality trait and Conservatism across ZCTAs. There is notable range in the strength and direction of association between personality traits and Conservatism.

Full size image

The overall zero-order correlations between SPI-5 factors and Conservatism for the entire sample, the ranges in slopes across ZCTAs, and the adjusted correlations between the random intercepts and random slopes (the interaction effect) demonstrate that while some personality traits have general associations with Conservatism, the strength and direction of the association depends on the average levels of Conservatism in the geographic location of interest (see Table 1).

For demonstration purposes, we present scatter plots from some of the most and least conservative ZCTAs in our sample and show the unique association between Agreeableness and Conservatism within these ZCTAs, demonstrating a Polarized CSPA (Fig. 4a and b). We also show this for Conscientiousness, demonstrating an Amplified CSPA (Fig. 4c and d). As shown in the scatter plots, individuals living in highly conservative Valdosta, GA scored higher on Conservatism if they were more Agreeable. Conversely, individuals living in less conservative Silver Lake, CA in Los Angeles County tended scored lower on Conservatism if they were more Agreeable. We see that despite an overall correlation of 0.00 between Agreeableness and Conservatism, which would have appeared null in the absence of a geographic analysis, there is much more nuance to the relationship between Agreeableness and Conservatism when studied on a context-dependent basis (i.e., interaction effect/adjusted r = 0.45, and the slope ranges from − 0.23 to + 0.33). We also observe that in highly conservative Pineville, LA, there is a strong positive association between Conscientiousness and Conservatism, but in the much less conservative Brooklyn, NY, the association is much weaker. While the zero-order correlation between Conservatism and Conscientiousness is small (0.19), the levels of Conservatism in the geographical context where the association is observed notably shapes the strength of the association between individual levels of Conservatism and Conscientiousness (i.e., interaction effect/adjusted r = 0.51, and the slope ranges from + 0.06 to + 0.62).

Fig. 4
figure 4

Examples of relationship between Agreeableness and Conscientiousness with Conservatism in ZCTAs with higher average Conservatism (Valdosta, Georgia & Pineville, Louisiana) and lower average Conservatism (Los Angeles/Silver Lake, California & Brooklyn, NY). This figure illustrates the unique association between Agreeableness and Conservatism within ZCTAs with some of the highest (Valdosta, GA) and lowest (Silver Lake, CA) Conservatism averages, highlighting a Polarized CSPA (a and b). The scatter plots reveal that in highly conservative Valdosta, GA, individuals who are more agreeable tend to score higher on Conservatism. In contrast, in the less conservative Silver Lake, CA, individuals who are more agreeable tend to score lower on Conservatism. Despite an overall correlation of 0.00 between Agreeableness and Conservatism, which might appear null without geographic analysis, the relationship is more nuanced when examined contextually. Furthermore, in highly conservative Pineville, LA, there is a strong positive association between Conscientiousness and Conservatism, whereas in the much less conservative Brooklyn, NY, this association is significantly weaker, suggesting an Amplified CSPA. See supplementary material, Appendix A, Table A1 for pattern of contextualized spatial personality associations (CSPAs) when the Conservatism scale is disaggregated into two distinct dimensions (i.e., religiosity and political orientation).

