Global Ecology and Geography of Gender Equality

Abstract

Proximal socio-economic drivers of gender equality tend to obscure its remote ecological origins. General systems theory predicts that the greater annual variability in daylength, temperature, and daily precipitation at higher latitudes requires greater psychosocial flexibility. We extend this prediction to gender equality as a likely consequence. Accordingly, for 87 pre-industrial societies after 1500 CE, we find more gender equality in more variable habitats, and that this link is mediated by greater subsistence flexibility—foraging rather than raising plants and animals. Mutatis mutandis, these ecological predictors of global gender equality replicate in 175 modern countries after 2000 CE. Gender equality was, and still is, lowest around the Equator, higher toward the North and South Poles, and invariant in east–west direction. The geographical positioning of gender equality in pre-industrial times can predict over 40% of the opposite north–south gradients of gender equality in the opposite Northern and Southern Hemispheres today.

Keywords

gender equality subsistence flexibility pre-industrial habitat variability

The ornithologist Cockburn (2006) has identified and located 3,529 bird species with biparental care, 529 with female-only care, and 50 with male-only care. Intriguingly, the common practice of biparental care is least common in tropical and subtropical realms (85%), increases in Nearctic and Palearctic realms toward the North Pole (90%), and peaks in Antarctica around the South Pole (98%). Similarly, in pre-industrial humans the contribution of males to parental care through provisioning was 48% in the tropics, 67% in clement climates, and 87% in the Arctic (Marlowe, 2000). These crude cross-species observations suggest that the supplementary role of males in the reproductive domain increases at higher latitudes. This adaptation mystery inspired us to explore habitational differences in degrees of gender equality in humans around the world.

Worldwide, compared to men, women usually have less access to opportunities and affordances in terms of education, work, and politics. Greater inequality of women is associated with more illness and violence, poorer education systems, lower economic growth, worse governance, and less subjective well-being (Boserup, 1970; Chaaban & Cunningham, 2011; Hudson et al., 2020; Klasen, 2002; Li et al., 2021). Consequently, achieving gender equality—defined here as valuing and institutionalizing equal access of women to resources, practices, and payoffs—has become a global policy concern (Amin et al., 2020; United Nations Development Programme, 2018; World Economic Forum, 2018). Against this background, we test an ecological explanation and a geographical description of gendered access to resources, practices, and payoffs in pre-industrial societies after 1500 CE, replicating the results in modern countries after 2000 CE, and arguing that greater habitat variability at higher latitudes facilitates greater gender equality.

The present study of global associations between ecology, geography, and gender equality concentrates on pre-industrial societies for four reasons. First, it is important to extend insight into deep historical roots of gender equality. Second, modern socio-economic drivers of gender equality—including relatively recent processes of industrialization, education, digitalization, and globalization—tend to obscure remote historical origins of gender equality. Third, in the more distant past fluctuations in daylength, climate, plant growth, and animal life likely required greater interchangeability in the subsistence roles of women and men than today, thus leaving more of an ecological imprint on gender equality. Fourth, the assumption of independent observations is easier to meet for more isolated pre-industrial societies than for more entwined contemporary societies due to less shared cultural ancestry and intercultural influence.

Theoretical Foundations

Ecology of Gender Equality

Gender equality is embedded within multiple layers of context, ranging from the family environment to the societal environment (e.g., culture, governance, economy), to the biotic habitat (animals, plants), to the abiotic habitat (e.g., day, night, cold, heat, drought, deluge). Our ecological foundation is the cross-disciplinary understanding that flexible and innovative responses are more effective in more variable environments whereas inflexible and routine responses are more effective in more stable environments (Giuliano & Nunn, 2021; Lawrence & Lorsch, 1967; Mintzberg, 1979; Schill et al., 2019). This insight derives from general systems theory (Ashby, 1958; Heylighen, 1992; Von Bertalanffy, 1968), which states that to thrive any system adjusts its internal flexibility to the variability of its external environment. Psychosocial systems generally also tend to follow this rule. Therefore, we argue that female and male members of these psychosocial systems are subject to pressures toward greater flexibility in gender roles required in more variable habitats.

This study starts to consider the impact of abiotic habitat variability on subsistence flexibility as a core aspect of psychosocial flexibility in pre-industrial times. Faced with greater seasonal variations in daylength and temperature, and greater daily variations in precipitation, pre-industrial inhabitants of higher latitudes were more dependent on the whims of nature, with the consequence that both women and men were more likely to adopt greater flexibility in daily functioning. Means by which they could develop more flexible psychosocial systems include flexibly foraging of wild plants and animals at different times and locations rather than place-bound raising and keeping of domesticated plants and animals, flexible nomadism rather than fixed sedentism, and living in small movable bands rather than large immovable tribes (Nolan & Lenski, 1999). Thus, if habitat variability drives subsistence flexibility, pre-industrial peoples will have evolved more nomadic gathering and hunting in smaller bands at higher latitudes with greater habitat variability, but more sedentary raising of plants and animals in larger tribal groups at lower latitudes with greater habitat stability.

We cumulatively propose that smaller scale relocation and restructuring of gathering and hunting activities can be easier maintained and improved when group members fulfill more equal and thus more interchangeable and flexible roles. Greater access of women to resources (e.g., property, information, skills) is thought to qualify them to help sustain flexible societal structures, processes, and strategies. Greater access to practices (e.g., tasks, acts, evaluations) is thought to enable them to better flexible societal arrangements. And greater access to payoffs (e.g., wages, favors, prestige) is thought to motivate them to sustain and improve the achieved societal flexibility. Compared to pre-industrial women in sedentary agricultural societies, pre-industrial women in nomadic foraging societies are therefore expected to own and earn more, know more, perform more; have more rights, obligations, and influence; and yield more respect, trust, support, and cooperation.

The empowerment of pre-industrial women in more variable habitats with greater subsistence flexibility will have gradually led to changes toward more female participation in public arenas, toward more authority positions, and even away from patrilineal toward matrilineal rule of descent—that is, toward greater gender equality. By comparison, the required coordination of subsistence cycles of plowing, irrigating, harvesting, animal husbandry, and marketing in more stable habitats is thought to have been easier to realize in larger communities embracing more traditional gender roles of women as housewives and men as breadwinners (Alesina et al., 2013; Hansen et al., 2015; Iversen & Rosenbluth, 2010). In summary, theoretical reasoning suggests the following hypothesis: In pre-industrial societies, greater habitat variability is associated with greater flexibility in subsistence-related strategies, which in turn is associated with greater gender equality.

Geography of Gender Equality

The proposed ecological explanation provides not only explicit information about the positive association between habitat variability and gender equality, but also implicit information about the geographical distribution of gender equality. Seasonal daylength variability, seasonal temperature variability, and daily precipitation variability increase at higher latitudes, and vary negligibly with longitude (Hut et al., 2013; Lisovski et al., 2017; Van de Vliert & Van Lange, 2019). It therefore follows that, if habitat variability and gender equality naturally coincide, gender equality would increase from the Equator toward the North and South Poles rather than from the Greenwich meridian to the east or the west. The Equator here should not be understood as the exact Equator (0°) because prior research has documented near-Equatorial reversal points of 6° N for average temperature (Aschoff, 1981) and 8° N for daily precipitation variability (Van de Vliert & Van Lange, 2019). We therefore predicted a U-curve distribution of pre-industrial gender equality between the North and South Poles.

Although this U-curve implies a reversal of the north–south gradient of pre-industrial gender equality, the precise location of the predicted near-Equatorial reversal point is left to be identified. In common with the near-Equatorial reversal points for habitat variability and subsistence flexibility, the near-Equatorial reversal point for gender equality can be computed as X_m + (−b₁ / 2b₂), where X_m = the mean midrange latitude of the sampled societies, b₁ = the regression coefficient for regressing the targeted variable on the centered linear measure of latitude (X), and b₂ = the regression coefficient for regressing the targeted variable on latitude-squared (X²). We note that support for opposite north–south gradients of pre-industrial gender equality in the opposite Northern and Southern Hemispheres provides a spatial validation of the ecological explanation as well as psychologically interesting information about poleward inclinations of gender equality.

