IntroductionWestern, educated, industrialized, rich, and democratic (WEIRD) societies continue to be overrepresented in samples and researchers in the psychological and behavioral science1,2, but in recent years, researchers have increasingly begun collecting data with participants from less-WEIRD countries. For example, researchers have compiled many cross-national datasets to investigate cultural variation in personality3, social norms4, honesty5, values6, cognitive styles7, and fundamental social motives8. Some of these cross-national comparisons seem to suggest that cultural differences in psychological effects are small. Notably, Klein et al.'s9 Many labs 2 project attempted to test variability in psychological effects across settings by testing 125 samples from 36 countries, and concluded that there was little heterogeneity in effects between more vs. less WEIRD cultures. However, all the cross-cultural studies cited above relied heavily on educated samples, such as university students, which may underrepresent the world’s true cultural diversity. In the present study, we use the cultural fixation index (CFST)10 to investigate how cross-national cultural differences are associated with individuals’ education, income, and subjective perception of status, across 95 nations and a diverse array of cultural traits measured by the World Values Survey (WVS)11. We test whether education, income, and status specifically predict cultural values that are more similar to values endorsed in WEIRD countries. Our results reveal that higher levels of education are associated with a pattern of cultural values emblematic of the cultural traits of the United States, Western Europe, and other Anglo-heritage countries, but income and subjective social status are not.Although we do not know the causal process that explains these patterns, these results are consistent with hypotheses linking formal education to WEIRD culture, that is, those patterns of cognition, attitudes, values, and behaviors that are more common in Western nations compared to the rest of the world, such as individualism, nonconformity, analytic thinking, and generalized trust10,12,13. The transmission hypothesis argues that formal education is an institutionalized means of efficiently transmitting cultural traits14,15,16,17. Sociologists have long argued that formal education is one institution that spreads global cultural values across modern nations18. If education is indeed an efficient means of transmitting culture and has contributed to the spread of Western culture specifically, this spread will have been enhanced by truancy laws, which have required children to spend a large part of their day, throughout childhood and adolescence, in classrooms. Children spend their time not only acquiring literacy, numeracy, and discipline, but entire ways of thinking19. The curricula of many formal education systems descend from and continue to be influenced by those developed in Western nations, often instituted during periods of European colonial rule20,21,22,23. Moreover, modern educational systems also promote learning through internet-based technologies24, and thus provide people around the world with skills necessary to access online content, which leads to cultural influences beyond one’s local community. Thus, education may transmit a broad suite of attitudes, behaviors, and other psychological tendencies that are common in WEIRD countries, and make it easier for people to access cultural outputs from WEIRD countries. If this were true, we would expect more highly educated individuals to be culturally closer to WEIRD countries. In sum, education is a direct means of cultural transmission. Income and subjective social status may also contribute to cultural transmission through the ability to travel, communicate, and otherwise connect to other societies around the world. However, in the present study we found no evidence that income or subjective social status decreased cultural distance to WEIRD countries.Many researchers have argued that socioeconomic status (an emergent property of several dimensions, including a person’s educational attainment, wealth, and occupational prestige) is an important dimension of social identity that generates distinct patterns of attitudes, cognition, and behaviors that result in cultural group differences between those high in education, income, and status and those who lack socioeconomic power25,26. For example, those with higher SES have been found, relative to low SES individuals living in the same country, to hold more independent views of themselves, to place less importance on interpersonal relationships and more importance on personal freedoms, and to display more analytic patterns of thinking. These patterns have been documented across several countries, including the United States, Russia, China, as well as in multinational comparisons27,28,29,30,31,32, although with potential evidence of inverse patterns of increased other-orientation among those with high SES in Confucian cultures, such as Japan30. In longitudinal studies within the United States, change in national SES over time predicts subsequent increases in individualism33. And in cross-national studies, individuals with higher SES have lower power distance orientation and greater individualism, masculinity, and comfort with ambiguity34. Parental SES is also associated with the types of traits parents prioritize teaching to their children, such as independence, self-confidence, and uniqueness, in both the United States and in multinational investigations35,36. Unfortunately, the socioeconomic background of participants in psychological studies is rarely reported (let alone meaningfully considered) in top-tier scientific publications37.As we argued above, of all the different components of socioeconomic status, formal education may be an especially powerful pathway for the transmission of cultural values, as formal educational settings operate with the explicit goal of transmitting socially valued information, skills, cognitive styles, and modes of behavior. There is substantial evidence, primarily from studies of the United States, that educational attainment is associated with within-country variation in cultural traits. For example, there is evidence that individuals with more education display less racial biases, such as greater support for racial integration, greater awareness of discrimination and rejection of out-group exclusionism (in both the United States and internationally, e.g.,38,39,40,41,42), although this association may be limited to certain contexts or valus (e.g., see refs. 43,44). Receiving a university education