Full size image

Discussion

The relationships between personality traits and Conservatism clearly vary across geographical locations and are moderated by the prevalence of Conservatism within those locations. Our findings suggest that individuals with personality traits deemed more adaptive (i.e., traits that promote optimal physical and social outcomes, such as high Agreeableness, Conscientiousness, and Extraversion, along with low Neuroticism) tend to exhibit more aligned ideologies with the prevailing ethos of their socio-cultural environments than counterparts lower on these traits. While general associations at the population level between personality traits and Conservatism are apparent, they do not tell the full story of these associations. That is, the zero-order correlations for Conservatism with Openness and Conscientiousness are apparent yet small, and for Agreeableness, Neuroticism, and Extraversion, these correlations are weak. When we examine how local context moderates these associations, we see that local levels of Conservatism meaningfully influence these associations. Notably, the largest moderating effect of a conservative geographical context was observed in the associations between Conservatism and the traits of Agreeableness and Conscientiousness. Gebauer and colleagues11 emphasized a socio-cultural motives perspective, proposing that Agreeableness and Conscientiousness are key drivers of individual alignment to socio-cultural contexts, which is consistent with our observation that these traits have the largest CSPAs. Our findings support this perspective, showing that while the overall association between Agreeableness and Conservatism (r = 0.00) appears negligible across all participants in our sample, there is evident range in the association across geographical locations (-0.23 to 0.33). That is, the association is positive in more conservative areas and negative in less conservative ones. Additionally, Conscientiousness only correlates with Conservatism in conservative settings. These personality dimensions are pivotal in determining whether individuals assimilate their ideologies and behaviors to their socio-cultural context. Furthermore, Extraversion and Neuroticism also influence the alignment and conformity of ideologies with the socio-cultural ethos. Neuroticism may be more negatively associated with Conservatism in Conservative regions due to what this implies about level of person-environment fit in a region. In other words, being conservative in a conservative region would help individuals maintain better “fit” with their environment2,3,4, resulting in higher emotional stability than individuals who are conservative in a less conservative region. This clear alignment of ideologies and beliefs, driven by higher levels of these personality aspects, embodies what we term Context Specific Personality Associations (CSPAs).

We also found that eleven of the SPI-27 personality facets had interaction effects (CSPAs) larger than +/- 0.2. These eleven facets with interaction effects larger than +/- 0.2, were Authoritarianism (0.57). Honesty (0.47), Wellbeing (0.47). Compassion (0.42), Industry (0.36), Self Control (0.24), Trust (0.22), Impulsivity (-0.22), Sociability (0.22), Perfectionism (0.21), and Anxiety (-0.20). We expect that the strongest facet-level Polarized CSPAs for Conservatism with Honesty, Wellbeing, and Compassion were observed (i.e., more strong and positive associations with Conservatism in conservative regions and more strong and negative associations with Conservatism in less conservative regions) because these associations allow for better maintenance of fit and social congruence within social contexts. For instance, having ideological beliefs more congruent with one’s environment may enable an individual to feel they can be more honest or compassionate with others in their social context because they experience a sense of fit. Having greater wellbeing is also related to having greater person-environment fit2,3,4. Additionally, the strong Amplified CSPA observed with Authoritarianism reveals that while Authoritarianism is always positively associated with Conservatism, the association is stronger in more Conservative contexts. This amplification in the magnitude of association likely occurs due to contextual amplification, which describes how political ideology (e.g., Conservatism) can become more severe, salient, and intense in social contexts exhibiting higher levels of that political ideology (e.g., places with higher levels of Conservatism)45.

The alignment of ideologies and beliefs to one’s socio-cultural environment is adaptive, facilitating individuals’ maintenance and support social ties, fostering social connectedness and belonging17,20,21,46,47 as well as enabling individuals to best get along and get ahead in their social environments. It also allows individuals to find acceptance, acquire and maintain resources and status, and foster a sense of purpose and meaning in their lives14,15. Additionally, we likely observe geographically varying associations between personality traits and Conservatism due to Person-Environment Fit, specifically through Attraction, Construction, and Conformation2,4. That is, individuals tend to seek out environments that match their values (Attraction), create new ones (Construction), or conform their beliefs to those of the geographical context they are in (Conformation). Attraction drives migration processes that contribute to ideological segregation43, while Conformity drives the alignment of individual attitudes and beliefs to the geographical context. Both mechanisms likely contribute to the variation in personality associations with Conservatism across geographic locations.

Future research should aim to understand if CSPAs are enacted across micro-level groups and specific situational contexts, in addition to geographic macro-level groups. Perhaps individuals higher on the “getting along” and “getting ahead” dimensions of personality would exhibit greater “situationally appropriate behaviors” see48. Situationally appropriate behaviors refer to actions that are most fitting, proper, or socially acceptable in a given situational context. Individuals higher on personality dimensions encompassing positive emotionality, sociability, order, and self-control tend to exhibit more situationally appropriate behaviors, ultimately allowing them to get along better in their environments. Hogan and Kaiser49 have demonstrated that adolescent students who navigate their social environments effectively through a combination of cooperative and agentic behavior are more successful both academically and socially. Therefore, examining the interplay between personality and various group and situational contexts by assessing how specific situational contexts and group settings elicit situationally appropriate behaviors from individuals with certain personality compositions would be incredibly informative and potentially expansive to the study of CSPAs. Additionally, our findings on geographically varying personality associations dependent on local levels of Conservatism may suggest possible limitations to studying personality as it relates to ideologies on more ideologically segregated university-based samples.