Ecology of Pre-Industrial Gender Equality

Ethnographers have categorized pre-industrial societies after 1500 CE as more male-dominated (e.g., Aymara, Ganda, Kapauku, Otoro) versus more gender-egalitarian (e.g., Klamath, Lapps, Yahgan, Yurok) cultural systems. The investigated societies are listed in Supplemental Table S1. The first analysis sought to answer the following question: Was greater habitat variability directly associated with greater subsistence flexibility, and indirectly with more equal access of women and men to resources, practices, and payoffs? To avoid confirmation bias (Klayman & Ha, 1987), we tested alternative predictors—causal or confounding factors—that might fit the explanatory evidence. The source data and codes for all analyses are available at https://doi.org/10.34894/LALJIK.

Sample

Data on pre-industrial gender equality were available for 87 societies of the Standard Cross-Cultural Sample of pre-industrial cultures (Murdock & White, 1969; Ross, 1983). This sample size can only detect direct and indirect effects of at least medium size (more than 13% of the variance) with 80% power assuming a two-sided test and a type I error level of .05 (Fritz & MacKinnon, 2007). Because the Standard Cross-Cultural Sample contains only one society from each world area, it maximizes world representativeness, minimizes societal relatedness because of geographical proximity (Bromham et al., 2018; Willig et al., 2003), and reduces violations of the statistical assumption of independent observations.

The potential sampling problem of dependence between the society-level observations (Mace & Holden, 2005) was indeed minimized as evidenced by the following information from the phylogenetic classifications of languages (source: Kirby et al., 2016): 26 language families in the Northern Hemisphere were absent from the Southern Hemisphere and 8 language families in the Southern Hemisphere were absent from the Northern Hemisphere (Supplemental Table S1). As a consequence, it is unlikely that flexibility-mediated associations between habitat variability and gender equality are due to spatially or culturally biased sampling of societies.

Outcome Measure: Gender Equality

As detailed below, using ethnographic reports, Ross (1983) coded the participation of pre-industrial women and men in public arenas, as well as the equal appointment of women to public offices. Similarly, Murdock and White (1969) coded pre-industrial societies’ gender equality in the rule of descent. After excluding three societies with only a single gender-equality code (Cayapa, Miskito, Suku), there remain 87 societies, 77 with three codes and 10 with two codes, the average of which served as the outcome variable reported in Supplemental Table S1.

Participation in public arenas: 1 = women generally excluded; 2 = some role for women; 3 = significant role for women but more participation of men; 4 = participation of women equal to or greater than that of men. Appointment to positions of authority: 1 = positions only open to men; 2 = positions rarely, if ever, held by women; 3 = positions occasionally held by women; 4 = positions commonly held by women (although men are eligible). Gender-egalitarian rule of descent: 1 = patrilineal society (n = 40); 2 = patrilineal dominant and matrilineal subordinate (Manus, Yapese) or quasi-patrilineal (Havasupi); 3 = non-lineal or ambilineal (n = 34); 4 = matrilineal (n = 10). The combined mean index of gender equality (eigenvalue λ = 1.958, R² = 0.653; Cronbach’s α = 0.727) approximates normality (M = 2.059, SD = 0.888; skewness = 0.231, SE = 0.258; kurtosis = −1.291, SE = 0.511).

Principal Predictors

Descriptive Predictors: Latitude and Longitude

To be able to estimate habitat variability (next paragraph), the societal locations were measured on the north–south coordinate of midrange latitude (negative below the Equator and positive above it) and the east–west coordinate of midrange longitude (negative west of the Greenwich meridian and positive east of it; source: Murdock & White, 1969; Supplemental Table S1).

Explanatory Predictor: Habitat Variability

Pre-industrial variabilities in daylength, temperature, precipitation, plant growth, and animal life are unknown. Given that the worldwide distribution of climate has negligibly changed since 1500 CE (Burroughs, 1997; Ditlevsen et al., 1996), past habitat variability in the 87 sampled societies was estimated on the basis of the known habitat variability in 175 contemporary countries. For each country, Supplemental Table S2 lists seasonal daylength variability, seasonal temperature variability, and daily precipitation variability as underlying components of overall habitat variability (eigenvalue λ = 2.268, R² = 0.756; Cronbach’s α = 0.836). Supplemental Table S3 additionally lists the location of each country in terms of midrange latitude and midrange longitude. Nowadays, latitude-squared (B_[171] = 0.536, p < .001, confidence interval [CI] = [0.499, 0.574]), latitude-linear (B_[171] = 0.665, p < .001, CI = [0.622, 0.708]), and longitude-linear (B_[171] = 0.005, p = .853, CI = [−0.046, 0.055]) account for 90% of the worldwide variation in habitat variability.

In Supplemental Table S2, this regression equation is broken down for today’s geographical positioning of seasonal daylength variability, seasonal temperature variability, and daily precipitation variability, respectively. Applying the three equations to the latitudes and longitudes of pre-industrial societies resulted in three estimates of variabilities in daylength, temperature, and precipitation Supplemental (Supplemental Table S1). The three virtually identical estimates yielded a reliable index of overall habitat variability in pre-industrial societies (eigenvalue λ = 2.944, R² = 0.981; Cronbach’s α = 0.982; Supplemental Table S1). The validity of this measure of pre-industrial habitat variability was reflected in its U-curve distribution between the North and South Poles, and the reversal of its north–south gradient at 1°20′ north of the Equator (Figure 1).

Figure 1.

Latitudinal Distributions of Habitat Variability, Foraging, Nomadism, and Small Community in 87 Pre-Industrial Societies.

Mediators: Subsistence Flexibility

As argued above, plausible mediators between habitat variability and gender equality are flexible strategies of subsistence. As reasoned and detailed in Supplemental Table S1, flexibility was measured by estimating foraging, nomadism, and small community to enable nomadism. The nominal data on foraging (Murdock & White, 1969) were recoded into an ordinal variable ranging from 1 = exclusively raising domesticated plants and animals to 4 = exclusively foraging wild plants and animals. Nomadism was represented by Murdock and Wilson’s (1972) 6-point scale ranging from 1 = permanent settlement to 6 = nomadic or migratory community. Finally, small community (Murdock & Wilson, 1972) ranged from 1 = 50,000 persons or more to 8 = fewer than 50 persons. Foraging, nomadism, and small community are overlapping yet distinct variables (0.449 < r_[85] < 0.556, ps < .001), all with increased values toward the North and South Poles (Figure 1).

Alternative Predictors

Pathogen Prevalence

The burden of infectious diseases may account for the results, in whole or in part, because it leads to slower life-history strategies and greater gender inequality (Varnum & Grossmann, 2017), and decreases toward the North and South Poles (Van de Vliert, 2020; Van de Vliert & Van Lange, 2019). Pre-industrial pathogen prevalence, retrieved from Cashdan and Steele’s (2013) study, was represented by eight infectious diseases: leishmaniasis, trypanosomes, malaria, schistosomes, filariae, dengue, typhus, and plague (Cronbach’s α = 0.760).

Food Insecurity

Famine crises, such as the Great Irish Famine (1845-1852) and the Chinese Great Leap Forward Famine (1959-1961), tend to increase gender inequality because of higher female mortality rates and son preferences (Mu, 2019). Therefore, occurrence of famine, severity of famine, and persistence of famine in pre-industrial societies, coded by Dirks (1993), were averaged (Cronbach’s α = 0.906) and then analyzed as a potential cause of gender inequality.