in the United States also predicts scores on tests of moral reasoning styles (e.g., the defining issues test45,46) and is associated with economic attitudes that are more in line with economists’ preferences47.There is also some evidence that higher education predicts lower religious beliefs in U.S. samples, including reduced supernatural and fundamentalist Christian beliefs48,49,50, less certain belief in God51, lower belief in religious evil52, less belief in divine involvement and control in daily life53, and less paranormal belief54, although other studies have failed to find robust associations with religiosity (e.g.,55,56). Studies across multiple nations have also found that educational attainment also predicts specific attitudes, such as attitudes towards refugees57 and climate change beliefs58, as well as the likelihood of various sexual behaviors59,60, in addition to robust effects on intelligence and abstract reasoning14,15,16. These previous studies suggest that formal education is a viable pathway for shifting cultural values, and we hypothesize that the cultural values of highly educated people will reflect a pattern that is similar to the culture of WEIRD countries.This evidence indicates that there are noticeable cultural differences between low education and highly educated individuals living in the same country and suggests that these cultural differences may be visible across many countries around the world. However, existing evidence is typically limited by focusing on particular socioeconomic status differences within a single nation or on a limited set of psychological traits.Therefore, in the present study, we conducted a broader test of whether socioeconomic status is associated with a variety of different cultural traits across 95 countries around the world. We apply the cultural fixation index to document patterns of cultural variation between members of different socioeconomic groups living in each country (defined by education, income, or self-perceived status), including the specific hypothesis that people with higher education are more culturally similar to people living in the United States. Processes of colonization, globalization, and the adoption of Western-style institutions for education, religion, and politics has led to the diffusion of WEIRD cultural norms around the world13. As a result, highly educated individuals may not merely be similar to one another overall, but specifically more similar to the pattern of attitudes, traits, and behaviors of WEIRD nations, including Anglosphere countries like the United States, Canada, the United Kingdom and Australia, and Western Europe more generally.ResultsAnalysis strategyTo measure cultural similarity between highly educated groups around the world, we examined patterns of values, attitudes, and behaviors reported in the WVS. We use the generic term “cultural values” in this paper to refer to responses to all questions within the WVS that could plausibly be culturally transmitted. This includes traits, such as participants’ reported moral, political, and religious beliefs, their attitudes towards other people and group memberships, and perspectives on child-rearing, sexuality, law, and economics, which together encompasses a broad range of different behaviors and preferences.We measured cultural distance using the cultural fixation index (CFST), which captures the degree of similarity between two different groups in their responses to the set of cultural traits captured by the WVS (see description by Muthukrishna et al. and colleagues10 for further statistical information and validation). CFST reflects the difference between populations in the variance of the distribution of individuals’ responses to cultural value questions in the WVS. In population genetics, FST similarly reflects the difference in variance of the distribution of particular alleles at a particular locus of a genome, such as the variance in eye color between two populations. FST is theoretically meaningful within population genetics and evolution more generally, because it measures how genotype frequencies for each subpopulation differ from expectations if there were random mating over the entire population; that is, it measures the degree to which the populations can be considered structured and separate (for example, due to geography or mating preferences)61,62. Muthukrishna et al.10 argue that this is directly analogous to between-group differentiation caused by selection, migration, and selective social-learning mechanisms, which structure how culture is transmitted. Thus, insofar as cultural evolution offers a compelling explanation for cultural transmission and change, CFST offers a theoretically informed approach to measuring cultural distance, with a long and established history within the broader biological and evolutionary sciences.Importantly, CFST scores are able to measure differences not merely in a single cultural trait, but aggregated across the full set of different cultural value questions contained within the WVS, including traits that are independent from one another or are on different response scales (discrete and continuous). These cultural distance scores thereby index the overall magnitude of cultural differences between groups, across a range of many different cultural traits. Values of CFST can range from 0 (when the distribution of traits is identical between two populations) to 1 (when equal sized populations have completely different traits, and within-group traits are homogeneous with individuals sharing the same characteristics). Intermediate values indicate the magnitude of cultural differences accounted for by being part of different groups, with larger CFST scores indicating greater between-group differences in cultural trait distributions. Robustness tests show that CFST values are not highly sensitive to the specific questions10, and remain stable even when half the questions are selected at random, which is not surprising given the tendency for cultures to cluster63 because of transmission mechanisms, such as common sources of information and the conformist bias in cultural learning64,65,66. Prior research has used CFST to quantify the cultural distance between people living in different countries10 and with different religious affiliations67.In the present research, we apply the cultural fixation index to document patterns of cultural variation between members of different socioeconomic groups living in each country (defined by education, income, or self-perceived status). We first test whether higher education (and other indicators of SES, including income and subjective social status) is associated