All in all, individual differences in personality are crucial to consider in understanding differences in behaviors and attitudes, but associations between personality traits, ideologies, and behaviors do not occur in isolation. Accounting for how geographical context moderates the effect of personality characteristics on outcomes is critical for understanding the mechanisms influencing individual differences in ideologies and behaviors. Considering differences in personality and socio-cultural context jointly leads to the greatest predictive accuracy for behaviors12. Previous literature identifying geographically varying associations for psychological characteristics [e.g.9,10,11] in conjunction with our previous findings on geographically varying associations between personality traits and pro-environmental behaviors8 moderated by local levels of pro-environmental behaviors as well as our current study identifying how conservative geographical context shapes the direction and strength of associations between personality traits and Conservatism, collectively support our notion of Context Specific Personality Associations (CSPAs). Ultimately, the “getting along” and “getting ahead” dimensions of personality are functional in facilitating positive outcomes in health, wellbeing, and social connectedness. Therefore, individuals higher on these dimensions are naturally more likely to exhibit more aligned behaviors and ideologies with the geographical context of which they are a part. What we propose is not necessarily novel and is supported by Person-Environment Fit theory, Socioanalytic theory, and Cultural Cognition Theory. However, we propose that more research focusing on the differential associations between personality characteristics, ideologies, and behaviors across geographical contexts is fundamental for a comprehensive understanding of the implications of personality traits for real-life behaviors and attitudes.

Method

Participants/Data

Synthetic Aperture Personality Assessment (SAPA) data collected from 2017 to 2019 were used for the current research. The data were restricted to participants residing in the US who, had a score for each personality indicator, reported their ZIP (Zone Improvement Plan) code, and resided in a ZCTA with at least 5 total SAPA participants (N = 154,960; 8,708 ZCTAs represented). The participant count per ZCTA ranged from n = 5 to n = 284, (Mean = 17.77, Median = 13, SD = 15.76). Participants ranged from 18 to 90 years of age (Mean = 40.82, SD = 16.51, Median = 39). The Sample comprised approximately 66.2% women (n = 102,083), 33.7% men (n = 51,959), and < 1% (n = 168) participants who reported ‘other’ for sex. Protocols and data collection procedures for the SAPA Project were reviewed and approved by the Northwestern University Institutional Review Board. Informed consent was obtained from all participants prior to data collection and the methods for this study followed all relevant regulations and guidelines.

SAPA data uses a Massively Missing Completely at Random (MMCAR) approach, meaning a planned missing design50,51. is utilized to collect data on over 6000 items across very large quantities of volunteering participants. Participants completed approximately 185 items each in the 2017 to 2019 SAPA data collection (Mean = 184.78, SD = 98.03, Median = 180, Minimum items completed = 8, Maximum items completed = 336). Using the MMCAR approach means that participants were randomly given only a subset of the available SAPA project items to increase the breadth of items administered across more participants. This approach is less limiting for non-geographical analyses due to the planned missing design and the use of covariance matrices. However, there are notable limitations when using MMCAR data for geographic analysis. While administering fewer items allowed for broader geographical representation across the USA, it also meant that we did not have complete coverage of the Conservatism scale for each participant. Participants received an average score based on the items they completed, so our findings should be interpreted with this limitation in mind.

Measures

SAPA personality inventory higher-order factors (SPI-5)

The five higher order personality factors used in this study (i.e., Agreeableness, Conscientiousness, Neuroticism, Extraversion, Openness)6 were assessed using the SAPA Personality Inventory (SPI) 5 factors measure7. Each personality factor was evaluated with 14 of the SPI-135 items (i.e., 5 higher-order factors x 14 items per factor = 70 SPI items). Response options ranged from 1 to 6. The SPI-5 reliability indices and scale unidimensionality52 are presented in Table 2.