Population Density

Flexibility in subsistence—the mediating variable in this study—is counteracted by the size-based density of the population (Nolan & Lenski, 1999). To address this confounding influence, we used a 7-point measure from Murdock and Wilson’s (1972) study. Population density varied from 0.6 person per square mile on average in purely foraging societies (1) to more than 100 on average in purely agrarian societies (7; M = 3.760, SD = 2.040; skewness = 0.118, SE = 0.260; kurtosis = −1.272, SE = 0.514).

Statehood

In pre-industrial times, the micropolitics about female and male control of resources, practices, and payoffs was heavily intertwined with the distributive politics at overarching levels of sovereignty (Coltrane, 1992). This complication made it necessary to take into account the number of levels of statehood: 1 = stateless society; 5 = sovereignty third or higher hierarchical level (source: Tuden & Marshall, 1972).

Religiosity

As another causal confound, “what we have seen throughout human history . . . is religion by and large upholding the subordination of women, particularly at the household level” (Hudson et al., 2020, pp. 369-370), and especially where God is masculine and powerful (Hofstede, 1999). A related proxy for measuring masculine religiosity is the high god variable (Murdock, 1967): 1 = absent or not reported; 2 = present but not active in human affairs; 3 = present and active in human affairs but not supportive of human morality; 4 = present, active, and specifically supportive of human morality.

Modernization

Cultural changes from traditional mindsets and practices toward inventive, innovative, and socio-economic progress tend to express themselves in greater gender equality (Inglehart & Norris, 2003), including equal freedom in embracing gendered preferences (Falk & Hermle, 2018). We therefore included Divale and Seda’s (2000) index of pre-industrial modernization as an alternative predictor that covers 33 changes in, for example, agriculture, technology, trade, transportation, governance, family organization, communication, and education (Cronbach’s α = 0.677).

Income Inequality

Inequal distributions of payoffs may be a normal cultural fact in need of attention because gender inequality is strongly confounded with overall income inequality (www.imf.org/external/pubs/ft/sdn/2015/sdn1520_info.pdf; Hudson et al., 2020). Income inequality in pre-industrial times increased to the extent that there were not only more rich people (1 = absence; 2 = presence, no percentage information; 3 = few; 4 = many), but also more poor people (1 = absence; 2 = presence, no percentage information; 3 = few; 4 = many) and more dispossessed people (1 = absence; 2 = presence, no percentage information; 3 = few; 4 = many) (source: Lang, 1998). Accordingly, we controlled for the average of these three indicators of income inequality in the sampled societies (Cronbach’s α = 0.814).

Aggression

Male domination is confounded with aggression, motivating us to also control for internal violence and external warfare (retrieved from Ross, 1983). Internal violence was measured with seven codes for the frequency and level of local conflict, the acceptability of violent reactions, the actual use of physical force, and the like (Cronbach’s α = .900). External warfare, represented by codes for the frequency of external conflict, hostility toward other societies, and overt fighting with them (Cronbach’s α = .868), was only weakly related to internal violence (r_[84] = 0.386, p < .001).

Results

Confirmatory Testing

Table 1 shows the expected positive correlations among pre-industrial gender equality, habitat variability, foraging, nomadism, and small community, with the marginal note that latitude-linear represses the association of gender equality and habitat variability (from r_[84] = 0.327, p < .001 to r_[85] = 0.205, p < .05). The proposed ecological explanation was tested by regressing gender equality on the proposed mediators of subsistence flexibility (foraging, nomadism, and small community) and the ultimate predictor of habitat variability. Hayes’s (2018) path analysis Template 81 was set to report results with heteroskedasticity-corrected standard errors. Moreover, the macro was set up to draw 5,000 bootstrap samples for constructing bias-corrected 95% CIs.

Table 1.

Correlations Among Main Study Variables Across 87 Pre-Industrial Societies.

Variable	1	2	3	4	5	6	7
1. Gender equality
2. Habitat variability	0.205*
3. Foraging	0.500***	0.424***
4. Nomadism	0.288**	0.293**	0.556***
5. Small community	0.368***	0.337***	0.504***	0.449***
6. Latitude-linear (z)	−0.060	0.664***	0.176	0.118	0.090
7. Latitude-squared	0.329**	0.728***	0.406***	0.285**	0.367***	−0.030
8. Longitude-linear (z)	−0.110	−0.232*	−0.214*	−0.236*	−0.045	−0.135	−0.193

p < .05. **p < .01. ***p < .001.

As can be seen in Table 2, the total effect of habitat variability on gender equality (R² = 0.270, p < .001) is not a direct effect. Instead, habitat variability has direct effects on each of the three manifestations of subsistence flexibility. Foraging, nomadism, and small community, in turn, are associated with gender equality, and thus standing out as potential mediators. Although five mediation paths were tested, the only path with a significant CI (effect = 0.168, CI = [0.048, 0.374]) was from habitat variability through foraging to gender equality. In support of the ecological hypothesis, subsistence flexibility in the form of gathering and hunting seems to have linked habitat variability to pre-industrial gender equality.

Table 2.

The Statistical Path From Habitat Variability (HV) Through Foraging to Gender Equality Across 87 Pre-Industrial Societies.

Total effect: R² = 0.270 (p < .001)	Effect	SE	p	95% CI
Direct effects
HV → GE (gender equality)	−0.029	0.102	.777	[−0.233, 0.174]
HV → FO (foraging)	0.544	0.161	<.001	[0.225, 0.864]
HV → NO (nomadism)	0.605	0.167	<.001	[0.274, 0.936]
HV → SC (small community)	0.546	0.130	<.001	[0.287, 0.806]
Indirect effects
HV → FO → GE	0.168	0.085		[0.048, 0.374]^a
HV → NO → GE	−0.001	0.013		[−0.033, 0.025]
HV → SC → GE	0.022	0.023		[−0.012, 0.078]
HV → FO → NO → GE	−0.005	0.025		[−0.052, 0.050]
HV → FO → SC → GE	0.028	0.022		[−0.010, 0.076]

Note. SE = standard error; CI = confidence interval

Based on 5,000 bootstrap samples with 95% bias-corrected CIs. No p-value is provided because the indirect effect is quantified as the product of paths.

Testing Alternative Predictors

As fully reported in the Supplemental Material, pathogen prevalence (Supplemental Table S4; n = 87), food insecurity (Supplemental Table S5; n = 85), population density (Supplemental Table S6; n = 86), statehood (Supplemental Table S7; n = 87), religiosity (Supplemental Table S8; n = 82), modernization (Supplemental Table S9; n = 75), income inequality (Supplemental Table S10; n = 55), internal violence (Supplemental Table S11; n = 86), and external war (Supplemental Table S12; n = 86) did not have any significant effect on pre-industrial gender equality over and above the mediation effect of foraging between habitat variability and gender equality. Table 3 provides the combined results for alternative predictors available for 79 pre-industrial societies.

Table 3.

Effects of Habitat Variability, Foraging, and Alternative Predictors on Gender Equality in 79 Pre-Industrial Societies.

Predictors	Total effect: R² = 0.339 (p < .001)
Predictors	B	95% CI	p
Habitat variability	0.005	[−0.250, 0.260]	.968
Foraging	0.257	[0.054, 0.460]	.014
Pathogen prevalence^a	−0.218	[−0.629, 0.194]	.294
Food insecurity^b	−0.028	[−0.238, 0.182]	.792
Population density^c	0.013	[−0.121, 0.147]	.849
Statehood^d	−0.003	[−0.219, 0.213]	.979
Religiosity^e	−0.088	[−0.244, 0.068]	.263
Internal violence^f	−0.063	[−0.002, 0.129]	.056
External war^g	−0.118	[−0.257, 0.020]	.093

Note. CI = confidence interval.