specifically with cultural traits emblematic of WEIRD nations. Prior to analysis, we preregistered the methods and analysis plan on the Open Science Framework (https://osf.io/d5ze2/). Our initial preregistered hypothesis, built on Muthukrishna et al. and colleagues10 and specifically predicted cultural distance from the United States, with a placebo test of cultural distance from China, across all traits in the WVS that can plausibly be considered culturally transmitted. After conducting these analyses, we amended the preregistration to also include additional robustness checks, that conducted the same analyses for each separate dimension of traits in the WVS, predicting that highly educated non-Americans will be more culturally similar to Americans across many cultural dimensions. There are two main reasons for our focus on the United States: (a) the US is still overrepresented in the psychological literature and the reference group for understanding many psychological effects, and (b) the CFST WEIRD scale of distance from the United States was shown to be most predictive of cultural and psychological differences in previous work10. However, while Hollywood and American universities have a large influence, this does not necessarily imply that higher education specifically makes people more American, but rather more similar to patterns of WEIRD cultural norms also found in other Western, Anglosphere countries.Below, we conduct further exploratory analyses using each possible country in the dataset as the reference country for comparison, to fully map for whom being highly educated predicts greater cultural similarity. This analysis further bolstered the overall finding that higher education is associated with the culture in WEIRD countries. As preregistered, we also tested whether higher income and subjective social status is associated with cultural distance from the United States, as we did not have a priori predictions that education would matter more than other facets of SES. But we focus our most comprehensive analyses below on differences based on education, as education showed the strongest association with cultural traits. We disclose all data exclusions, alternative and exploratory analyses, and measures (in the article and in the accompanying SI).Pre-registered analysis: cultural distance from the United StatesOur main analyses tested whether high education, high income, or high subjective status was associated with cultural similarity to people living in WEIRD countries. We begin by looking at cultural distance from the United States, a prototypical WEIRD country. We analyzed the average cultural distance between Americans and non-Americans with high education (i.e., greater than secondary education, such as university degrees or certificates), and compared it to the average cultural distance between Americans and non-Americans with low education (i.e., only primary education). Results indicate that, on average across all non-US countries, more highly educated people, CFST = 0.167, 95% CI [0.152, 0.182], and those with intermediate levels of education (only secondary education), CFST = 0.161 [0.147, 0.174] were more culturally similar to people in the United States, than those with low levels of education, CFST = 0.235 [0.218, 0.252].Figure 1 depicts the cultural distance from the United States among highly educated people and people with low education in each country. For example, people in Singapore with low levels of education were quite culturally distant from the United States, CFST = 0.241, but highly educated people in Singapore were much more culturally similar to the United States, CFST = 0.059, more similar in magnitude to the cultural distance between highly educated people in the Australia and the United States, CFST = 0.052. Likewise, Russians with low levels of education were very culturally distant from the United States, CFST = 0.358, whereas highly educated Russians were much more similar to the United States, CFST = 0.086, more similar in magnitude to the cultural distance between were highly educated people in New Zealand, CFST = 0.081. Looking across all countries (with n ≥ 100 of both highly educated and low-educated people for comparison), in 70% of countries, highly educated people were significantly culturally closer to the United States than people with low education in those same countries, as indicated by nonoverlapping 95% confidence intervals between the high- and low-education groups.Fig. 1: Cultural distance from the United States, based on education, income, and status.Full size imageCultural distance (CFST) from the United States among people in other countries with high levels of education and low levels of education (left), high and low levels of income (middle), and high and low levels of status (right). Higher scores indicate greater cultural differences from the United States, and smaller scores indicate greater cultural similarity. Color represents the magnitude of the difference, with red indicating countries where highly educated people are more similar to the United States, and blue indicating countries where highly educated people are more culturally different from the United States, compared to low-education people in the same country. Countries with no difference (or countries missing a high or low comparison group) are grey. Dot size corresponds to sample size within each education group.These results appear to be specific to education. Cultural distance from the United States did not significantly differ (as indicated by overlapping 95% confidence intervals) between those with high vs. low levels of income, high average CFST = 0.217 [0.199, 0.236], low average CFST = 0.199 [0.183, 0.214], or between those with high vs. low subjective social status, high average CFST = 0.182 [0.166, 0.198], low average CFST = 0.181 [0.167, 0.195].We tested the statistical significance of these effects using mixed-effects regression models that predicted cultural distances to the USA from the education group, with random intercepts for each comparison country. We also controlled for several features of the country (national population size, GDP per capita, years of education for citizens older than 15, and continent), which may predict a country’s overall similarity to the USA by virtue of also being large, rich, or highly educated overall. (Inspection of plots revealed that data generally met the assumptions of multilevel regression models, such as homogeneity of variance and normality of residuals.) These analyses (Table 1) confirmed that low-education and high-education groups differ in distance from the United States, t(80.44) = −9.11, p