SAPA personality inventory lower-order factors (SPI-27)

The 27 lower-order interstitial personality facets used in this study (SPI-27; Condon, 2018) were assessed using the SAPA Personality Inventory (SPI) 27 lower-order factors measure7. Each personality factor was evaluated with 5 of the SPI-135 items (i.e., 27 higher-order factors x 5 items per factor = 135 SPI items). Response options ranged from 1 to 6. The SPI-27 reliability indices and scale unidimensionality are presented in Table 2.

Table 2 Reliability indices (i.e., Omega hierarchical [ωh], alpha [α], and Omega total [ωt]) and scale unidimensionality (u) for the SAPA personality inventory 5 factors and 27 interstitial Facets.
Full size table

Conservatism

The Conservatism scale for this study was obtained from the SAPA Personality Inventory (SPI) 27 facets7. Conservatism is treated as an interstitial personality facet within the SAPA framework and assessed via a 5-item scale with response options ranging from 1 to 6. Participants responded to between 1 and 6 of the scale items. The items included in the Conservatism scale are unique to the Conservatism measure in that none of these items are used to score any of the SPI-5 factors. The Conservatism scale items, reliability indices, and scale unidimensionality are presented in Table 3. Most participants (n = 61,571) completed all five Conservatism items. The mean number of items responded to by participants was 3.48, and the median number of items responded to by participants was 4 items. A histogram for the distribution of the total number of Conservatism items responded to by each participant is available in Supplementary material, Appendix A, Fig. A1.

Table 3 Conservatism scale items, general factor loadings, and participant response count per item (Scale reliabilities are ωh = 0.54, α = 0.81, ωt = 0.88, U = 0.85).
Full size table

ZIP code tabulation area (ZCTAs)

For reporting residential information, participants were simply asked to report the ZIP (Zone Improvement Plan) code that they reside in. These ZIP codes were then converted to ZIP Code Tabulation Area (ZCTAs), which are more interpretable as they group ZIP codes together into slightly larger geographic units on the same level used for most US census data. ZCTAs have better defined boundaries than ZIP codes, change less over time, and are commonly used in geographical research.

Analytic plan

Variation in the relationship for higher-order personality trait factors (SPI-5) and lower-order personality trait factors (SPI-27) with Conservatism across ZCTAs was assessed using a multilevel modeling (MLM) procedure that enabled the generation of ZCTA-specific slopes (i.e., strength and direction of the relationship between each personality indicator and Conservatism within each individual ZCTA). ZCTA identification was treated as our random variable, and intercepts and slopes to vary across ZCTAs. That is, a unique intercept (ZCTA-specific average levels of Conservatism) and unique slope (ZCTA-specific relationship between each personality factor and outcome) were generated for each ZCTA. We grand-mean centered the SPI-5 higher-order and SPI-27 lower-order factors prior to employing the MLMs, which means our findings should be interpreted in terms of whether an individual was higher or lower on a personality indicator in comparison to others in the whole sample.

All analyses and data visualizations were performed in R53. The MLMs were done with the lme444 and nlme54 packages. Psych55 and psychTools56 were used for descriptive analyses, calculating reliabilities and unidimensionality, as well as data visualizations. After running the MLMs, the ZCTA-specific slopes, which represented the unique relationship between each personality indicator and Conservatism within each individual ZCTA were correlated with the ZCTA-specific average levels of Conservatism (intercepts). These correlations were inherently adjusted for any within-cluster dependencies, sample size, and the overall fixed effect of the personality trait factor on Conservatism44. The adjusted correlations can be interpreted as the extent to which conservative sociocultural context within ZCTAs moderates the association between each personality factor and Conservatism. Notable systematic variation was found between ZCTAs in average levels of Conservatism, ranging from 1.29 to 5.70 on a six-point scale.

The calculated ICC for the multilevel structure of the Conservatism model was 0.104 indicating that about 10% of the variance in Conservatism is attributable to between ZCTA differences, suggesting a significant multilevel structure to the data. When comparing the random intercept with the random intercept + random slope models, model fit improved for each of the SPI-5 and SPI-27 indicators. That is, the AIC, BIC, and deviance statistics decreased with the inclusion of a random slope, and log likelihood increased for each of the SPI-5 and SPI-27 models. The improved fit across all multilevel models when including a random slope infers that accounting for the geographically varying associations between SPI-5 and SPI-27 traits and Conservatism more accurately reflects the structure of the observed data for each of the SPI-5 and SPI-27 traits.