The prevalence of leishmaniasis, trypanosomes, malaria, schistosomes, filariae, dengue, typhus, and plague (Cronbach’s α = 0.760). For single-factor model, see Supplemental Table S4.

The occurrence, severity, and persistence of famine (Cronbach’s α = 0.906). For single-factor model, see Supplemental Table S5.

The number of persons per square mile. For single-factor model, see Supplemental Table S6.

The number of levels of statehood. For single-factor model, see Supplemental Table S7.

The extent to which a high god is active in human affairs. For single-factor model, see Supplemental Table S8.

The frequency and level of local conflict, the acceptability of violent reactions, and the actual use of physical force (Cronbach’s α = .900). For single-factor model, see Supplemental Table S11.

The frequency of external conflict, hostility toward other societies, and overt fighting with them (Cronbach’s α = .868). For single-factor model, see Supplemental Table S12.

A slight concern was that the results were somewhat inflated by the fact that 28 out of the 87 societies belonged to three dominant language families (Supplemental Table S1: Austronesian 13, Atlantic-Congo 9, Afro-Asiatic 6). To address this concern, the Supplemental Material also shows results when assigning lower weights to the societies with higher linguistic relatedness (source: Kirby et al., 2016). Across the 81 societies with weights for linguistic relatedness, phylogenetic dependence had no distorting influence on the mediatory role of foraging between habitat variability and gender equality (see Supplemental Tables S4-S12). On the contrary, the foraging-mediated impact of habitat variability increased rather than decreased when controlling for the societies’ linguistic relatedness. Specifically, the unweighted models predicted significantly lower proportions of the variation in gender equality (mean R² = 0.241) than did the weighted models (mean R² = 0.329; z = −2.666, p = .008). Thus, our ecological hypothesis has been tested in a conservative fashion.

Geography of Pre-Industrial Gender Equality

The second analysis tested the habitat-based prediction that pre-industrial gender equality had a U-curve distribution between the North and South Poles, while varying negligibly in east–west direction. The sample, the measure of gender equality, and the measures of latitude and longitude were the same as in the preceding analysis. For reasons clarified above, we standardized and squared midrange latitude, used latitude-linear to control for confounding differences between the Northern and Southern Hemisphere including the larger landmasses toward the North Pole, used latitude-squared to capture curvilinearity, and used both latitude-linear and latitude-squared to estimate the near-Equatorial reversal point of the north–south gradient of pre-industrial gender equality. Longitude-linear represented the unidirectional longitude scale with the Greenwich meridian as an imaginary and meaningless midpoint. Zero-order correlations are reported in Table 1.

As predicted, gender equality increased toward the North and South Poles (R² = 0.114; B_[83] = 0.208, p = .004, CI = [0.070, 0.347] for latitude-squared), doing so to roughly the same extent (B_[83] = −0.056, p = .545, CI = [−0.238, 0.127] for latitude-linear), while varying negligibly with longitude (B_[83] = −0.051, p = .591, CI = [−0.239, 0.137]). This U-curve distribution of gender equality reverses its north–south gradient at 17°21′ north of the Equator. However, a detailed visualization of the results (Figure 2a) shows steep southward increases below −10° S (r_[11] = −0.603, p = .029) and steep northward increases above 25° N (r_[27] = 0.492, p = .007). By contrast, within that zone between −10° S and 25° N, gender equality decreases rather than increases toward the North and South Poles (R² = 0.162; B_[42] = −2.033, p = .030, CI = [−3.863, −0.202] for latitude-squared; B_[42] = −1.796, p = .007, CI = [−3.072, −0.520] for latitude-linear). This is a point for later discussion: Why did pre-industrial societies have a small peak in gender equality at 6°46′ north of the Equator?

Figure 2.

Latitudinal Distributions of Gender Equality in (a) 87 Pre-Industrial Societies and (b) 175 Contemporary Countries.

Ecology of Current Gender Equality

Replicated Outcome Measure: Gender Equality

To investigate whether the tentative results for pre-industrial societies can be replicated in modern countries, we constructed a measure of current gender equality in health, education, rights and liberties, economic and political participation, salaried employment, and life satisfaction. Specifically, we averaged six standardized indices of inequality versus equality (detailed in Table 4; eigenvalue λ = 4.296, R² = 0.716; Cronbach’s α = 0.917). As can be seen in Supplemental Table S3, this 175-country index ranges from −2.30 (Yemenis) and −2.25 (Chadians) for gender inequality to 1.46 (Finns) and 1.53 (Icelanders) for gender equality (M = −0.103, SD = 0.857; skewness = −0.430, SE = 0.184; kurtosis = −0.644, SE = 0.365). We retained large countries (e.g., Australia, Brazil, Canada, China, India, Russia, and the United States) because weighting for larger country size and larger population size had negligible effects on the results.

Table 4.

Global Assessments of Current Gender Inequality Versus Equality.

Indices	N ^a	Aspects of differential access of women and men to resources, practices, and payoffs
Gender inequality
World Economic Forum (2018)	151	Health, educational attainment, economic participation, political empowerment
Stoet and Geary (2019)	134	Healthy life span, educational attainment, literacy rate, overall life satisfaction
Sullivan (1991)	101	Life expectancy, marriage breakdown, literacy rate, salaried employment, freedom from discrimination
Gender equality
United Nations Development Programme (2018)	162	Educational attainment, labor force participation, parliamentary seats
Organisation for Economic Co-operation and Development (2019)	121	Non-discriminatory laws, physical protection, daughter–son equivalence, civil rights and liberties
Social Watch (2012)	156	Educational attainment, economic activity, socio-economic empowerment
Aggregate index	175	Overall gender equality (Cronbach’s α = 0.917)

N is number of contemporary countries.

To reproduce the objective nature of the indicators of gender equality in pre-industrial societies, the country indices in Table 4 are not based on self-reports of experienced gender equality today. However, there is reason to assume that our objective index also represents subjective mindsets. The seventh wave of the World Values Surveys measured how strongly a country’s inhabitants believe that (a) “education is more important for a boy than a girl,” (b) “when jobs are scarce, men should have priority over women to get a job,” and (c) “men make better political leaders than women.” The resulting subjective index of current gender equality (retrieved from https://www.worldvaluessurvey.org/WVSDocumentationWV7.jsp) is strongly associated with the objective index of current gender equality used here (r_[84] = 0.783, p < .001, CI = [0.685, 0.853]).

Replicated Principal Predictor: Habitat Variability

As already discussed, seasonal daylength variability, seasonal temperature variability, and daily precipitation variability were combined into a reliable 175-country index of overall habitat variability (Supplemental Table S2), which served as explanatory predictor.

Replicated Mediators: Subsistence Flexibility

There is no modern replica of the pre-industrial variable of place-bound raising and keeping of domesticated plants and animals versus flexibly foraging of wild plants and animals. The best available estimate of subsistence flexibility is each country’s present position on the historical continuum from sedentary agricultural practices to more flexible industrial and service activities. Proxies are the national percentages of employment in the agrarian sector (arable farming, animal farming, and fishing), the industrial sector (manufacturing, mining, building, and public utilities), and the service sector (trade, transport, restaurants, hotels, finances, communications, and community and personal services). These percentages, retrieved from the World Factbook (https://www.cia.gov/the-world-factbook/field/labor-force-by-occupation), were factored into a single index of increasingly flexible subsistence activities (Supplemental Table S2; eigenvalue λ = 2.128, R² = 0.709; Cronbach’s α = 0.789).

Replicated Alternative Predictors

Pathogen Prevalence

To test whether infectious disease threats may account for the results, the current burden of human-to-human transmitted diseases (e.g., measles, cholera, hookworm, leishmaniasis, leprosy, dengue fever) was retrieved from Fincher and Thornhill’s (2012) study (Cronbach’s α = 0.760).

Food Insecurity

As second potential cause of current gender equality, a modern version of the occurrence and severity of famine in pre-industrial times—the World Bank’s rating of food insecurity in the population—was harvested from https://data.worldbank.org/indicator/SN.ITK.MSFI.ZS.

Population Density

As possible confounding factor, the number of country inhabitants per square kilometer, log-transformed to reduce skewness, was available from Parker’s (1997) study.

Statehood

The next alternative predictor—the difference today between a non-governed and a well-governed country—was captured by combining the World Bank’s indicators of government effectiveness, regulatory quality, and rule of law (https://info.worldbank.org/governance/wgi; Cronbach’s α = 0.977).

Religiosity

The extent to which a god is present and active in human affairs by facilitating the subordination of women was derived from the Gallup World Polls. Gallup asked inhabitants of 159 countries whether religion is or is not an important part of their daily life (https://www.gallup.com/analytics/318923/world-poll-public-datasets.aspx). The national percentage of “Yes” responses was used to represent the extent of religious belief.

National Wealth

Because Divale and Seda’s (2001) socio-economic modernization in pre-industrial societies covered 33 changes in societal functioning, it cannot be meaningfully replicated in contemporary countries. Instead, we used national wealth as a well-known proxy determinant of female empowerment and gender equality (Inglehart & Norris, 2003), operationalizing it as the average log-transformed income per head in international dollars at the beginning of this century (United Nations Development Programme, 2002, 2004, 2006; Cronbach’s α = 0.998; for country scores, see Supplemental Table S2).

Income Inequality

The extent to which the distribution of income among individual households within a country deviates from a perfectly equal distribution was downloaded from https://en.wikipedia.org/wiki/List_of_countries_by_income_equality (ultimate sources: the World Bank and the World Factbook).

Aggression

Internal violence was measured as a conglomerate of domestic warfare and violence, press repression, and business costs of aggressive crime and violence (borrowed from Van de Vliert and Van Lange [2019]; Cronbach’s α = 0.702). There is no modern replica for external warfare in pre-industrial times, which was always local and never on an intercontinental scale.

Results

Confirmatory Testing

Mirroring the results in Table 1, there are positive associations among current gender equality, habitat variability, and our proxy of current subsistence flexibility (Table 5). Subsequent mediation analysis (Hayes, 2018; Template 4) suggested the existence of two effects (R² = 0.587): a direct path from habitat variability to gender equality (effect = 0.441, p < .001; CI = [0.350, 0.531]), and an indirect path from habitat variability through subsistence flexibility to gender equality (effect = 0.220, CI = [0.152, 0.298]; see the “Confirmatory testing” row in Table 6). In support of the ecological hypothesis, subsistence flexibility in the form of less agriculture still seems to help link together habitat variability and gender equality today.

Table 5.

Correlations Among Main Study Variables Across 175 Contemporary Countries.

Variable	1	2	3	4	5
1. Gender equality
2. Habitat variability	0.669***
3. Subsistence flexibility: less agriculture	0.664***	0.513***
4. Latitude-linear (z)	0.343***	0.660***	0.419***
5. Latitude-squared (z-squared)	0.503***	0.588***	0.244***	−0.132
6. Longitude-linear	−0.114	0.066	−0.063	−0.083	−0.001

***

p < .001.

Table 6.

Controlling the Statistical Paths From Habitat Variability to Gender Equality for Alternative Predictors.

Predictors	Total effect			Principal predictors^b
	Total effect			Direct path: HV → GE		Indirect path: HV → SF → GE
	N ^a	R ²	p	Effect	95% CI	Effect	95% CI
Principal predictors^b
Confirmatory testing	175	0.587	<.001	0.441	[0.350, 0.531]	0.220	[0.152, 0.298]
Alternative predictors^c
Pathogen prevalence^d	175	0.597	<.001	0.382	[0.282, 0.483]	0.052	[0.006, 0.109]
Food insecurity^e	139	0.687	<.001	0.367	[0.263, 0.471]	0.040	[0.000, 0.090]
Population density^f	174	0.590	<.001	0.453	[0.357, 0.548]	0.217	[0.143, 0.295]
Statehood^g	173	0.656	<.001	0.309	[0.216, 0.402]	0.057	[0.019, 0.106]
Religiosity^h	159	0.627	<.001	0.201	[0.068, 0.335]	0.063	[0.012, 0.132]
National wealthⁱ	175	0.631	<.001	0.372	[0.277, 0.467]	0.010	[−0.004, 0.043]
Income inequality^j	161	0.633	<.001	0.542	[0.429, 0.656]	0.229	[0.149, 0.319]
Aggression^k	166	0.619	<.001	0.380	[0.283, 0.477]	0.136	[0.075, 0.210]
All alternative predictors predictors	123	0.803	<.001	0.203	[0.060, 0.346]	−0.001	[−0.016, 0.016]

Note. HV = habitat variability; GE = gender equality; SF = subsistence flexibility; CI = confidence interval.

N is number of contemporary countries.

The principal predictors are HV and SF. SF is represented by less agriculture, that is, more industrial and service activities.

The direct path HV → GE and the indirect path HV → SF → GE are controlled, first for single alternative predictors, and then for these alternative predictors in combination.

The number of cases of human-to-human transmitted diseases (r_[173] = −0.640, p < .001 with GE).

The World Bank’s rating of food insecurity in the population (r_[137] = −0.728, p < .001 with GE).

The log-transformed number of country inhabitants per square kilometer (r_[172] = 0.109, p = .154 with GE).

The World Bank’s rating of government effectiveness, regulatory quality, and rule of law (r_[171] = 0.728, p < .001 with GE).

The extent to which religious belief is an important part of daily life (r_[157] = −0.726, p < .001 with GE).

The average log-transformed income per head in international dollars (r_[173] = 0.722, p < .001 with GE).

The extent to which household incomes are unequally distributed (r_[159] = −0.178, p = .024 with GE).

Domestic violence, press repression, and business costs of aggressive crime and violence (r_[164] = −0.561, p < .001 with GE).

Testing Alternative Predictors

Repeating the above mediation analysis while controlling for pathogen prevalence, food insecurity, population density, statehood, religiosity, national wealth, income inequality, and aggression hardly changed the results. As reported in Table 6, the direct and indirect paths from habitat variability to gender equality are robust to controlling for these alternative predictors, with only one exception. Higher national wealth overrides the mediating impact of more subsistence flexibility (less agriculture) on greater gender equality, which seems to indicate that national wealth is an independent cause of current gender equality.

An alternative interpretation of this result is that contemporary countries with less agriculture (more industrial and service activities) have developed higher national wealth, which then serves as an extra mediator between habitat variability and gender equality. In line with this speculation, a supplementary analysis with two mediators (Hayes, Template 6; Table 7) revealed three effects (R² = 0.631): (a) a direct path from habitat variability to gender equality (effect = 0.372, CI = [0.277, 0.467]), (b) a short indirect path through national wealth (effect = 0.069, CI = [0.023, 0.138]), and (c) a complete path from habitat variability through subsistence flexibility and then via national wealth to gender equality (effect = 0.146, CI = [0.075, 0.230]). These statistical paths suggest that, in addition to subsistence flexibility, national wealth also helps link habitat variability to gender equality.

Table 7.

Extended Statistical Paths From Habitat Variability to Gender Equality Across 175 Contemporary Countries.

Total effect: R² = 0.631 (p < .001)	Effect	SE	p	95% CI
Direct effects
HV → GE	0.372	0.048	<.001	[0.277, 0.467]
HV → SF^a	0.591	0.056	<.001	[0.480, 0.702]
HV → NW	0.643	0.065	<.001	[0.516, 0.770]
Indirect effects
HV → SF → GE	0.074	0.052		[−0.028, 0.177]^b
HV → NW → GE	0.069	0.029		[0.023, 0.138]
HV → SF → NW → GE	0.146	0.039		[0.075, 0.230]
HV → NW → SF → GE	0.064	0.043		[−0.023, 0.148]

Note. HV = habitat variability; GE = gender equality; SF = subsistence flexibility; NW = national wealth; SE = standard error; CI = confidence interval.

SF is represented by less agriculture, that is, more industrial and service activities.

Based on 5,000 bootstrap samples with 95% bias-corrected CIs. No p-value is provided because the indirect effect is quantified as the product of paths.

Geography of Current Gender Equality

Specifying the correlations in Table 5, latitude-squared (R² = 0.307; B_[171] = 0.431, p < .001, CI = [0.343, 0.518]), latitude-linear (R² = 0.180; B_[171] = 0.376, p < .001, CI = [0.276, 0.475]), and longitude-linear (R² = 0.022; B_[171] = −0.154, p = .010, CI = [−0.271, −0.037]) accounted for 44% of the variation in contemporary gender equality. Just like pre-industrial gender equality (Figure 2a), current gender equality (Figure 2b) appears to be lowest around the Equator, increases toward the North and South Poles, and varies statistically significant but theoretically negligible with longitude (accounting for only 2.2% of the variation). Just like pre-industrial gender equality (17°21′ N), current gender equality (7°06′ N) has its near-Equatorial reversal point north of the Equator.

The geographical positioning of gender equality in pre-industrial times enabled us to forecast the geographical positioning of gender equality in the modern era. Specifically, the spatial regression equation for pre-industrial gender equality (Y = [0.208] × [z-latitude × z-latitude] + [−0.056 × z-latitude] + [−0.051 × z-longitude]) was applied to the latitudes and longitudes of the 175 contemporary countries. The forecasted levels of current gender equality (Supplemental Table S3) were then shown to be associated with the measured levels of current gender equality (Supplemental Table S3) in both the Northern Hemisphere (r_[137] = 0.638, p < .001) and Southern Hemisphere (r_[34] = 0.666, p < .001). These results reflect clear support for the ecological explanation and long-term prediction of current gender equality.

Discussion

The existence of latitudinal gradients of biparental care in birds (Cockburn, 2006) and pre-industrial humans (Marlowe, 2000) inspired us to explore the global ecology and geography of past and present gender equality. As a first finding, habitat variability seems to qualify as a distal ecological predictor of gender equality. Strengthening this finding is the fact that industrialization, education, digitalization, globalization, and other modern socio-economic drivers of gender equality cannot have caused gender equality in pre-industrial times. Second, flexible foraging by gathering wild plants and hunting wild animals appears to stand out as a mediator between habitat variability and pre-industrial gender equality. The mediating function of subsistence flexibility in the past seems to be consolidated by two mediators between habitat variability and current gender equality: subsistence flexibility in the form of less agriculture, and national wealth enabling flexibility in how to subsist and thrive.

Responding to calls for enriching explanatory modeling with more predictive modeling (Hofman et al., 2021; Yarkoni & Westfall, 2017) yielded four more insights. In agreement with the ecological explanation, pre-industrial inhabitants of higher latitude habitats tended to have stronger habits of subsistence flexibility and gender equality (Figures 1 and 2). In further agreement with the explanation, the pre-industrial north–south distributions of habitat variability, flexible foraging, and gender equality share near-Equatorial reversal points (1°20′ N, 6°58′ N, and 17°21′ N, respectively). Furthermore, pre-industrial manifestations of flexible foraging and gender equality did not significantly vary in east–west direction. Finally, the geographical positioning of gender equality in pre-industrial times can forecast over 40% of each of the oppositely sloping north–south gradients of gender equality in the opposite Northern and the Southern Hemispheres today.

These results suggest that habitat variability is in part directly associated with subsistence flexibility and indirectly associated with gender equality (Figure 3). The replication of these results in pre-industrial societies and contemporary countries and the long-term predictability of current gender equality should not be misunderstood as a claim that there are no temporal changes in psychosocial gender systems. Indeed, the continuity of the ecological and geographical patterning of gender equality around the globe does not exclude that local levels of gender equality are in flux due to local circumstances. As a case in point, it has been shown that decreases in pathogen prevalence in the United States since the middle of the 20^th century are linked to increases in gender equality (Varnum & Grossmann, 2017), although this local progress toward gender equality has slowed or stalled in recent decades (England et al., 2020).

Figure 3.

Greater Habitat Variability as a Core Explanation of Greater Gender Equality at Higher Latitudes.

Figure 3 is anything but self-evident. Notably, it has been proposed that greater gender equality leads to more fluid social networks and flexible work organization, at least among Agta, Mbendjele, Ache, and Ju/’hoansi hunter-gatherers (Dyble et al., 2015). Our habitat-based model reverses the proposed direction of relationships by suggesting that habitat variability is more directly associated with subsistence flexibility than with gender equality. Integrating these sets of findings raises the intriguing question of whether there is an extended pathway, leading from greater habitat variability at higher latitudes through more equality in gender roles to more flexibility in broader patterns of social contacts, verbal communications, and behavioral interactions. For example, Hudson et al. (2020) may be right in claiming that greater gender equality shapes greater food security, better health, more peace and stability, higher economic performance, better governance, and more environmental protection worldwide.

Importantly, any investigation into gender differences in personality or psychosocial functioning is inevitably done among women and men residing at a given latitude with a given level of gender equality. Indeed, a resulting statistical effect is inevitably based on the upward or downward deviation from the mean or baseline level of gendered mindsets and practices in daily life. For example, sexual harassment will be observed more often among higher latitude inhabitants with higher baselines of gender equality (e.g., Swedes) than among lower latitude inhabitants with lower baselines of gender equality (e.g., Indians). As a further consequence, gender research is insufficiently informative and generalizable as long as the latitudinal baseline level of gender equality is left out of consideration. For example, the recent finding that gender inequality undermines well-being of both women and men in liberal but not in conservative societies (Li et al., 2021) gains psychological breadth and depth against the context of parallel poleward increases in gender equality (this study) and in liberal ideology and subjective well-being (Van de Vliert & Van Lange, 2019).

All this should be read in the light of the strengths and weaknesses of our study. The strength that the results rest on representative sampling of mutually independent pre-industrial societies comes with the weakness that the analyses were not pre-registered and thus rather exploratory in nature. To reduce this weakness, we established the convergent, discriminant, and predictive validity of the geographical distribution of pre-industrial gender equality, and replicated the results in modern countries. The convergent validity was established through spatial reversibility—the opposite directions of the north–south gradients of gender equality in the opposite Northern and Southern Hemispheres. The discriminant validity of the north–south gradients was apparent from the negligible importance of east–west gradients of gender equality. Predictive validity was assessed by demonstrating that the habitat-based geography of past gender equality in 87 societies was able to forecast local levels of present gender equality in 175 countries.

The well-considered focus on the remote ecological origins of gender equality in pre-industrial times has the inherent disadvantage of leaving a variety of relevant precursors of present gender equality out of consideration. Our present-day analysis does cover the effects of current subsistence conditions, pathogen prevalence, population density, statehood, religiosity, national wealth, income inequality, and aggression. By design, however, this cross-time replication of the ecology and geography of pre-industrial gender equality is silent about the effects of, for example, colonial settlement patterns (e.g., Alesina et al., 2013), intergenerational transmission of gender roles (e.g., Cano & Hofmeister, 2023), current fertility conditions (e.g., McDonald, 2000), educational opportunities (e.g., Stoet & Geary, 2019), and the almost worldwide availability of social media (Ahmed, 2020).

An important commonality of these non-habitat covariates of present gender equality, and of present gender equality itself, is the almost axiomatic assumption that they cannot have caused variability in daylength, temperature, and precipitation, let alone opposite north–south gradients of such habitat variability in the opposite Northern and Southern Hemispheres. Hence, the more likely direction of the observed impact is from latitudinal habitat variability through subsistence flexibility to gender equality (Figure 3). The same latitude-by-longitude design can be employed to explore the flexibility-mediated impact of habitat variability on cross-cultural differences in, for example, collectivism, hierarchism, creativity, aggression, and life satisfaction (Van de Vliert, 2020; Van de Vliert & Van Lange, 2019).

Of course, supplementary habitat-variability explanations of gender inequality remain possible. A notable case in point may be the small peak of pre-industrial gender equality around the hottest latitude, the thermal Equator at about 6° N (Aschoff, 1981; Figure 2a). Indeed, sudden heat and drought waves may have been responsible for why food-gathering tribes (Andamanese, Mbuti, Tiwi, Warrau) and horticultural tribes (Cuna, Marshalese, Mende, Iban, Tiv), formerly residing around 6° N, evolved more flexible adjustments to the variability inherent in life-threatening weather disasters, with more egalitarian gender roles in their wake. Contemporary inhabitants of territories around 6° N seem to have embraced more male-dominated gender roles (Figure 2b), perhaps because their modes of subsistence have become less susceptible to heat and drought waves. To what extent life-threatening weather disasters promote gender equality, including biparental care for offspring, is a promising topic for further research.

Irrespective of how the pre-industrial and current prevalence of gender equality around the Equator might be explained, policymakers can make use of our habitat-based geographical perspective. Achieving equal rights, opportunities, and outcomes for girls and boys and women and men is the United Nations’ sustainable development goal Number 5 (United Nations Development Programme, 2018). If our tentative research findings are correct, achieving gender parity requires a north–south rather than east–west agenda—or, better still, an Equatorial-Zone program—for promoting female autonomy, inclusion, and empowerment. The most widely recognized target domains are better access of women to health, education, work, payment, and public influence (Table 4). Of course, tropical habitat invariability cannot be changed. It can, however, inform policymakers on where more efforts are needed to advance girls’ and women’s empowerment.

Supplemental Material

sj-docx-1-psp-10.1177_01461672241237383 – Supplemental material for Global Ecology and Geography of Gender Equality

Supplemental material, sj-docx-1-psp-10.1177_01461672241237383 for Global Ecology and Geography of Gender Equality by Evert Van de Vliert and Esther S. Kluwer in Personality and Social Psychology Bulletin

Footnotes

Acknowledgements

The authors thank Joshua Ackerman, Carsten De Dreu, Xu Huang, Tim Johnson, Paul Van Lange, Chris Welzel, and five anonymous reviewers for providing helpful comments and suggestions. The authors received no specific funding for this work.

Author Contributions

E.V.d.V. designed and performed the study and analyzed the data. E.V.d.V. and E.S.K. wrote and edited the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Evert Van de Vliert

Esther S. Kluwer

Supplemental Material

Supplemental material is available online at .

References

Ahmed

(2020). Concurrent media news use and gender-based political participation inequality in a low-income democracy. International Journal of Public Opinion Research, 32, 815–828. https://doi.org/10.1093/ijpor/edaa001

Alesina

Giuliano

Nunn

(2013). On the origins of gender roles: Women and the plough. The Quarterly Journal of Economics, 128, 469–530. https://doi.org/10.1093/qje/qjt005

Amin

Abreu Lopes

George

(2020). Structural determinants of gender inequality: Why they matter for adolescent girls’ sexual and reproductive health. BMJ, 368, 16985. https://doi.org/10.1136/bmj.16985

Aschoff

(1981). Annual rhythms in man. In Aschoff

(Ed.), Handbook of behavioral neurobiology: Vol. 4. Biological rhythms (pp. 475–487). Plenum Press.

Ashby

W. R.

(1958). Requisite variety and implications for control of complex systems. Cybernetica, 1, 83–99.

Boserup

(1970). Woman’s role in economic development. Allen & Unwin.

Bromham

Hua

Cardillo

Schneemann

Greenhill

S. J.

(2018). Parasites and politics: Why cross-cultural studies must control for relatedness, proximity and covariation. Royal Society Open Science, 5, 181100. https://doi.org/10.1098/rsos.181100

Burroughs

W. J.

(1997). Does the weather really matter? The social implication of climate change. Cambridge University Press.

Cano

Hofmeister

(2023). The intergenerational transmission of gender: Paternal influences on children’s gender attitudes. Journal of Marriage and Family, 85, 193–214. https://doi.org/10.1111/jomf.12863

10.

Cashdan

Steele

(2013). Pathogen prevalence, group bias, and collectivism in the standard cross-cultural sample. Human Nature, 24, 59–75. https://doi.org/10.1007/s12110-012-9159-3

11.

Chaaban

Cunningham

(2011). Measuring the economic gain of investing in girls: The girl effect dividend. World Bank. https://doi.org/10.1596/1813-9450-5753

12.

Cockburn

(2006). Prevalence of different modes of parental care in birds. Proceedings of the Royal Society B: Biological Sciences, 273, 1375–1383. https://doi.org/10.1098/rspb.2005.3458

13.

Coltrane

(1992). The micropolitics of gender in nonindustrial societies. Gender and Society, 6, 86–107. https://doi.org/10.1177/089124392006001006

14.

Dirks

(1993). Starvation and famine: Cross-cultural codes and some hypothesis tests. Cross-Cultural Research, 27, 28–69. https://doi.org/10.1177/106939719302700103

15.

Ditlevsen

P. D.

Svensmark

Johnsen

(1996). Contrasting atmospheric and climate dynamics of the last-glacial and Holocene periods. Nature, 379, 810–812. https://doi.org/10.1038/379810a0

16.

Divale

Seda

(2000). Cross-cultural codes of modernization. World Cultures, 11, 152–170.

17.

Divale

Seda

(2001). Modernization as changes in cultural complexity: New cross-cultural measurements. Cross-Cultural Research, 35, 127–153. https://doi.org/10.1177/106939710103500203

18.

Dyble

Salali

G. D.

Chaudhary

Page

Smith

Thompson

Vinicius

Mace

Migliano

A. B.

(2015). Sex equality can explain the unique social structure of hunter-gatherer bands. Science, 348(6236), 796–798. https://doi.org/10.1126/science.aaa5139

19.

England

Levine

Mishel

(2020). Progress toward gender equality in the United States has slowed or stalled. Proceedings of the National Academy of Sciences USA, 117, 6990–6997. https://doi.org/10.1073/pnas.1918891117

20.

Falk

Hermle

(2018). Relationship of gender differences in preferences to economic development and gender equality. Science, 362, Article eaas9899. https://doi.org/10.1126/science.aas9899

21.

Fincher

C. L.

Thornhill

(2012). Parasite-stress promotes in-group assortative sociality: The cases of strong family ties and heightened religiosity. Behavioral and Brain Sciences, 35, 61–79. https://doi.org/10.1017/S0140525X11000021

22.

Fritz

M. S.

MacKinnon

D. P.

(2007). Required sample size to detect the mediated effect. Psychological Science, 18, 233–239. https://doi.org/10.1111/j.1467-9280.2007.01882.x

23.

Giuliano

Nunn

(2021). Understanding cultural persistence and change. The Review of Economic Studies, 88, 1541–1581. https://doi.org/10.1093/restud/rdaa074

24.

Hansen

C. W.

Jensen

P. S.

Skovsgaard

C. V.

(2015). Modern gender roles and agricultural history: The Neolithic inheritance. Journal of Economic Growth, 20, 365–404. https://doi.org/10.1007/s10887-015-9119-y

25.

Hayes

A. F.

(2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed.). Guilford Press.

26.

Heylighen

(1992). Principles of systems and cybernetics: An evolutionary perspective. In Trappl

(Ed.), Cybernetics and systems (Vol. 1, pp. 3–10). World Scientific.

27.

Hofman

J. M.

Watts

D. J.

Athey

Garip

Griffiths

T. L.

Kleinberg

Margetts

Mullainathan

Salganik

M. J.

Vazire

Vespignani

Yarkoni

(2021). Integrating explanation and prediction in computational social science. Nature, 595, 181–188. https://doi.org/10.1038/s41586-021-03659-0

28.

Hofstede

(1999). Religion, masculinity, and sex. In Hofstede

(Ed.), Masculinity and femininity: The taboo dimension of national cultures (pp. 192–209). Sage.

29.

Hudson

V. M.

Bowen

D. L.

Nielsen

P. L.

(2020). The first political order: How sex shapes governance and national security worldwide. Columbia University Press.

30.

Hut

R. A.

Paolucci

Dor

Kyriacou

C. P.

Daan

(2013). Latitudinal clines: An evolutionary view on biological rhythms. Proceedings of the Royal Society B, 280, 20130433. https://doi.org/10.1098/rspb.2013.0433

31.

Inglehart

Norris

(2003). Rising tide: Gender equality and cultural change around the world. Cambridge University Press.

32.

Iversen

Rosenbluth

(2010). Women, work, and politics: The political economy of gender inequality. Yale University Press.

33.

Kirby

K. R.

Gray

R. D.

Greenhill

S. J.

Jordan

F. M.

Gomes-Ng

Bibiko

H.-J.

Blasi

D. E.

Botero

C. A.

Bowern

Ember

C. R.

Leehr

Low

B. S.

McCarter

Divale

Gavin

M. C.

(2016). D-PLACE: A global database of cultural, linguistic and environmental diversity. PLOS ONE, 11(7), Article e0158391. https://doi.org/10.1371/journal.pone.0158391

34.

Klasen

(2002). Low schooling for girls, slower growth for all? Cross-country evidence on the effect of gender inequality in education on economic development. The World Bank Economic Review, 16, 345–373. https://doi.org/10.1093/wber/lhf004

35.

Klayman

Y.-W.

(1987). Confirmation, disconfirmation, and information in hypothesis testing. Psychological Review, 94, 211–228. https://doi.org/10.1037/0033-295X.94.2.211

36.

Lang

(1998). CONAN: An electronic code-text data-base for cross-cultural studies. World Cultures, 9, 13–56.

37.

Lawrence

P. R.

Lorsch

J. W.

(1967). Organization and environment: Managing differentiation and integration. Harvard University Press.

38.

Zuckerman

Diener

(2021). Culture moderates the relation between gender inequality and well-being. Psychological Science, 32, 823–835. https://doi.org/10.1177/0956797620972492

39.

Lisovski

Ramenofsky

Wingfield

J. C.

(2017). Defining the degree of seasonality and its significance for future research. Integrative and Comparative Biology, 57, 934–942. https://doi.org/10.1093/icb/icx040

40.

Mace

Holden

C. J.

(2005). A phylogenetic approach to cultural evolution. Trends in Ecology and Evolution, 20, 116–121. https://doi.org/10.1016/j.tree.2004.12.002

41.

Marlowe

(2000). Paternal investment and the human mating system. Behavioural Processes, 51, 45–61. https://doi.org/10.1016/S0376-6357(00)00118-2

42.

McDonald

(2000). Gender equity in theories of fertility transition. Population and Development Review, 26, 427–439. www.jstor.org/stable/172314

43.

Mintzberg

(1979). The structuring of organizations. Prentice-Hall.

44.

(2019). Aspects of gender in famine: Evidence from the Chinese great leap forward famine. In Preedy

V. R.

Patel

V. B.

(Eds.), Handbook of famine, starvation, and nutrient deprivation (pp. 61–76). Springer. https://doi.org/10.1007/978-3-319-55387-0_105

45.

Murdock

G. P.

White

D. R.

(1969). Standard cross-cultural sample. Ethnology, 8, 329–369. https://doi.org/10.2307/3772907

46.

Murdock

G. P.

(1967). Ethnographic atlas: A summary. Ethnology, 6, 109–236. https://doi.org/10.2307/3772751

47.

Murdock

G. P.

Wilson

S. F.

(1972). Settlement patterns and community organization. Ethnology, 11, 254–295. https://doi.org/10.2307/3773219

48.

Nolan

Lenski

(1999). Human societies: An introduction to macrosociology (8th ed.). McGraw-Hill.

49.

Organisation for Economic Co-operation and Development. (2019). SIGI 2019 global report: Transforming challenges into opportunities. Social Institutions and Gender Index. https://doi.org/10.1787/bc56d212-en

50.

Parker

P. M.

(1997). National cultures of the world: A statistical reference. Greenwood Press.

51.

Ross

M. H.

(1983). Political decision making and conflict: Additional cross-cultural codes and scales. Ethnology, 22, 169–192. https://doi.org/10.2307/3773578

52.

Ross

M. H.

(1993). The culture of conflict. Yale University Press.

53.

Schill

Anderies

J. M.

Lindahl

Folke

Polasky

Cárdenas

J. C.

Crépin

A.-S.

Janssen

M. A.

Norberg

Schlüter

(2019). A more dynamic understanding of human behaviour for the Anthropocene. Nature Sustainability, 2, 1075–1082. https://doi.org/10.1038/s41893-019-0419-7

54.

Social Watch. (2012). Gender equity index. www.socialwatch.org/book/export/html/14367

55.

Stoet

Geary

D. C.

(2019). A simplified approach to measuring national gender inequality. PLOS ONE, 14, Article e0205349. https://doi.org/10.1371/journal.pone.0205349

56.

Sullivan

I. I. I.

(1991). Measuring global values: The ranking of 162 countries. Greenwood.

57.

Tuden

Marshall

(1972). Political organization: Cross-cultural codes 4. Ethnology, 11, 436–464. https://doi.org/10.2307/3773074

58.

United Nations Development Programme. (2002). Human development report 2002. Oxford University Press.

59.

United Nations Development Programme. (2004). Human development report 2004. Oxford University Press.

60.

United Nations Development Programme. (2006). Human development report 2006. Palgrave Macmillan.

61.

United Nations Development Programme. (2018). Human development indices and indicators.

62.

Van de Vliert

(2020). The global ecology of differentiation between us and them. Nature Human Behaviour, 4, 270–278. https://doi.org/10.1038/s41562-019-0783-3

63.

Van de Vliert

Van Lange

P. A. M.

(2019). Latitudinal psychology: An ecological perspective on creativity, aggression, happiness, and beyond. Perspectives on Psychological Science, 14, 860–884. https://doi.org/10.1177/1745691619858067

64.

Varnum

M. E. W.

Grossmann

(2017). Pathogen prevalence is associated with cultural changes in gender equality. Nature Human Behaviour, 1, Article 0003. https://doi.org/10.1038/s41562-016-0003

65.

Von Bertalanffy

. (1968). General system theory. George Braziller.

66.

Willig

M. R.

Kaufman

D. M.

Stevens

R. D.

(2003). Latitudinal gradients of biodiversity: Pattern, process, scale, and synthesis. Annual Review of Ecology, Evolution, and Systematics, 34, 273–309. https://doi.org/10.1146/annurev.ecolsys.34.012103.144032

67.

World Economic Forum. (2018). The global gender gap report 2018. World Economic Forum.

68.

Yarkoni

Westfall

(2017). Choosing prediction over explanation in psychology: Lessons from machine learning. Perspectives on Psychological Science, 12, 1100–1122. https://doi.org/10.1177/1745691617693393

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB