The obedience-autonomy disposition

People who name obedience and religious faith over independence and tolerance as important qualities for a child place themselves further right in most of the 103103 countries and territoriesEstimandcount of countries with a usable correlation between left-right and the full indexWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries in the World Values Survey (WVS) and European Values Study (EVS). The median correlationCorrelationA number from minus 1 to plus 1 for how closely two measures move together in a straight-line pattern. Zero means no straight-line pattern; a correlation does not say that one measure causes the other. between that measure and left-right placementLeft-right self-placementA survey question that asks people to place themselves on a scale from left to right. The scale is the same for everyone, but the meaning attached to left and right differs across people, countries and years. is +0.04 (95% CI 0.03 to 0.06)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.03 to 0.06n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, and the correlation is above zero in 8484 countries and territoriesEstimandcount of countries where the point estimate is above zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries. A correlation of zero means that two answers are unrelated, so a median this close to zero means that the child-rearing answer predicts left-right placement only weakly.

The chapter follows the measure through three main surveys, the WVS and EVS, the American National Election Studies (ANES) and the European Social Survey (ESS), and adds the General Social Survey (GSS) as a second American series. The ESS has no question about children and stands in with a neighboring measure of how much people value conformity, tradition and security. All four surveys correlate the measure with the same kind of outcome, a left-right or party question, and the chapter returns to that overlap below. Chapter 7.4, Left and right across countries shows how differently the left-right question attaches to religion, economics and moral attitudes from one country to the next, and Chapter 8.1, Basic values and left-right compares Schwartz's basic values with the scale.

The surveys measure the disposition with different questions

Stenner argued that a lasting preference for sameness and authority over difference and choice can be read from child-rearing preferences, because those questions do not mention politics [1]. Feldman described the same disposition as a conflict between conformity and personal autonomy [2], and Hetherington and Weiler used four American forced choices to study polarization [3]. Engelhardt, Feldman and Hetherington assess the child-rearing items as a measure of authoritarianism [4]. The WVS checklist also feeds the traditional versus secular-rational dimension of the WVS cultural map [5].

Figure 1

Three surveys ask about children and one asks about values, and each fixes the obedience pole in advance

Questions, the answers that count as each pole and the measure built from them, by survey

Survey and years	Question	Obedience versus autonomy pole	Measure
World Values Survey and European Values Study, fieldwork 2005-2023	Check-all list: which qualities are especially important for a child to learn at home (up to five; ten qualities in the cleaned file)	Obedience and religious faith, against independence and tolerance and respect for other people	Obedience and faith minus independence and tolerance, from a low of −2 to a high of +2
American National Election Studies, presidential-year studies 1992-2024	Four forced choices: independence or respect for elders, curiosity or good manners, self-reliance or obedience, considerate or well-behaved	Respect for elders, good manners, obedience and being well-behaved, against independence, curiosity, self-reliance and being considerate	One minus the mean of the four answers, from 0 (autonomy on every pair) to 1 (obedience on every pair)
General Social Survey, 1986-2024	Rank five qualities by importance for a child to learn, including obeying and thinking for oneself	Obeying, against thinking for oneself	Rank of thinking for oneself minus rank of obeying, from −4 to +4
European Social Survey, rounds 3 to 10, 2006-2022	Twenty-one portraits of a person; how much the respondent resembles each. No child-rearing item	Conformity, tradition and security values (conservation); openness to change is the opposite pole and is not measured directly	Conservation score, centered on the respondent's mean rating

Show data

Data for Three surveys ask about children and one asks about values, and each fixes the obedience pole in advance

measure	poles	question	survey
Obedience and faith minus independence and tolerance, from a low of −2 to a high of +2	Obedience and religious faith, against independence and tolerance and respect for other people	Check-all list: which qualities are especially important for a child to learn at home (up to five; ten qualities in the cleaned file)	World Values Survey and European Values Study, fieldwork 2005-2023
One minus the mean of the four answers, from 0 (autonomy on every pair) to 1 (obedience on every pair)	Respect for elders, good manners, obedience and being well-behaved, against independence, curiosity, self-reliance and being considerate	Four forced choices: independence or respect for elders, curiosity or good manners, self-reliance or obedience, considerate or well-behaved	American National Election Studies, presidential-year studies 1992-2024
Rank of thinking for oneself minus rank of obeying, from −4 to +4	Obeying, against thinking for oneself	Rank five qualities by importance for a child to learn, including obeying and thinking for oneself	General Social Survey, 1986-2024
Conservation score, centered on the respondent's mean rating	Conformity, tradition and security values (conservation); openness to change is the opposite pole and is not measured directly	Twenty-one portraits of a person; how much the respondent resembles each. No child-rearing item	European Social Survey, rounds 3 to 10, 2006-2022

Each survey fixes the obedience pole from the wording of its answers before any correlation with left-right is computed. In the WVS and EVS, respondents pick up to five qualities from a list, the cleaned file carries ten of them, and the full index counts obedience and religious faith on one side and independence and tolerance and respect for other people on the other. The ANES sets an autonomy quality against an obedience quality in each of its four pairs, and its index is the mean of the obedience answers.Each correlation is survey-weighted, and its 95% interval (built on the Kish effective sample sizeEffective sample sizeThe number of equally weighted respondents that would give the same precision as the weighted sample. It is smaller than the raw count when the weights vary a lot, and it sets how wide the confidence interval is. [6] and Fisher z [7]) treats the weightedSurvey weightsNumbers that scale each respondent so the sample matches the population on known traits such as age, race, education and region. A weighted average counts a respondent with a large weight for more than one with a small weight. sample as a simple random sample, so it is narrower than an interval that follows the sampling design. Intervals on medians and means across countries come from a bootstrap over countries.

The obedience side is larger in lower-income countries

Figure 2 counts respondents on the obedience side, those who name more obedience-side than autonomy-side qualities, by country income group (the World Bank groups in the WVS context file) and birth decade. In high-income countries the share is 11.511.5 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 9.0 to 14.1Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in high-income countries born in the 1940s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 36 countries.% among people born in the 1940s and 7.47.4 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 5.8 to 9.2Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in high-income countries born in the 1990s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 36 countries.% among those born in the 1990s, a change of −4.1 (95% CI −5.6 to −2.6)−4.1 percentage pointsEstimandmean across countries of the share above zero among those born in the 1990s minus those born in the 1940s (bootstrap over countries)Interval95% CI −5.6 to −2.6Weightsweight (S017) within each country; every country counted once across countriesPopulationhigh-income countries, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 36 countries. points. The figures are 25.625.6 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 20.7 to 30.8Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in upper-middle-income countries born in the 1940s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 22 countries.% and 20.920.9 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 17.0 to 24.9Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in upper-middle-income countries born in the 1990s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 22 countries.% in upper-middle-income countries (a change of −4.7 (95% CI −7.3 to −2.4)−4.7 percentage pointsEstimandmean across countries of the share above zero among those born in the 1990s minus those born in the 1940s (bootstrap over countries)Interval95% CI −7.3 to −2.4Weightsweight (S017) within each country; every country counted once across countriesPopulationupper-middle-income countries, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 22 countries.) and 31.531.5 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 23.3 to 40.9Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in lower-income countries born in the 1940s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 10 countries.% and 27.927.9 percent of respondents above zero on the indexEstimandmean across countries of the weighted share above zero on the full index (bootstrap over countries)Interval95% CI 20.8 to 36.9Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in lower-income countries born in the 1990s, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 10 countries.% in lower-income countries (−3.6 (95% CI −10.0 to 2.5)−3.6 percentage pointsEstimandmean across countries of the share above zero among those born in the 1990s minus those born in the 1940s (bootstrap over countries)Interval95% CI −10.0 to 2.5Weightsweight (S017) within each country; every country counted once across countriesPopulationlower-income countries, fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 10 countries., an interval that includes zero). The gap between lower-income and high-income countries is larger than the change across birth decades in either group.

Figure 2

The obedience side is largest in lower-income countries and shrinks across birth decades in richer ones

Percent of respondents on the obedience side (more obedience-side than autonomy-side qualities), by birth decade and country income group; fixed country panels; WVS and EVS fieldwork 2005-2023

Show data

Data for The obedience side is largest in lower-income countries and shrinks across birth decades in richer ones

Country group	Birth decade	Percent on the obedience side	Interval low	Interval high	Countries
High income	1940	11.5	9.0	14.1	36
High income	1950	9.7	7.4	12.2	36
High income	1960	8.9	6.7	11.4	36
High income	1970	8.4	6.5	10.5	36
High income	1980	8.3	6.4	10.2	36
High income	1990	7.4	5.8	9.2	36
Lower income	1940	31.5	23.3	40.9	10
Lower income	1950	30.0	22.8	38.6	10
Lower income	1960	28.9	21.3	38.2	10
Lower income	1970	29.7	22.4	38.6	10
Lower income	1980	27.8	21.0	36.1	10
Lower income	1990	27.9	20.8	36.9	10
Upper-middle income	1940	25.6	20.7	30.8	22
Upper-middle income	1950	24.8	19.9	29.5	22
Upper-middle income	1960	23.9	19.0	29.1	22
Upper-middle income	1970	23.3	18.6	28.2	22
Upper-middle income	1980	21.7	17.1	26.8	22
Upper-middle income	1990	20.9	17.0	24.9	22

A cohortCohortPeople born in the same period, such as the same decade. Comparing cohorts shows whether a difference comes from when people were born, as distinct from their age or the year they were surveyed. profile from surveys spread over 2005 to 2023 mixes age with birth decade, because a later survey measures the same birth decade at an older age. Comparing the same birth decade measured in 2014 to 2023 with 2005 to 2013 gives a mean difference of −0.3 (95% CI −0.9 to 0.2)−0.3 percentage pointsEstimandmean difference in the share above zero for the same birth decade between fieldwork in 2014-2023 and 2005-2013 (bootstrap over cells)Interval95% CI −0.9 to 0.2n312Weightsweight (S017) within each country; every country counted once across countriesPopulationcountry-by-birth-decade cells in the fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points across country-by-decade cells, which rules out a uniform shift between the two periods, but 2828 percent of cellsEstimandshare of cells where the two windows differ by more than 5 percentage points in either directionn312Weightseach country-by-birth-decade cell counted once; shares inside a cell use weight (S017)Populationcountry-by-birth-decade cells in the fixed panelPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS% of the cells differ by more than 55 percentage pointsEstimandcutoff on the absolute difference in the share above zeroWeightsnone (constant)Populationdefinition of a large difference between the two survey windowsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points, about the size of the cohort changes in the two richer groups, so period effects in single countries are not ruled out.

The correlation is positive but small in most countries

Figure 3 shows the correlation for each of the 103103 countries and territoriesEstimandcount of countries with a usable correlation between left-right and the full indexWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries with a usable estimate, computed on 281,678281,678 respondentsEstimandcount of respondentsWeightsnone (count)Populationrespondents with a left-right answer and all four child-quality answers, in the countries with a usable correlationPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS respondents. The median is +0.04 (95% CI 0.03 to 0.06)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.03 to 0.06n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and the mean is +0.06 (95% CI 0.04 to 0.07)+0.06 correlation coefficient, from −1 to +1Estimandmean across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.04 to 0.07n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS. The 55 countries and territoriesEstimandcount of countries where the 95% interval is below zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries with an interval below zero are Bulgaria, Czechia, Egypt, Moldova and Zambia.

Figure 3

The correlation is positive in 84 of 103 countries and +0.04 in the median country

Weighted correlation between left-right self-placement (10 = right) and the obedience measure, one dot per country; WVS and EVS fieldwork 2005-2023, ESS rounds 3 to 10

WVS and EVS, full indexWVS and EVS, without faithESS, conservation valuesESS, net of age, sex and education

• Interval above zero
• Interval includes zero
• Interval below zero

Show data

Data for The correlation is positive in 84 of 103 countries and +0.04 in the median country

Country	Region	Correlation	Interval low	Interval high	Respondents	Measure
Albania	Europe & Central Asia	+0.06	+0.02	+0.11	2,253	WVS and EVS, full index
Algeria	Middle East & North Africa	+0.06	−0.02	+0.14	545	WVS and EVS, full index
Andorra	Europe & Central Asia	+0.20	+0.15	+0.24	1,824	WVS and EVS, full index
Argentina	Latin America & Caribbean	+0.15	+0.11	+0.20	2,346	WVS and EVS, full index
Armenia	Europe & Central Asia	+0.05	+0.01	+0.08	3,776	WVS and EVS, full index
Australia	East Asia & Pacific	+0.17	+0.13	+0.20	4,488	WVS and EVS, full index
Austria	Europe & Central Asia	+0.17	+0.13	+0.21	2,685	WVS and EVS, full index
Azerbaijan	Europe & Central Asia	+0.04	0.00	+0.09	2,174	WVS and EVS, full index
Bangladesh	South Asia	0.00	−0.06	+0.06	1,200	WVS and EVS, full index
Belarus	Europe & Central Asia	0.00	−0.03	+0.04	3,222	WVS and EVS, full index
Belgium	Europe & Central Asia	+0.07	+0.01	+0.12	1,422	WVS and EVS, full index
Bolivia	Latin America & Caribbean	+0.05	+0.01	+0.10	1,800	WVS and EVS, full index
Bosnia Herzegovina	Europe & Central Asia	+0.10	+0.06	+0.15	2,315	WVS and EVS, full index
Brazil	Latin America & Caribbean	+0.06	+0.03	+0.10	3,680	WVS and EVS, full index
Bulgaria	Europe & Central Asia	−0.06	−0.10	−0.02	2,665	WVS and EVS, full index
Burkina Faso	Sub-Saharan Africa	+0.02	−0.04	+0.08	1,126	WVS and EVS, full index
Canada	North America	+0.25	+0.22	+0.28	5,642	WVS and EVS, full index
Chile	Latin America & Caribbean	+0.06	+0.01	+0.12	2,127	WVS and EVS, full index
Colombia	Latin America & Caribbean	+0.08	+0.06	+0.11	5,023	WVS and EVS, full index
Croatia	Europe & Central Asia	+0.16	+0.12	+0.21	2,196	WVS and EVS, full index
Cyprus	Europe & Central Asia	−0.01	−0.05	+0.03	2,894	WVS and EVS, full index
Czechia	Europe & Central Asia	−0.06	−0.09	−0.02	3,615	WVS and EVS, full index
Denmark	Europe & Central Asia	+0.09	+0.06	+0.13	4,688	WVS and EVS, full index
Ecuador	Latin America & Caribbean	+0.08	+0.04	+0.12	2,244	WVS and EVS, full index
Egypt	Middle East & North Africa	−0.14	−0.17	−0.10	4,245	WVS and EVS, full index
Estonia	Europe & Central Asia	−0.01	−0.05	+0.02	3,319	WVS and EVS, full index
Ethiopia	Sub-Saharan Africa	+0.02	−0.02	+0.07	1,931	WVS and EVS, full index
Finland	Europe & Central Asia	+0.11	+0.07	+0.15	2,955	WVS and EVS, full index
France	Europe & Central Asia	+0.13	+0.10	+0.17	3,862	WVS and EVS, full index
Georgia	Europe & Central Asia	+0.01	−0.02	+0.05	4,248	WVS and EVS, full index
Germany	Europe & Central Asia	+0.12	+0.09	+0.14	8,805	WVS and EVS, full index
Ghana	Sub-Saharan Africa	+0.04	−0.01	+0.08	2,334	WVS and EVS, full index
Greece	Europe & Central Asia	+0.18	+0.14	+0.22	2,201	WVS and EVS, full index
Guatemala	Latin America & Caribbean	−0.03	−0.10	+0.05	728	WVS and EVS, full index
Haiti	Latin America & Caribbean	+0.02	−0.03	+0.06	1,945	WVS and EVS, full index
Hong Kong SAR	East Asia & Pacific	+0.08	+0.04	+0.11	3,960	WVS and EVS, full index
Hungary	Europe & Central Asia	+0.07	+0.03	+0.10	3,444	WVS and EVS, full index
Iceland	Europe & Central Asia	+0.06	+0.01	+0.11	2,322	WVS and EVS, full index
India	South Asia	+0.01	−0.01	+0.04	5,826	WVS and EVS, full index
Indonesia	East Asia & Pacific	+0.01	−0.03	+0.04	4,093	WVS and EVS, full index
Iraq	Middle East & North Africa	+0.12	+0.05	+0.18	978	WVS and EVS, full index
Italy	Europe & Central Asia	+0.16	+0.13	+0.19	3,552	WVS and EVS, full index
Japan	East Asia & Pacific	+0.06	+0.03	+0.09	3,609	WVS and EVS, full index
Kazakhstan	Europe & Central Asia	+0.02	−0.03	+0.07	1,500	WVS and EVS, full index
Kenya	Sub-Saharan Africa	+0.02	−0.04	+0.07	1,164	WVS and EVS, full index
Kosovo	Europe & Central Asia	−0.02	−0.10	+0.06	662	WVS and EVS, full index
Kyrgyzstan	Europe & Central Asia	+0.08	+0.03	+0.13	1,461	WVS and EVS, full index
Latvia	Europe & Central Asia	−0.01	−0.05	+0.04	1,985	WVS and EVS, full index
Lebanon	Middle East & North Africa	+0.04	−0.02	+0.11	837	WVS and EVS, full index
Libya	Middle East & North Africa	+0.01	−0.04	+0.06	2,606	WVS and EVS, full index
Lithuania	Europe & Central Asia	+0.07	+0.02	+0.11	1,831	WVS and EVS, full index
Luxembourg	Europe & Central Asia	+0.22	+0.16	+0.28	1,256	WVS and EVS, full index
Macau SAR	East Asia & Pacific	−0.01	−0.08	+0.05	997	WVS and EVS, full index
Malaysia	East Asia & Pacific	+0.02	−0.02	+0.06	2,613	WVS and EVS, full index
Maldives	South Asia	+0.04	−0.02	+0.10	1,022	WVS and EVS, full index
Mali	Sub-Saharan Africa	0.00	−0.06	+0.05	1,193	WVS and EVS, full index
Malta	Middle East & North Africa	+0.05	−0.03	+0.12	764	WVS and EVS, full index
Mexico	Latin America & Caribbean	+0.04	+0.01	+0.07	4,785	WVS and EVS, full index
Moldova	Europe & Central Asia	−0.07	−0.12	−0.02	1,610	WVS and EVS, full index
Mongolia	East Asia & Pacific	+0.04	−0.01	+0.09	1,613	WVS and EVS, full index
Montenegro	Europe & Central Asia	+0.03	−0.03	+0.10	1,211	WVS and EVS, full index
Morocco	Middle East & North Africa	+0.16	+0.12	+0.20	1,916	WVS and EVS, full index
Netherlands	Europe & Central Asia	+0.20	+0.18	+0.22	7,886	WVS and EVS, full index
New Zealand	East Asia & Pacific	+0.15	+0.10	+0.20	1,423	WVS and EVS, full index
Nicaragua	Latin America & Caribbean	+0.03	−0.04	+0.09	871	WVS and EVS, full index
Nigeria	Sub-Saharan Africa	+0.02	−0.02	+0.06	2,929	WVS and EVS, full index
North Macedonia	Europe & Central Asia	+0.05	−0.03	+0.13	766	WVS and EVS, full index
Northern Ireland	Europe & Central Asia	+0.16	+0.06	+0.25	642	WVS and EVS, full index
Norway	Europe & Central Asia	+0.08	+0.04	+0.11	3,154	WVS and EVS, full index
Pakistan	South Asia	−0.04	−0.09	+0.02	1,200	WVS and EVS, full index
Palestine	Middle East & North Africa	−0.01	−0.09	+0.06	730	WVS and EVS, full index
Peru	Latin America & Caribbean	+0.02	−0.01	+0.06	3,345	WVS and EVS, full index
Philippines	East Asia & Pacific	+0.03	−0.02	+0.07	2,389	WVS and EVS, full index
Poland	Europe & Central Asia	+0.19	+0.16	+0.22	3,599	WVS and EVS, full index
Portugal	Europe & Central Asia	+0.05	0.00	+0.11	1,942	WVS and EVS, full index
Puerto Rico	Latin America & Caribbean	0.00	−0.06	+0.06	1,049	WVS and EVS, full index
Romania	Europe & Central Asia	−0.02	−0.05	+0.01	4,539	WVS and EVS, full index
Russia	Europe & Central Asia	+0.01	−0.02	+0.05	4,611	WVS and EVS, full index
Rwanda	Sub-Saharan Africa	+0.06	+0.03	+0.10	2,555	WVS and EVS, full index
Serbia	Europe & Central Asia	+0.03	−0.01	+0.07	3,739	WVS and EVS, full index
Singapore	East Asia & Pacific	+0.07	+0.02	+0.11	1,745	WVS and EVS, full index
Slovakia	Europe & Central Asia	+0.03	−0.01	+0.07	3,260	WVS and EVS, full index
Slovenia	Europe & Central Asia	+0.17	+0.13	+0.20	3,091	WVS and EVS, full index
South Africa	Sub-Saharan Africa	−0.02	−0.06	+0.01	5,594	WVS and EVS, full index
South Korea	East Asia & Pacific	+0.04	0.00	+0.08	3,640	WVS and EVS, full index
Spain	Europe & Central Asia	+0.20	+0.17	+0.23	4,298	WVS and EVS, full index
Sweden	Europe & Central Asia	+0.04	0.00	+0.07	4,254	WVS and EVS, full index
Switzerland	Europe & Central Asia	+0.15	+0.12	+0.17	5,134	WVS and EVS, full index
Taiwan ROC	Not classified	0.00	−0.04	+0.04	3,563	WVS and EVS, full index
Tajikistan	Europe & Central Asia	−0.03	−0.09	+0.02	1,200	WVS and EVS, full index
Thailand	East Asia & Pacific	+0.06	+0.02	+0.09	3,763	WVS and EVS, full index
Trinidad and Tobago	Latin America & Caribbean	+0.01	−0.05	+0.07	1,213	WVS and EVS, full index
Tunisia	Middle East & North Africa	−0.01	−0.06	+0.04	1,678	WVS and EVS, full index
Turkey	Europe & Central Asia	+0.16	+0.13	+0.19	4,467	WVS and EVS, full index
Ukraine	Europe & Central Asia	+0.05	+0.01	+0.10	4,565	WVS and EVS, full index
United Kingdom	Europe & Central Asia	+0.13	+0.11	+0.16	6,150	WVS and EVS, full index
United States	North America	+0.28	+0.25	+0.31	5,900	WVS and EVS, full index
Uruguay	Latin America & Caribbean	+0.14	+0.10	+0.18	2,662	WVS and EVS, full index
Uzbekistan	Europe & Central Asia	+0.01	−0.07	+0.09	615	WVS and EVS, full index
Venezuela	Latin America & Caribbean	−0.06	−0.11	0.00	1,190	WVS and EVS, full index
Vietnam	East Asia & Pacific	+0.11	+0.06	+0.16	1,408	WVS and EVS, full index
Zambia	Sub-Saharan Africa	−0.07	−0.14	−0.01	959	WVS and EVS, full index
Zimbabwe	Sub-Saharan Africa	+0.02	−0.03	+0.06	2,627	WVS and EVS, full index
Albania	Europe & Central Asia	0.00	−0.04	+0.05	2,253	WVS and EVS, without the religious-faith item
Algeria	Middle East & North Africa	+0.06	−0.02	+0.14	545	WVS and EVS, without the religious-faith item
Andorra	Europe & Central Asia	+0.16	+0.12	+0.21	1,824	WVS and EVS, without the religious-faith item
Argentina	Latin America & Caribbean	+0.15	+0.11	+0.20	2,346	WVS and EVS, without the religious-faith item
Armenia	Europe & Central Asia	+0.05	+0.01	+0.08	3,783	WVS and EVS, without the religious-faith item
Australia	East Asia & Pacific	+0.14	+0.11	+0.18	4,488	WVS and EVS, without the religious-faith item
Austria	Europe & Central Asia	+0.16	+0.12	+0.20	2,693	WVS and EVS, without the religious-faith item
Azerbaijan	Europe & Central Asia	+0.04	−0.01	+0.08	2,174	WVS and EVS, without the religious-faith item
Bangladesh	South Asia	+0.01	−0.05	+0.07	1,200	WVS and EVS, without the religious-faith item
Belarus	Europe & Central Asia	+0.01	−0.03	+0.04	3,223	WVS and EVS, without the religious-faith item
Belgium	Europe & Central Asia	+0.05	0.00	+0.11	1,422	WVS and EVS, without the religious-faith item
Bolivia	Latin America & Caribbean	+0.05	0.00	+0.10	1,800	WVS and EVS, without the religious-faith item
Bosnia Herzegovina	Europe & Central Asia	+0.05	0.00	+0.09	2,315	WVS and EVS, without the religious-faith item
Brazil	Latin America & Caribbean	+0.06	+0.03	+0.09	3,680	WVS and EVS, without the religious-faith item
Bulgaria	Europe & Central Asia	−0.07	−0.12	−0.03	2,689	WVS and EVS, without the religious-faith item
Burkina Faso	Sub-Saharan Africa	−0.02	−0.08	+0.04	1,126	WVS and EVS, without the religious-faith item
Canada	North America	+0.21	+0.18	+0.24	5,642	WVS and EVS, without the religious-faith item
Chile	Latin America & Caribbean	+0.04	−0.02	+0.09	2,127	WVS and EVS, without the religious-faith item
Colombia	Latin America & Caribbean	+0.07	+0.04	+0.09	5,023	WVS and EVS, without the religious-faith item
Croatia	Europe & Central Asia	+0.08	+0.03	+0.13	2,208	WVS and EVS, without the religious-faith item
Cyprus	Europe & Central Asia	−0.01	−0.05	+0.03	2,898	WVS and EVS, without the religious-faith item
Czechia	Europe & Central Asia	−0.06	−0.10	−0.03	3,663	WVS and EVS, without the religious-faith item
Denmark	Europe & Central Asia	+0.08	+0.04	+0.11	4,688	WVS and EVS, without the religious-faith item
Ecuador	Latin America & Caribbean	+0.07	+0.03	+0.11	2,244	WVS and EVS, without the religious-faith item
Egypt	Middle East & North Africa	−0.13	−0.16	−0.09	4,245	WVS and EVS, without the religious-faith item
Estonia	Europe & Central Asia	−0.01	−0.04	+0.03	3,322	WVS and EVS, without the religious-faith item
Ethiopia	Sub-Saharan Africa	+0.04	0.00	+0.08	1,931	WVS and EVS, without the religious-faith item
Finland	Europe & Central Asia	+0.08	+0.04	+0.12	2,955	WVS and EVS, without the religious-faith item
France	Europe & Central Asia	+0.13	+0.09	+0.16	3,862	WVS and EVS, without the religious-faith item
Georgia	Europe & Central Asia	+0.02	−0.01	+0.05	4,250	WVS and EVS, without the religious-faith item
Germany	Europe & Central Asia	+0.11	+0.09	+0.14	8,805	WVS and EVS, without the religious-faith item
Ghana	Sub-Saharan Africa	+0.01	−0.04	+0.06	2,334	WVS and EVS, without the religious-faith item
Greece	Europe & Central Asia	+0.09	+0.05	+0.13	2,201	WVS and EVS, without the religious-faith item
Guatemala	Latin America & Caribbean	−0.03	−0.10	+0.04	759	WVS and EVS, without the religious-faith item
Haiti	Latin America & Caribbean	+0.01	−0.03	+0.06	1,945	WVS and EVS, without the religious-faith item
Hong Kong SAR	East Asia & Pacific	+0.09	+0.05	+0.12	3,960	WVS and EVS, without the religious-faith item
Hungary	Europe & Central Asia	+0.02	−0.01	+0.06	3,446	WVS and EVS, without the religious-faith item
Iceland	Europe & Central Asia	+0.04	0.00	+0.09	2,323	WVS and EVS, without the religious-faith item
India	South Asia	−0.01	−0.03	+0.02	5,826	WVS and EVS, without the religious-faith item
Indonesia	East Asia & Pacific	+0.01	−0.03	+0.04	4,093	WVS and EVS, without the religious-faith item
Iraq	Middle East & North Africa	+0.02	−0.05	+0.08	978	WVS and EVS, without the religious-faith item
Italy	Europe & Central Asia	+0.14	+0.11	+0.17	3,554	WVS and EVS, without the religious-faith item
Japan	East Asia & Pacific	+0.05	+0.02	+0.09	3,609	WVS and EVS, without the religious-faith item
Kazakhstan	Europe & Central Asia	+0.01	−0.04	+0.06	1,500	WVS and EVS, without the religious-faith item
Kenya	Sub-Saharan Africa	+0.02	−0.04	+0.08	1,164	WVS and EVS, without the religious-faith item
Kosovo	Europe & Central Asia	+0.04	−0.04	+0.11	662	WVS and EVS, without the religious-faith item
Kyrgyzstan	Europe & Central Asia	+0.08	+0.02	+0.13	1,461	WVS and EVS, without the religious-faith item
Latvia	Europe & Central Asia	−0.01	−0.05	+0.04	2,014	WVS and EVS, without the religious-faith item
Lebanon	Middle East & North Africa	0.00	−0.07	+0.07	837	WVS and EVS, without the religious-faith item
Libya	Middle East & North Africa	+0.01	−0.04	+0.06	2,606	WVS and EVS, without the religious-faith item
Lithuania	Europe & Central Asia	+0.05	0.00	+0.10	1,834	WVS and EVS, without the religious-faith item
Luxembourg	Europe & Central Asia	+0.18	+0.12	+0.23	1,259	WVS and EVS, without the religious-faith item
Macau SAR	East Asia & Pacific	0.00	−0.07	+0.06	997	WVS and EVS, without the religious-faith item
Malaysia	East Asia & Pacific	0.00	−0.04	+0.04	2,613	WVS and EVS, without the religious-faith item
Maldives	South Asia	+0.06	−0.01	+0.12	1,022	WVS and EVS, without the religious-faith item
Mali	Sub-Saharan Africa	0.00	−0.06	+0.06	1,193	WVS and EVS, without the religious-faith item
Malta	Middle East & North Africa	+0.04	−0.04	+0.11	766	WVS and EVS, without the religious-faith item
Mexico	Latin America & Caribbean	+0.03	+0.01	+0.06	4,786	WVS and EVS, without the religious-faith item
Moldova	Europe & Central Asia	−0.07	−0.11	−0.02	1,616	WVS and EVS, without the religious-faith item
Mongolia	East Asia & Pacific	0.00	−0.05	+0.05	1,613	WVS and EVS, without the religious-faith item
Montenegro	Europe & Central Asia	−0.03	−0.09	+0.03	1,212	WVS and EVS, without the religious-faith item
Morocco	Middle East & North Africa	+0.11	+0.07	+0.16	1,916	WVS and EVS, without the religious-faith item
Netherlands	Europe & Central Asia	+0.18	+0.16	+0.20	7,889	WVS and EVS, without the religious-faith item
New Zealand	East Asia & Pacific	+0.14	+0.09	+0.19	1,423	WVS and EVS, without the religious-faith item
Nicaragua	Latin America & Caribbean	+0.01	−0.06	+0.07	871	WVS and EVS, without the religious-faith item
Nigeria	Sub-Saharan Africa	+0.01	−0.03	+0.05	2,929	WVS and EVS, without the religious-faith item
North Macedonia	Europe & Central Asia	+0.05	−0.03	+0.13	804	WVS and EVS, without the religious-faith item
Northern Ireland	Europe & Central Asia	+0.12	+0.02	+0.21	650	WVS and EVS, without the religious-faith item
Norway	Europe & Central Asia	+0.06	+0.02	+0.09	3,154	WVS and EVS, without the religious-faith item
Pakistan	South Asia	−0.05	−0.11	+0.01	1,200	WVS and EVS, without the religious-faith item
Palestine	Middle East & North Africa	−0.03	−0.10	+0.05	730	WVS and EVS, without the religious-faith item
Peru	Latin America & Caribbean	+0.02	−0.01	+0.06	3,345	WVS and EVS, without the religious-faith item
Philippines	East Asia & Pacific	+0.01	−0.03	+0.06	2,389	WVS and EVS, without the religious-faith item
Poland	Europe & Central Asia	+0.10	+0.06	+0.13	3,604	WVS and EVS, without the religious-faith item
Portugal	Europe & Central Asia	+0.01	−0.04	+0.07	1,944	WVS and EVS, without the religious-faith item
Puerto Rico	Latin America & Caribbean	+0.02	−0.04	+0.08	1,049	WVS and EVS, without the religious-faith item
Romania	Europe & Central Asia	−0.02	−0.05	+0.01	4,544	WVS and EVS, without the religious-faith item
Russia	Europe & Central Asia	0.00	−0.04	+0.03	4,620	WVS and EVS, without the religious-faith item
Rwanda	Sub-Saharan Africa	+0.04	0.00	+0.08	2,555	WVS and EVS, without the religious-faith item
Serbia	Europe & Central Asia	0.00	−0.03	+0.04	3,743	WVS and EVS, without the religious-faith item
Singapore	East Asia & Pacific	+0.03	−0.02	+0.08	1,745	WVS and EVS, without the religious-faith item
Slovakia	Europe & Central Asia	+0.03	−0.01	+0.07	3,270	WVS and EVS, without the religious-faith item
Slovenia	Europe & Central Asia	+0.09	+0.06	+0.13	3,093	WVS and EVS, without the religious-faith item
South Africa	Sub-Saharan Africa	−0.01	−0.05	+0.02	5,594	WVS and EVS, without the religious-faith item
South Korea	East Asia & Pacific	+0.03	−0.01	+0.07	3,640	WVS and EVS, without the religious-faith item
Spain	Europe & Central Asia	+0.14	+0.11	+0.17	4,304	WVS and EVS, without the religious-faith item
Sweden	Europe & Central Asia	+0.04	0.00	+0.07	4,254	WVS and EVS, without the religious-faith item
Switzerland	Europe & Central Asia	+0.15	+0.13	+0.18	5,134	WVS and EVS, without the religious-faith item
Taiwan ROC	Not classified	−0.01	−0.04	+0.03	3,563	WVS and EVS, without the religious-faith item
Tajikistan	Europe & Central Asia	−0.04	−0.10	+0.01	1,200	WVS and EVS, without the religious-faith item
Thailand	East Asia & Pacific	+0.05	+0.02	+0.08	3,763	WVS and EVS, without the religious-faith item
Trinidad and Tobago	Latin America & Caribbean	0.00	−0.05	+0.06	1,213	WVS and EVS, without the religious-faith item
Tunisia	Middle East & North Africa	−0.07	−0.12	−0.02	1,680	WVS and EVS, without the religious-faith item
Turkey	Europe & Central Asia	+0.07	+0.04	+0.10	4,505	WVS and EVS, without the religious-faith item
Ukraine	Europe & Central Asia	+0.01	−0.03	+0.05	4,577	WVS and EVS, without the religious-faith item
United Kingdom	Europe & Central Asia	+0.13	+0.10	+0.16	6,155	WVS and EVS, without the religious-faith item
United States	North America	+0.22	+0.19	+0.25	5,900	WVS and EVS, without the religious-faith item
Uruguay	Latin America & Caribbean	+0.12	+0.08	+0.16	2,662	WVS and EVS, without the religious-faith item
Uzbekistan	Europe & Central Asia	+0.01	−0.07	+0.09	615	WVS and EVS, without the religious-faith item
Venezuela	Latin America & Caribbean	+0.01	−0.04	+0.07	1,190	WVS and EVS, without the religious-faith item
Vietnam	East Asia & Pacific	+0.12	+0.06	+0.17	1,408	WVS and EVS, without the religious-faith item
Zambia	Sub-Saharan Africa	−0.04	−0.11	+0.02	959	WVS and EVS, without the religious-faith item
Zimbabwe	Sub-Saharan Africa	+0.03	−0.01	+0.08	2,627	WVS and EVS, without the religious-faith item
Albania	Europe & Central Asia	0.00	−0.08	+0.09	1,041	ESS, conservation values
Austria	Europe & Central Asia	+0.22	+0.19	+0.24	7,715	ESS, conservation values
Belgium	Europe & Central Asia	+0.08	+0.06	+0.09	13,202	ESS, conservation values
Bulgaria	Europe & Central Asia	−0.23	−0.25	−0.21	9,818	ESS, conservation values
Croatia	Europe & Central Asia	+0.15	+0.12	+0.18	5,353	ESS, conservation values
Cyprus	Europe & Central Asia	−0.02	−0.06	+0.02	4,044	ESS, conservation values
Czechia	Europe & Central Asia	−0.23	−0.25	−0.21	13,804	ESS, conservation values
Denmark	Europe & Central Asia	+0.12	+0.09	+0.14	8,952	ESS, conservation values
Estonia	Europe & Central Asia	−0.06	−0.08	−0.05	12,434	ESS, conservation values
Finland	Europe & Central Asia	+0.10	+0.08	+0.12	14,133	ESS, conservation values
France	Europe & Central Asia	+0.20	+0.19	+0.22	14,593	ESS, conservation values
Germany	Europe & Central Asia	+0.16	+0.15	+0.18	18,649	ESS, conservation values
Greece	Europe & Central Asia	+0.10	+0.07	+0.13	6,131	ESS, conservation values
Hungary	Europe & Central Asia	−0.01	−0.04	+0.01	10,753	ESS, conservation values
Iceland	Europe & Central Asia	+0.10	+0.07	+0.14	3,146	ESS, conservation values
Ireland	Europe & Central Asia	+0.15	+0.13	+0.17	15,004	ESS, conservation values
Israel	Middle East & North Africa	+0.09	+0.07	+0.11	10,886	ESS, conservation values
Italy	Europe & Central Asia	+0.06	+0.03	+0.09	6,404	ESS, conservation values
Kosovo	Europe & Central Asia	+0.10	0.00	+0.19	809	ESS, conservation values
Latvia	Europe & Central Asia	−0.02	−0.07	+0.03	2,290	ESS, conservation values
Lithuania	Europe & Central Asia	+0.01	−0.01	+0.04	8,096	ESS, conservation values
Montenegro	Europe & Central Asia	+0.03	−0.04	+0.10	1,613	ESS, conservation values
Netherlands	Europe & Central Asia	+0.14	+0.12	+0.16	13,145	ESS, conservation values
North Macedonia	Europe & Central Asia	−0.02	−0.12	+0.08	879	ESS, conservation values
Norway	Europe & Central Asia	+0.08	+0.05	+0.10	11,607	ESS, conservation values
Poland	Europe & Central Asia	+0.13	+0.11	+0.15	9,787	ESS, conservation values
Portugal	Europe & Central Asia	+0.07	+0.04	+0.09	10,831	ESS, conservation values
Romania	Europe & Central Asia	−0.06	−0.12	0.00	1,464	ESS, conservation values
Russia	Europe & Central Asia	−0.04	−0.07	−0.02	7,993	ESS, conservation values
Serbia	Europe & Central Asia	0.00	−0.06	+0.07	1,274	ESS, conservation values
Slovakia	Europe & Central Asia	−0.19	−0.21	−0.16	8,503	ESS, conservation values
Slovenia	Europe & Central Asia	+0.14	+0.12	+0.16	8,061	ESS, conservation values
Spain	Europe & Central Asia	+0.20	+0.19	+0.22	11,913	ESS, conservation values
Sweden	Europe & Central Asia	+0.04	+0.02	+0.06	10,896	ESS, conservation values
Switzerland	Europe & Central Asia	+0.22	+0.20	+0.24	11,856	ESS, conservation values
Turkey	Europe & Central Asia	+0.13	+0.08	+0.18	1,885	ESS, conservation values
Ukraine	Europe & Central Asia	−0.05	−0.09	−0.01	4,541	ESS, conservation values
United Kingdom	Europe & Central Asia	+0.17	+0.15	+0.19	14,926	ESS, conservation values
Albania	Europe & Central Asia	+0.04	−0.04	+0.13	1,035	ESS, conservation values net of age, sex and education
Austria	Europe & Central Asia	+0.18	+0.15	+0.21	5,705	ESS, conservation values net of age, sex and education
Belgium	Europe & Central Asia	+0.08	+0.05	+0.10	9,741	ESS, conservation values net of age, sex and education
Bulgaria	Europe & Central Asia	−0.09	−0.12	−0.06	7,237	ESS, conservation values net of age, sex and education
Switzerland	Europe & Central Asia	+0.18	+0.16	+0.20	8,462	ESS, conservation values net of age, sex and education
Cyprus	Europe & Central Asia	+0.05	0.00	+0.10	2,137	ESS, conservation values net of age, sex and education
Czechia	Europe & Central Asia	−0.14	−0.16	−0.12	11,912	ESS, conservation values net of age, sex and education
Germany	Europe & Central Asia	+0.14	+0.12	+0.16	13,439	ESS, conservation values net of age, sex and education
Denmark	Europe & Central Asia	+0.06	+0.03	+0.09	5,987	ESS, conservation values net of age, sex and education
Estonia	Europe & Central Asia	−0.04	−0.06	−0.02	10,008	ESS, conservation values net of age, sex and education
Spain	Europe & Central Asia	+0.18	+0.16	+0.21	8,217	ESS, conservation values net of age, sex and education
Finland	Europe & Central Asia	+0.08	+0.06	+0.10	10,694	ESS, conservation values net of age, sex and education
France	Europe & Central Asia	+0.17	+0.15	+0.19	10,749	ESS, conservation values net of age, sex and education
United Kingdom	Europe & Central Asia	+0.15	+0.12	+0.17	10,388	ESS, conservation values net of age, sex and education
Greece	Europe & Central Asia	+0.05	+0.01	+0.08	4,368	ESS, conservation values net of age, sex and education
Croatia	Europe & Central Asia	+0.14	+0.11	+0.18	4,179	ESS, conservation values net of age, sex and education
Hungary	Europe & Central Asia	+0.04	+0.01	+0.06	8,354	ESS, conservation values net of age, sex and education
Ireland	Europe & Central Asia	+0.10	+0.08	+0.12	11,842	ESS, conservation values net of age, sex and education
Israel	Middle East & North Africa	+0.10	+0.08	+0.13	8,536	ESS, conservation values net of age, sex and education
Iceland	Europe & Central Asia	+0.09	+0.06	+0.13	3,130	ESS, conservation values net of age, sex and education
Italy	Europe & Central Asia	+0.05	+0.02	+0.08	6,305	ESS, conservation values net of age, sex and education
Lithuania	Europe & Central Asia	+0.03	0.00	+0.05	8,031	ESS, conservation values net of age, sex and education
Latvia	Europe & Central Asia	−0.12	−0.21	−0.02	638	ESS, conservation values net of age, sex and education
Montenegro	Europe & Central Asia	+0.05	−0.02	+0.12	1,580	ESS, conservation values net of age, sex and education
North Macedonia	Europe & Central Asia	−0.03	−0.13	+0.06	867	ESS, conservation values net of age, sex and education
Netherlands	Europe & Central Asia	+0.13	+0.11	+0.15	9,653	ESS, conservation values net of age, sex and education
Norway	Europe & Central Asia	+0.07	+0.04	+0.09	8,655	ESS, conservation values net of age, sex and education
Poland	Europe & Central Asia	+0.14	+0.11	+0.16	7,022	ESS, conservation values net of age, sex and education
Portugal	Europe & Central Asia	+0.07	+0.04	+0.10	7,603	ESS, conservation values net of age, sex and education
Serbia	Europe & Central Asia	+0.07	0.00	+0.14	1,265	ESS, conservation values net of age, sex and education
Russia	Europe & Central Asia	0.00	−0.04	+0.04	5,059	ESS, conservation values net of age, sex and education
Sweden	Europe & Central Asia	+0.03	0.00	+0.05	7,774	ESS, conservation values net of age, sex and education
Slovenia	Europe & Central Asia	+0.15	+0.12	+0.17	5,947	ESS, conservation values net of age, sex and education
Slovakia	Europe & Central Asia	−0.12	−0.15	−0.08	5,338	ESS, conservation values net of age, sex and education
Ukraine	Europe & Central Asia	−0.03	−0.08	+0.02	2,309	ESS, conservation values net of age, sex and education
Kosovo	Europe & Central Asia	+0.11	+0.01	+0.20	809	ESS, conservation values net of age, sex and education

The United States (+0.28 (95% CI 0.25 to 0.31)+0.28 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.25 to 0.31n5,900Weightsweight (S017) within the countryPopulationrespondents in United States with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2011, 2017, waves pooledSourceWorld Values Survey and EVS) and Canada (+0.25 (95% CI 0.22 to 0.28)+0.25 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.22 to 0.28n5,642Weightsweight (S017) within the countryPopulationrespondents in Canada with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2020, waves pooledSourceWorld Values Survey and EVS) have the largest correlations, followed by Luxembourg (+0.22+0.22 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.16 to 0.28n1,256Weightsweight (S017) within the countryPopulationrespondents in Luxembourg with a left-right answer and all four child-quality answersPeriodfieldwork 2008, waves pooledSourceWorld Values Survey and EVS), the Netherlands (+0.20+0.20 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.18 to 0.22n7,886Weightsweight (S017) within the countryPopulationrespondents in Netherlands with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2008, 2012, 2017, 2022, waves pooledSourceWorld Values Survey and EVS), Spain (+0.20+0.20 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.17 to 0.23n4,298Weightsweight (S017) within the countryPopulationrespondents in Spain with a left-right answer and all four child-quality answersPeriodfieldwork 2007, 2008, 2011, 2017, waves pooledSourceWorld Values Survey and EVS) and Andorra (+0.20+0.20 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.15 to 0.24n1,824Weightsweight (S017) within the countryPopulationrespondents in Andorra with a left-right answer and all four child-quality answersPeriodfieldwork 2005, 2018, waves pooledSourceWorld Values Survey and EVS). Poland is at +0.19+0.19 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.16 to 0.22n3,599Weightsweight (S017) within the countryPopulationrespondents in Poland with a left-right answer and all four child-quality answersPeriodfieldwork 2005, 2008, 2012, 2017, waves pooledSourceWorld Values Survey and EVS, Germany at +0.12+0.12 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.09 to 0.14n8,805Weightsweight (S017) within the countryPopulationrespondents in Germany with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2008, 2013, 2017, 2018, waves pooledSourceWorld Values Survey and EVS, Sweden at +0.04+0.04 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI 0.00 to 0.07n4,254Weightsweight (S017) within the countryPopulationrespondents in Sweden with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2009, 2011, 2017, waves pooledSourceWorld Values Survey and EVS and India at +0.01 (95% CI −0.01 to 0.04)+0.01 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI −0.01 to 0.04n5,826Weightsweight (S017) within the countryPopulationrespondents in India with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2012, 2023, waves pooledSourceWorld Values Survey and EVS, an interval that includes zero. The correlation is negative in Czechia (−0.06−0.06 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI −0.09 to −0.02n3,615Weightsweight (S017) within the countryPopulationrespondents in Czechia with a left-right answer and all four child-quality answersPeriodfieldwork 2008, 2017, 2022, waves pooledSourceWorld Values Survey and EVS), Bulgaria (−0.06−0.06 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI −0.10 to −0.02n2,665Weightsweight (S017) within the countryPopulationrespondents in Bulgaria with a left-right answer and all four child-quality answersPeriodfieldwork 2006, 2008, 2017, waves pooledSourceWorld Values Survey and EVS) and Egypt (−0.14−0.14 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with the full indexInterval95% CI −0.17 to −0.10n4,245Weightsweight (S017) within the countryPopulationrespondents in Egypt with a left-right answer and all four child-quality answersPeriodfieldwork 2008, 2013, 2018, waves pooledSourceWorld Values Survey and EVS).

The median also differs by the income groups of Figure 2: +0.10 (95% CI 0.07 to 0.15)+0.10 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.07 to 0.15Weightsweight (S017) within each country; every country counted once across countriesPopulationhigh-income countries with a usable correlationPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 44 countries. in high-income countries, +0.04 (95% CI 0.02 to 0.05)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.02 to 0.05Weightsweight (S017) within each country; every country counted once across countriesPopulationupper-income countries with a usable correlationPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 28 countries. in upper-middle-income countries and +0.02 (95% CI 0.00 to 0.03)+0.02 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.00 to 0.03Weightsweight (S017) within each country; every country counted once across countriesPopulationlower-income countries with a usable correlationPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 30 countries. in lower-income countries. High-income countries are separated from upper-middle-income countries (a difference in medians of +0.061 (95% CI 0.020 to 0.116)+0.061 correlation coefficient, from −1 to +1Estimanddifference between the median correlations of the two income groups, high-income minus upper-middle-income (bootstrap that redraws the countries of each group)Interval95% CI 0.020 to 0.116Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation and an income groupPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 72 countries.), and upper-middle-income and lower-income countries are not separated from each other (+0.024 (95% CI −0.001 to 0.043)+0.024 correlation coefficient, from −1 to +1Estimanddifference between the median correlations of the two income groups, upper-middle-income minus lower-income (bootstrap that redraws the countries of each group)Interval95% CI −0.001 to 0.043Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation and an income groupPeriodfieldwork 2005-2023SourceWorld Values Survey and EVSNoteBased on 58 countries., an interval that includes zero). The three groups hold 102102 countries and territoriesEstimandcount of countries with a usable correlation and an income group in the WVS context fileWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS of the 103103 countries and territoriesEstimandcount of countries with a usable correlation between left-right and the full indexWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries, because Taiwan ROC has no income group in the context file. Income group also tracks region, religion and party system, so the comparison describes where the correlation is larger and does not say why.

The ESS views of Figure 3 use the neighboring measure, the importance of conformity, tradition and security, which comes from the circle of Schwartz's basic valuesSchwartz basic valuesTen broad motivations, such as security, tradition, achievement and universalism, that Shalom Schwartz's theory arranges in a circle. Neighboring values are compatible and opposite values conflict. [8]. Its median correlation with left-right is +0.08 (95% CI 0.01 to 0.11)+0.08 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and Schwartz conservation values (bootstrap over countries)Interval95% CI 0.01 to 0.11n38Weightsanalysis weight within each country; every country counted oncePopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey across 3838 countriesEstimandcount of countries with a usable correlationWeightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey countries, positive in 2727 countriesEstimandcount of countries where the point estimate is above zeroWeightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey and with an interval below zero in 77 countriesEstimandcount of countries where the 95% interval is below zeroWeightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey. The country pattern resembles the WVS one. The correlation is +0.20+0.20 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI 0.19 to 0.22n11,913Weightsanalysis weight within the countryPopulationrespondents in Spain with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey in Spain and +0.22+0.22 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI 0.19 to 0.24n7,715Weightsanalysis weight within the countryPopulationrespondents in Austria with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey in Austria, +0.04+0.04 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI 0.02 to 0.06n10,896Weightsanalysis weight within the countryPopulationrespondents in Sweden with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey in Sweden, and negative in Czechia (−0.23−0.23 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI −0.25 to −0.21n13,804Weightsanalysis weight within the countryPopulationrespondents in Czechia with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey), Bulgaria (−0.23−0.23 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI −0.25 to −0.21n9,818Weightsanalysis weight within the countryPopulationrespondents in Bulgaria with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey) and Slovakia (−0.19−0.19 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI −0.21 to −0.16n8,503Weightsanalysis weight within the countryPopulationrespondents in Slovakia with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey).

All 77 countriesEstimandcount of countries where the 95% interval is below zeroWeightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey ESS countries with an interval below zero are post-communist, and 1010 countriesEstimandcount of post-communist countries where the point estimate is below zero (post-communist: governed by a communist party before 1990 in eastern Europe, the former Soviet Union or the former Yugoslavia)Weightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey of the 1818 countriesEstimandcount of post-communist countries with a usable correlation (post-communist: governed by a communist party before 1990 in eastern Europe, the former Soviet Union or the former Yugoslavia)Weightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey post-communist countries have a negative point estimate, against 11 countriesEstimandcount of other countries where the point estimate is below zero (post-communist: governed by a communist party before 1990 in eastern Europe, the former Soviet Union or the former Yugoslavia)Weightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey of the 2020 countriesEstimandcount of other countries with a usable correlation (post-communist: governed by a communist party before 1990 in eastern Europe, the former Soviet Union or the former Yugoslavia)Weightsnone (count)PopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey other countries. Poland, Slovenia and Croatia are post-communist and have positive correlations in both surveys, so the region is not uniform. Work on post-communist publics finds that having lived under communism is associated with more support for state welfare and less support for markets [9], and this chapter does not test whether that legacy explains the negative signs.

Faith alone gives the same median as the full index

The WVS index contains religious faith, and faith alone has an interval above zero in 5252 countries and territoriesEstimandcount of countries where the 95% interval of the correlation with the religious faith alone is above zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries. Figure 4 recomputes the median country correlation for ten definitions of the measure and for the full index with age, sex and education held fixed. Without faith, the median is +0.03 (95% CI 0.01 to 0.05)+0.03 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the without faith: obedience minus independence and tolerance (bootstrap over countries)Interval95% CI 0.01 to 0.05n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS against +0.04 (95% CI 0.03 to 0.06)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index: obedience and faith minus independence and tolerance (bootstrap over countries)Interval95% CI 0.03 to 0.06n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS with it, a difference of +0.010 (95% CI 0.005 to 0.027)+0.010 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation with the full index minus the median for the without faith: obedience minus independence and tolerance, on the countries usable under both (bootstrap over countries, one draw applied to both)Interval95% CI 0.005 to 0.027n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation under both definitionsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS. Without faith the correlation is positive in 8080 countries and territoriesEstimandcount of countries where the point estimate for the without faith: obedience minus independence and tolerance is above zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries, and the interval is above zero in 4141 countries and territoriesEstimandcount of countries where the 95% interval of the correlation with the without faith: obedience minus independence and tolerance is above zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries against 4949 countries and territoriesEstimandcount of countries where the 95% interval is above zeroWeightsnone (count)Populationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS with it. The country-by-country correlations with and without faith are close (+0.93 (95% CI 0.90 to 0.95)+0.93 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the two sets of country correlations (Fisher z interval)Interval95% CI 0.90 to 0.95n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with usable correlations on both measuresPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS across countries), so the ordering of countries does not depend on the faith item.

Figure 4

Religious faith alone gives a median correlation of +0.04, the same as the full index (+0.04), and independence or tolerance alone give +0.02 and +0.02

Median across countries of the correlation of left-right with each definition of the measure, with 95% bootstrap intervals over countries; WVS and EVS fieldwork 2005-2023

• Full index
• Without faith
• Single item
• Age, sex, education fixed

Show data

Data for Religious faith alone gives a median correlation of +0.04, the same as the full index (+0.04), and independence or tolerance alone give +0.02 and +0.02

Measure	Definition	Median correlation	Interval low	Interval high	Countries	Interval above zero	Interval includes zero	Interval below zero	Correlation across countries with the full index
Full index (obedience and faith vs independence and tolerance)	Full index: obedience and faith minus independence and tolerance	+0.04	+0.03	+0.06	103	49	49	5	1.00
Full index, imagination for tolerance	Imagination version: obedience and faith minus independence and imagination	+0.04	+0.03	+0.07	103	52	44	7	0.95
No faith item	Without faith: obedience minus independence and tolerance	+0.03	+0.01	+0.05	103	41	57	5	0.93
No faith, imagination added	Without faith: obedience minus independence, tolerance and imagination	+0.03	+0.02	+0.06	103	43	53	7	0.91
Obedience minus independence	Without faith: obedience minus independence	+0.03	+0.02	+0.04	103	37	59	7	0.87
Obedience only	Obedience alone	+0.03	+0.02	+0.04	104	37	64	3	0.65
Faith only	Religious faith alone	+0.04	+0.03	+0.06	104	52	48	4	0.80
Independence only (reversed)	Independence alone, reversed	+0.02	+0.01	+0.03	104	31	63	10	0.76
Tolerance only (reversed)	Tolerance and respect alone, reversed	+0.02	+0.01	+0.03	104	32	66	6	0.59
Imagination only (reversed)	Imagination alone, reversed	+0.02	+0.01	+0.04	104	41	52	11	0.66
Full index, net of age, sex, education	Full index net of age, sex and education	+0.05	+0.03	+0.06	103	49	51	3	0.99

Religious faith alone gives a median of +0.04 (95% CI 0.03 to 0.06)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the religious faith alone (bootstrap over countries)Interval95% CI 0.03 to 0.06n104Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, and the difference between the full index and faith alone (+0.006 (95% CI −0.008 to 0.020)+0.006 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation with the full index minus the median for the religious faith alone, on the countries usable under both (bootstrap over countries, one draw applied to both)Interval95% CI −0.008 to 0.020n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation under both definitionsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS) includes zero. Obedience alone gives +0.03 (95% CI 0.02 to 0.04)+0.03 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the obedience alone (bootstrap over countries)Interval95% CI 0.02 to 0.04n104Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, above the three autonomy items alone. Independence (+0.02 (95% CI 0.01 to 0.03)+0.02 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the independence alone, reversed (bootstrap over countries)Interval95% CI 0.01 to 0.03n104Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS), tolerance (+0.02 (95% CI 0.01 to 0.03)+0.02 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the tolerance and respect alone, reversed (bootstrap over countries)Interval95% CI 0.01 to 0.03n104Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS) and imagination (+0.02 (95% CI 0.01 to 0.04)+0.02 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the imagination alone, reversed (bootstrap over countries)Interval95% CI 0.01 to 0.04n104Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS) alone give lower medians, each scored so that a higher value marks the obedience pole, and the median for the full index is higher than each of them by +0.027 (95% CI 0.016 to 0.041)+0.027 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation with the full index minus the median for the independence alone, reversed, on the countries usable under both (bootstrap over countries, one draw applied to both)Interval95% CI 0.016 to 0.041n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation under both definitionsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, +0.026 (95% CI 0.013 to 0.044)+0.026 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation with the full index minus the median for the tolerance and respect alone, reversed, on the countries usable under both (bootstrap over countries, one draw applied to both)Interval95% CI 0.013 to 0.044n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation under both definitionsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and +0.024 (95% CI 0.004 to 0.039)+0.024 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation with the full index minus the median for the imagination alone, reversed, on the countries usable under both (bootstrap over countries, one draw applied to both)Interval95% CI 0.004 to 0.039n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with a usable correlation under both definitionsPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS. The version that uses imagination in place of tolerance gives +0.04 (95% CI 0.03 to 0.07)+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the imagination version: obedience and faith minus independence and imagination (bootstrap over countries)Interval95% CI 0.03 to 0.07n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, close to the full index.

Age, sex and education change little in the WVS

Age, sex and education could stand behind the correlation if they relate to both the measure and left-right placement. In the WVS and EVS, holding them fixed leaves the median at +0.05 (95% CI 0.03 to 0.06)+0.05 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index net of age, sex and education (bootstrap over countries)Interval95% CI 0.03 to 0.06n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, and the adjusted and unadjusted country correlations are nearly the same (+0.99 (95% CI 0.98 to 0.99)+0.99 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the two sets of country correlations (Fisher z interval)Interval95% CI 0.98 to 0.99n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with usable correlations on both measuresPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS). In the ESS the adjustment changes individual countries more: the correlation in Czechia moves from −0.23−0.23 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with Schwartz conservation valuesInterval95% CI −0.25 to −0.21n13,804Weightsanalysis weight within the countryPopulationrespondents in Czechia with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey to −0.14 (95% CI −0.16 to −0.12)−0.14 correlation coefficient, from −1 to +1Estimandweighted partial correlation of left-right with Schwartz conservation values, holding age, sex and education fixedInterval95% CI −0.16 to −0.12n11,912Weightsanalysis weight within the countryPopulationrespondents in Czechia with a left-right answer and conservation scorePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey, and Czechia, Bulgaria and Slovakia stay below zero, while the median is +0.07 (95% CI 0.05 to 0.10)+0.07 correlation coefficient, from −1 to +1Estimandmedian across countries of the partial correlation of left-right with Schwartz conservation values, holding age, sex and education fixed (bootstrap over countries)Interval95% CI 0.05 to 0.10n36Weightsanalysis weight within each country; every country counted oncePopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey against +0.08+0.08 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and Schwartz conservation values (bootstrap over countries)Interval95% CI 0.01 to 0.11n38Weightsanalysis weight within each country; every country counted oncePopulationESS countries with an effective sample of 300 or morePeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey (the last view of Figure 3).

The surveys agree as closely on age and religiosity

Figure 5 sets the WVS and EVS correlation for each country against the ESS one. Across 3636 countriesEstimandcount of countries in both surveysWeightsnone (count)Populationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey countries in both surveys the correlation between the two sets is +0.81 (95% CI 0.66 to 0.90)+0.81 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations with left-right (Fisher z interval)Interval95% CI 0.66 to 0.90n36Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey, the two surveys give a correlation of the same sign in 3030 countriesEstimandcount of countries where the two surveys give a correlation of the same signWeightsnone (count)Populationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey countries and the mean absolute difference is 0.060.06 absolute difference in correlation unitsEstimandmean across countries of the absolute difference between the two surveys' correlationsWeightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey. Czechia and Bulgaria are negative in both. Slovakia and Hungary are close to zero or slightly positive in the WVS and lower in the ESS. The interval on the cross-country correlation is wide, because it rests on only 3636 countriesEstimandcount of countries in both surveysWeightsnone (count)Populationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey countries.

Figure 5

The two surveys rank countries alike, with a correlation of +0.81 across 36 countries

Correlation of left-right with the WVS and EVS obedience index (horizontal) and with the ESS conservation values (vertical), countries in both surveys

Show data

Data for The two surveys rank countries alike, with a correlation of +0.81 across 36 countries

Country	WVS and EVS correlation	WVS interval low	WVS interval high	ESS correlation	ESS interval low	ESS interval high	WVS respondents	ESS respondents	Smaller of the two counts
Albania	+0.06	+0.02	+0.11	0.00	−0.08	+0.09	2,253	1,041	1,041
Austria	+0.17	+0.13	+0.21	+0.22	+0.19	+0.24	2,685	7,715	2,685
Belgium	+0.07	+0.01	+0.12	+0.08	+0.06	+0.09	1,422	13,202	1,422
Bulgaria	−0.06	−0.10	−0.02	−0.23	−0.25	−0.21	2,665	9,818	2,665
Croatia	+0.16	+0.12	+0.21	+0.15	+0.12	+0.18	2,196	5,353	2,196
Cyprus	−0.01	−0.05	+0.03	−0.02	−0.06	+0.02	2,894	4,044	2,894
Czechia	−0.06	−0.09	−0.02	−0.23	−0.25	−0.21	3,615	13,804	3,615
Denmark	+0.09	+0.06	+0.13	+0.12	+0.09	+0.14	4,688	8,952	4,688
Estonia	−0.01	−0.05	+0.02	−0.06	−0.08	−0.05	3,319	12,434	3,319
Finland	+0.11	+0.07	+0.15	+0.10	+0.08	+0.12	2,955	14,133	2,955
France	+0.13	+0.10	+0.17	+0.20	+0.19	+0.22	3,862	14,593	3,862
Germany	+0.12	+0.09	+0.14	+0.16	+0.15	+0.18	8,805	18,649	8,805
Greece	+0.18	+0.14	+0.22	+0.10	+0.07	+0.13	2,201	6,131	2,201
Hungary	+0.07	+0.03	+0.10	−0.01	−0.04	+0.01	3,444	10,753	3,444
Iceland	+0.06	+0.01	+0.11	+0.10	+0.07	+0.14	2,322	3,146	2,322
Italy	+0.16	+0.13	+0.19	+0.06	+0.03	+0.09	3,552	6,404	3,552
Kosovo	−0.02	−0.10	+0.06	+0.10	0.00	+0.19	662	809	662
Latvia	−0.01	−0.05	+0.04	−0.02	−0.07	+0.03	1,985	2,290	1,985
Lithuania	+0.07	+0.02	+0.11	+0.01	−0.01	+0.04	1,831	8,096	1,831
Montenegro	+0.03	−0.03	+0.10	+0.03	−0.04	+0.10	1,211	1,613	1,211
Netherlands	+0.20	+0.18	+0.22	+0.14	+0.12	+0.16	7,886	13,145	7,886
North Macedonia	+0.05	−0.03	+0.13	−0.02	−0.12	+0.08	766	879	766
Norway	+0.08	+0.04	+0.11	+0.08	+0.05	+0.10	3,154	11,607	3,154
Poland	+0.19	+0.16	+0.22	+0.13	+0.11	+0.15	3,599	9,787	3,599
Portugal	+0.05	0.00	+0.11	+0.07	+0.04	+0.09	1,942	10,831	1,942
Romania	−0.02	−0.05	+0.01	−0.06	−0.12	0.00	4,539	1,464	1,464
Russia	+0.01	−0.02	+0.05	−0.04	−0.07	−0.02	4,611	7,993	4,611
Serbia	+0.03	−0.01	+0.07	0.00	−0.06	+0.07	3,739	1,274	1,274
Slovakia	+0.03	−0.01	+0.07	−0.19	−0.21	−0.16	3,260	8,503	3,260
Slovenia	+0.17	+0.13	+0.20	+0.14	+0.12	+0.16	3,091	8,061	3,091
Spain	+0.20	+0.17	+0.23	+0.20	+0.19	+0.22	4,298	11,913	4,298
Sweden	+0.04	0.00	+0.07	+0.04	+0.02	+0.06	4,254	10,896	4,254
Switzerland	+0.15	+0.12	+0.17	+0.22	+0.20	+0.24	5,134	11,856	5,134
Turkey	+0.16	+0.13	+0.19	+0.13	+0.08	+0.18	4,467	1,885	1,885
Ukraine	+0.05	+0.01	+0.10	−0.05	−0.09	−0.01	4,565	4,541	4,541
United Kingdom	+0.13	+0.11	+0.16	+0.17	+0.15	+0.19	6,150	14,926	6,150

Both surveys correlate their measure with the same outcome, and left-right placement means different things in different countries. Agreement between the two surveys could therefore come from how the label attaches to any characteristic, and not from the measure. Figure 6 repeats the comparison for four other predictors of left-right placement, on the same countries. The correlation between the two surveys is +0.87 (95% CI 0.76 to 0.93)+0.87 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations of left-right with religiosity (Fisher z interval)Interval95% CI 0.76 to 0.93n36Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey for religiosity, +0.90 (95% CI 0.82 to 0.95)+0.90 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations of left-right with age (Fisher z interval)Interval95% CI 0.82 to 0.95n36Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey for age, +0.70 (95% CI 0.47 to 0.84)+0.70 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations of left-right with education (Fisher z interval)Interval95% CI 0.47 to 0.84n34Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey for education and +0.72 (95% CI 0.52 to 0.85)+0.72 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations of left-right with female (Fisher z interval)Interval95% CI 0.52 to 0.85n36Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey for sex, against +0.81 (95% CI 0.66 to 0.90)+0.81 correlation coefficient, from −1 to +1EstimandPearson correlation across countries of the WVS/EVS and ESS correlations with left-right (Fisher z interval)Interval95% CI 0.66 to 0.90n36Weightsweight within country in each survey; every country counted oncePopulationcountries with usable correlations in both surveysPeriodWVS/EVS fieldwork 2005-2023 and ESS rounds 3 to 10SourceEuropean Social Survey for the obedience measure. Age and religiosity give higher point estimates than the obedience measure, and no predictor's interval separates from the one for the obedience measure.

Figure 6

The two surveys agree as closely on age and religiosity as on obedience, so agreement is not specific to the measure

Correlation across countries between the WVS and EVS and the ESS correlations of left-right with each predictor; 95% Fisher z intervals

• Obedience measure
• Reference predictor

Show data

Data for The two surveys agree as closely on age and religiosity as on obedience, so agreement is not specific to the measure

Predictor	Definition	Correlation between surveys	Interval low	Interval high	Countries	Spread of WVS correlations (standard deviation)
Obedience measure	WVS and EVS full index against ESS conservation values	0.81	0.66	0.90	36	0.08
Religiosity	Importance of God in the WVS and EVS against attendance and prayer in the ESS	0.87	0.76	0.93	36	0.10
Education	Three levels	0.70	0.47	0.84	34	0.08
Age	Years	0.90	0.82	0.95	36	0.10
Sex (female)	Female against male	0.72	0.52	0.85	36	0.05

The agreement between the surveys shows that both rank countries alike on how left-right attaches to individual traits, and it does not show that both are measuring one disposition. Older people place themselves further left in Czechia and Bulgaria, where the obedience correlation is negative in both surveys, and further right in Spain and Germany, in both surveys.Correlation of left-right with age in years, WVS and EVS first and ESS second: Czechia −0.17−0.17 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (WVS/EVS)Interval95% CI −0.20 to −0.14n3,990Weightsweight (S017) within the countryPopulationrespondents in Czechia with a left-right answerPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and −0.23−0.23 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (ESS)Interval95% CI −0.25 to −0.22n14,107Weightsanalysis weight within the countryPopulationrespondents in Czechia with a left-right answerPeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey, Bulgaria −0.22−0.22 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (WVS/EVS)Interval95% CI −0.26 to −0.18n2,809Weightsweight (S017) within the countryPopulationrespondents in Bulgaria with a left-right answerPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and −0.24−0.24 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (ESS)Interval95% CI −0.27 to −0.22n9,869Weightsanalysis weight within the countryPopulationrespondents in Bulgaria with a left-right answerPeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey, Spain +0.14+0.14 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (WVS/EVS)Interval95% CI 0.12 to 0.17n4,322Weightsweight (S017) within the countryPopulationrespondents in Spain with a left-right answerPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and +0.08+0.08 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (ESS)Interval95% CI 0.06 to 0.09n14,089Weightsanalysis weight within the countryPopulationrespondents in Spain with a left-right answerPeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey, Germany +0.09+0.09 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (WVS/EVS)Interval95% CI 0.07 to 0.12n8,805Weightsweight (S017) within the countryPopulationrespondents in Germany with a left-right answerPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS and +0.11+0.11 correlation coefficient, from −1 to +1Estimandweighted correlation of left-right with age in years (ESS)Interval95% CI 0.09 to 0.12n26,424Weightsanalysis weight within the countryPopulationrespondents in Germany with a left-right answerPeriodESS rounds 3 to 10, 2006-2022SourceEuropean Social Survey. Every interval excludes zero.

The US tie to party rose, except among Black respondents

Figure 7 follows the ANES correlations from 1992 to 2024, by race in its first view and for all respondents in its second. Among all respondents the correlation with party identification is −0.04 (95% CI −0.09 to 0.00)−0.04 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included)Interval95% CI −0.09 to 0.00n2,023Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File in 1992, +0.14 (95% CI 0.10 to 0.18)+0.14 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included)Interval95% CI 0.10 to 0.18n3,600Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2016SourceANES Cumulative Data File in 2016 and +0.28 (95% CI 0.24 to 0.32)+0.28 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included)Interval95% CI 0.24 to 0.32n4,668Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in 2024, and each study after 2008 has a higher point estimate than the one before it, so the rise is a climb through successive studies and not a single step. The correlation with liberal-conservative self-placement starts higher, +0.22 (95% CI 0.17 to 0.26)+0.22 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point liberal-conservative self-placementInterval95% CI 0.17 to 0.26n1,498Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File in 1992, and reaches +0.43 (95% CI 0.39 to 0.46)+0.43 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point liberal-conservative self-placementInterval95% CI 0.39 to 0.46n4,085Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in 2024.

Figure 7

The correlation with party rose among white and Hispanic and other respondents and stayed near zero among Black respondents

Correlation of the child-rearing measure with party identification among ANES respondents, by race and study year, with 95% bands; the other views add liberal-conservative self-placement and the GSS

By raceParty, ideology and GSSNet of college, age and raceFace-to-face onlyWeb only

Show data

Data for The correlation with party rose among white and Hispanic and other respondents and stayed near zero among Black respondents

Study year	Series	Respondents	Mode	Correlation	Interval low	Interval high	Respondents in cell
1992	ANES party	All respondents	All modes	−0.04	−0.09	0.00	2,023
1992	ANES party	White non-Hispanic	All modes	+0.04	−0.01	+0.09	1,551
1992	ANES party	Black non-Hispanic	All modes	+0.02	−0.11	+0.14	261
1992	ANES party	Hispanic and other	All modes	−0.12	−0.25	+0.02	206
1992	ANES party	Net of college, age and white non-Hispanic	All modes	+0.06	+0.02	+0.11	1,971
1992	ANES ideology	All respondents	All modes	+0.22	+0.17	+0.26	1,498
1992	ANES ideology	White non-Hispanic	All modes	+0.28	+0.23	+0.33	1,217
1992	ANES ideology	Black non-Hispanic	All modes	+0.07	−0.09	+0.23	147
1992	ANES ideology	Hispanic and other	All modes	+0.14	−0.03	+0.30	131
1992	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.24	+0.19	+0.29	1,457
2000	ANES party	All respondents	All modes	0.00	−0.05	+0.06	1,527
2000	ANES party	White non-Hispanic	All modes	+0.08	+0.02	+0.14	1,184
2000	ANES party	Black non-Hispanic	All modes	−0.03	−0.21	+0.15	151
2000	ANES party	Hispanic and other	All modes	+0.02	−0.15	+0.18	183
2000	ANES party	Net of college, age and white non-Hispanic	All modes	+0.08	+0.02	+0.14	1,510
2000	ANES ideology	All respondents	All modes	+0.24	+0.15	+0.32	575
2000	ANES ideology	White non-Hispanic	All modes	+0.24	+0.15	+0.34	470
2000	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.23	+0.15	+0.32	568
2004	ANES party	All respondents	All modes	+0.05	−0.01	+0.12	1,041
2004	ANES party	White non-Hispanic	All modes	+0.17	+0.09	+0.25	741
2004	ANES party	Black non-Hispanic	All modes	0.00	−0.18	+0.18	152
2004	ANES party	Hispanic and other	All modes	+0.09	−0.10	+0.28	132
2004	ANES party	Net of college, age and white non-Hispanic	All modes	+0.14	+0.07	+0.21	1,025
2004	ANES ideology	All respondents	All modes	+0.26	+0.19	+0.33	817
2004	ANES ideology	White non-Hispanic	All modes	+0.32	+0.24	+0.39	614
2004	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.25	+0.18	+0.32	810
2008	ANES party	All respondents	All modes	+0.05	0.00	+0.11	2,050
2008	ANES party	White non-Hispanic	All modes	+0.16	+0.09	+0.23	1,022
2008	ANES party	Black non-Hispanic	All modes	0.00	−0.10	+0.10	484
2008	ANES party	Hispanic and other	All modes	+0.08	−0.03	+0.19	524
2008	ANES party	All respondents	Face-to-face	+0.05	0.00	+0.11	2,050
2008	ANES party	Net of college, age and white non-Hispanic	All modes	+0.14	+0.09	+0.20	1,994
2008	ANES ideology	All respondents	All modes	+0.27	+0.21	+0.33	1,473
2008	ANES ideology	White non-Hispanic	All modes	+0.32	+0.25	+0.38	815
2008	ANES ideology	Black non-Hispanic	All modes	+0.21	+0.08	+0.33	266
2008	ANES ideology	Hispanic and other	All modes	+0.27	+0.14	+0.40	378
2008	ANES ideology	All respondents	Face-to-face	+0.27	+0.21	+0.33	1,473
2008	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.28	+0.22	+0.34	1,433
2012	ANES party	All respondents	All modes	+0.08	+0.05	+0.12	5,453
2012	ANES party	White non-Hispanic	All modes	+0.21	+0.17	+0.25	3,255
2012	ANES party	Black non-Hispanic	All modes	−0.11	−0.21	−0.01	949
2012	ANES party	Hispanic and other	All modes	+0.06	−0.02	+0.15	1,233
2012	ANES party	All respondents	Face-to-face	+0.05	−0.02	+0.11	1,888
2012	ANES party	All respondents	Web	+0.10	+0.06	+0.14	3,565
2012	ANES party	Net of college, age and white non-Hispanic	All modes	+0.16	+0.13	+0.20	5,329
2012	ANES ideology	All respondents	All modes	+0.23	+0.20	+0.27	4,912
2012	ANES ideology	White non-Hispanic	All modes	+0.31	+0.27	+0.35	3,100
2012	ANES ideology	Black non-Hispanic	All modes	+0.12	+0.01	+0.23	752
2012	ANES ideology	Hispanic and other	All modes	+0.17	+0.08	+0.26	1,047
2012	ANES ideology	All respondents	Face-to-face	+0.20	+0.13	+0.27	1,365
2012	ANES ideology	All respondents	Web	+0.24	+0.20	+0.29	3,547
2012	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.25	+0.21	+0.28	4,805
2016	ANES party	All respondents	All modes	+0.14	+0.10	+0.18	3,600
2016	ANES party	White non-Hispanic	All modes	+0.29	+0.25	+0.33	2,601
2016	ANES party	Black non-Hispanic	All modes	+0.02	−0.11	+0.15	338
2016	ANES party	Hispanic and other	All modes	+0.03	−0.07	+0.12	643
2016	ANES party	All respondents	Face-to-face	+0.11	+0.04	+0.18	1,038
2016	ANES party	All respondents	Web	+0.15	+0.11	+0.20	2,562
2016	ANES party	Net of college, age and white non-Hispanic	All modes	+0.20	+0.16	+0.24	3,482
2016	ANES ideology	All respondents	All modes	+0.35	+0.32	+0.39	2,856
2016	ANES ideology	White non-Hispanic	All modes	+0.42	+0.38	+0.46	2,175
2016	ANES ideology	Black non-Hispanic	All modes	+0.22	+0.05	+0.38	200
2016	ANES ideology	Hispanic and other	All modes	+0.26	+0.16	+0.36	467
2016	ANES ideology	All respondents	Face-to-face	+0.32	+0.24	+0.39	821
2016	ANES ideology	All respondents	Web	+0.37	+0.32	+0.41	2,035
2016	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.34	+0.30	+0.38	2,760
2020	ANES party	All respondents	All modes	+0.21	+0.18	+0.24	7,314
2020	ANES party	White non-Hispanic	All modes	+0.40	+0.37	+0.43	5,332
2020	ANES party	Black non-Hispanic	All modes	−0.03	−0.14	+0.08	638
2020	ANES party	Hispanic and other	All modes	+0.12	+0.05	+0.20	1,280
2020	ANES party	Net of college, age and white non-Hispanic	All modes	+0.25	+0.22	+0.28	6,929
2020	ANES ideology	All respondents	All modes	+0.39	+0.36	+0.41	6,288
2020	ANES ideology	White non-Hispanic	All modes	+0.47	+0.44	+0.50	4,764
2020	ANES ideology	Black non-Hispanic	All modes	+0.16	+0.03	+0.28	425
2020	ANES ideology	Hispanic and other	All modes	+0.34	+0.26	+0.41	1,049
2020	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.37	+0.34	+0.40	5,966
2024	ANES party	All respondents	All modes	+0.28	+0.24	+0.32	4,668
2024	ANES party	White non-Hispanic	All modes	+0.43	+0.38	+0.47	3,379
2024	ANES party	Black non-Hispanic	All modes	+0.05	−0.08	+0.18	418
2024	ANES party	Hispanic and other	All modes	+0.26	+0.17	+0.35	833
2024	ANES party	All respondents	Face-to-face	+0.29	+0.21	+0.38	905
2024	ANES party	All respondents	Web	+0.26	+0.20	+0.32	1,736
2024	ANES party	Net of college, age and white non-Hispanic	All modes	+0.31	+0.27	+0.35	4,429
2024	ANES ideology	All respondents	All modes	+0.43	+0.39	+0.46	4,085
2024	ANES ideology	White non-Hispanic	All modes	+0.51	+0.47	+0.54	3,067
2024	ANES ideology	Black non-Hispanic	All modes	+0.25	+0.09	+0.39	290
2024	ANES ideology	Hispanic and other	All modes	+0.41	+0.32	+0.49	694
2024	ANES ideology	All respondents	Face-to-face	+0.45	+0.37	+0.53	784
2024	ANES ideology	All respondents	Web	+0.44	+0.38	+0.49	1,473
2024	ANES ideology	Net of college, age and white non-Hispanic	All modes	+0.41	+0.37	+0.45	3,866
1986	GSS party	All respondents	All modes	−0.03	−0.11	+0.06	728
1987	GSS party	All respondents	All modes	−0.05	−0.11	+0.01	1,776
1988	GSS party	All respondents	All modes	+0.03	−0.04	+0.10	975
1989	GSS party	All respondents	All modes	−0.04	−0.11	+0.03	985
1990	GSS party	All respondents	All modes	−0.02	−0.10	+0.06	861
1991	GSS party	All respondents	All modes	+0.03	−0.04	+0.10	970
1993	GSS party	All respondents	All modes	+0.04	−0.03	+0.11	1,016
1994	GSS party	All respondents	All modes	+0.02	−0.04	+0.07	1,914
1996	GSS party	All respondents	All modes	+0.04	−0.01	+0.09	1,856
1998	GSS party	All respondents	All modes	0.00	−0.05	+0.05	1,830
2000	GSS party	All respondents	All modes	+0.02	−0.03	+0.07	1,817
2002	GSS party	All respondents	All modes	+0.02	−0.06	+0.09	876
2004	GSS party	All respondents	All modes	−0.03	−0.11	+0.05	856
2004	GSS party	All respondents	Face-to-face	0.00	−0.09	+0.09	692
2006	GSS party	All respondents	All modes	+0.06	−0.02	+0.14	980
2006	GSS party	All respondents	Face-to-face	+0.07	−0.01	+0.15	918
2008	GSS party	All respondents	All modes	+0.08	+0.01	+0.14	1,319
2008	GSS party	All respondents	Face-to-face	+0.08	+0.01	+0.15	1,169
2010	GSS party	All respondents	All modes	−0.03	−0.10	+0.04	1,312
2010	GSS party	All respondents	Face-to-face	−0.04	−0.11	+0.03	1,170
2012	GSS party	All respondents	All modes	+0.07	0.00	+0.13	1,274
2012	GSS party	All respondents	Face-to-face	+0.06	−0.01	+0.14	1,109
2014	GSS party	All respondents	All modes	+0.04	−0.02	+0.09	1,620
2014	GSS party	All respondents	Face-to-face	+0.03	−0.03	+0.10	1,391
2016	GSS party	All respondents	All modes	+0.11	+0.05	+0.16	1,860
2016	GSS party	All respondents	Face-to-face	+0.12	+0.06	+0.18	1,735
2018	GSS party	All respondents	All modes	+0.10	+0.03	+0.16	1,470
2018	GSS party	All respondents	Face-to-face	+0.08	+0.02	+0.15	1,320
2021	GSS party	All respondents	All modes	+0.12	+0.07	+0.16	2,479
2021	GSS party	All respondents	Web	+0.12	+0.07	+0.17	2,171
2022	GSS party	All respondents	All modes	+0.12	+0.05	+0.18	2,209
2022	GSS party	All respondents	Face-to-face	+0.11	+0.01	+0.20	962
2022	GSS party	All respondents	Web	+0.13	+0.03	+0.22	1,042
2024	GSS party	All respondents	All modes	+0.13	+0.07	+0.19	2,045
2024	GSS party	All respondents	Face-to-face	+0.12	+0.01	+0.22	691
2024	GSS party	All respondents	Web	+0.18	+0.10	+0.26	1,135

The all-respondent series combines three race groups. Among white non-Hispanic respondents the correlation with party identification rose from +0.04 (95% CI −0.01 to 0.09)+0.04 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) among white non-Hispanic respondentsInterval95% CI −0.01 to 0.09n1,551Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File in 1992 to +0.43 (95% CI 0.38 to 0.47)+0.43 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) among white non-Hispanic respondentsInterval95% CI 0.38 to 0.47n3,379Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in 2024. Among Hispanic and other respondents, a residual group that grew from 1010 percent of respondentsEstimandweighted share of the party identification analysis sample who are neither white nor Black non-HispanicInterval95% CI 9 to 11n2,018Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File% to 2424 percent of respondentsEstimandweighted share of the party identification analysis sample who are neither white nor Black non-HispanicInterval95% CI 22 to 26n4,630Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File% of the weighted sample, it went from −0.12 (95% CI −0.25 to 0.02)−0.12 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) among Hispanic and other respondents (neither white nor Black non-Hispanic)Interval95% CI −0.25 to 0.02n206Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File to +0.26 (95% CI 0.17 to 0.35)+0.26 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) among Hispanic and other respondents (neither white nor Black non-Hispanic)Interval95% CI 0.17 to 0.35n833Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File.

Among Black non-Hispanic respondents the interval includes zero in 77 study yearsEstimandcount of years where the 95% interval of the correlation with party identification includes zeroWeightsnone (count)PopulationANES presidential-year studies with 100 or more Black non-Hispanic respondentsPeriod1992-2024SourceANES Cumulative Data File of 88 study yearsEstimandcountWeightsnone (count)PopulationANES presidential-year studies with 100 or more Black non-Hispanic respondentsPeriod1992-2024SourceANES Cumulative Data File studies and is below zero in the other one, 2012 (−0.11 (95% CI −0.21 to −0.01)−0.11 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) among Black non-Hispanic respondentsInterval95% CI −0.21 to −0.01n949Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2012SourceANES Cumulative Data File). In 2024 they score higher on the obedience side than white non-Hispanic respondents, 0.63 (95% CI 0.58 to 0.67)0.63 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.58 to 0.67n421Weightsweight (VCF0009z)PopulationBlack non-Hispanic adults who answered the four child-rearing pairsPeriod2024SourceANES Cumulative Data File against 0.46 (95% CI 0.44 to 0.47)0.46 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.44 to 0.47n3,390Weightsweight (VCF0009z)PopulationWhite non-Hispanic adults who answered the four child-rearing pairsPeriod2024SourceANES Cumulative Data File on the ANES index (which runs from autonomy on every pair to obedience on every pair), and 78.878.8 percent of respondentsEstimandweighted share identifying with the Democratic partyInterval95% CI 73.8 to 83.8n503Weightsweight (VCF0009z)PopulationBlack non-Hispanic adults, party identification 1 to 3 counted as Democratic (leaners included)Period2024SourceANES Cumulative Data File% of them identify as Democrats, leanersPartisan leanerA respondent who first says they are independent but, when asked, says they lean toward one party. Most figures group leaners with the party they lean toward, and each chapter says which grouping it uses. included. The rise in the all-respondent series therefore comes from the white group and the Hispanic and other group, and it is absent among Black respondents.

Holding a college degree, age and whether the respondent is white non-Hispanic fixed (a partial correlation, which removes what those three variables explain of each measure), the correlation with party identification is +0.06 (95% CI 0.02 to 0.11)+0.06 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) net of college degree, age and raceInterval95% CI 0.02 to 0.11n1,971Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod1992SourceANES Cumulative Data File in 1992 and +0.31 (95% CI 0.27 to 0.35)+0.31 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) net of college degree, age and raceInterval95% CI 0.27 to 0.35n4,429Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in 2024.

Studies disagree on whether the child-rearing scale works the same way for Black and white respondents. Pérez and Hetherington report that it does not [10]. A later study reports partial measurement invariance in the 1992 to 2008 studies, with one item, respect for elders, working differently, and full invariance in 2012 and 2016 [11]. This chapter reports the groups separately and does not test invariance itself.

The ANES 2012 and 2016 studies have separate face-to-face and web samples, and the 2020 and 2024 studies mix interview modesSurvey modeHow respondents were interviewed: in person, by phone, on paper or online. Mode can change answers, so a change in mode can look like a change in opinion.. The 2020 study has no mode weight, and in 2024 4343 percent of respondents (unweighted count)Estimandshare of analytic respondents with neither a face-to-face nor a web weight (panel and paper cases)n4,668Weightsnone (count)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File% of the respondents in the party analysis (2,0272,027 respondentsEstimandcount of analytic respondents with neither a face-to-face nor a web weightWeightsnone (count)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File of 4,6684,668 respondentsEstimandcount of analytic respondents in the party identification correlationWeightsnone (count)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File) have neither a face-to-face nor a web weight, because they come from a panel or a paper questionnaire, so they are in neither mode estimate. Among respondents with a mode weight, the 2024 correlation with party identification is +0.29 (95% CI 0.21 to 0.38)+0.29 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) in the face-to-face sampleInterval95% CI 0.21 to 0.38n905Weightsweight_ftf (face-to-face)Populationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in the face-to-face sample and +0.26 (95% CI 0.20 to 0.32)+0.26 correlation coefficient, from −1 to +1Estimandweighted correlation of the ANES child-rearing index with 7-point party identification (leaners included) in the web sampleInterval95% CI 0.20 to 0.32n1,736Weightsweight_webPopulationadults who answered the four child-rearing pairs and the outcomePeriod2024SourceANES Cumulative Data File in the web sample, so the high 2024 value does not come from one mode.

The GSS gives a second series that does not use the four forced choices. Its measure is the rank of obeying against the rank of thinking for oneself, and its correlation with party identification is 0.00 (95% CI −0.03 to 0.02)0.00 correlation coefficient, from −1 to +1Estimandweighted correlation of obedience ranked above thinking for oneself (rank difference) with 7-point party identificationInterval95% CI −0.03 to 0.02n9,225Weightsweight_usePopulationadults ranking the GSS child qualities, party identification 0 to 6Period1986-1994SourceGeneral Social Survey in 1986 to 1994 and +0.11 (95% CI 0.08 to 0.14)+0.11 correlation coefficient, from −1 to +1Estimandweighted correlation of obedience ranked above thinking for oneself (rank difference) with 7-point party identificationInterval95% CI 0.08 to 0.14n10,063Weightsweight_usePopulationadults ranking the GSS child qualities, party identification 0 to 6Period2016-2024SourceGeneral Social Survey in 2016 to 2024.The share ranking obedience above thinking for oneself is +6.9 (95% CI 4.3 to 9.4)+6.9 percentage pointsEstimanddifference in the share ranking obedience above thinking for oneselfInterval95% CI 4.3 to 9.4n7,914Weightsweight_usePopulationadults ranking the GSS child qualities, Republican minus Democratic identifiers (leaners included)Period2016-2024SourceGeneral Social Survey pointsPercentage pointThe unit for the difference between two percentages. A change from one share to another is stated in points, not in percent, so that absolute and relative change are not confused. higher among Republican than among Democratic identifiers, leaners included, in 2016 to 2024, against −1.0 (95% CI −3.3 to 1.4)−1.0 percentage pointsEstimanddifference in the share ranking obedience above thinking for oneselfInterval95% CI −3.3 to 1.4n8,103Weightsweight_usePopulationadults ranking the GSS child qualities, Republican minus Democratic identifiers (leaners included)Period1986-1994SourceGeneral Social Survey in 1986 to 1994.

The correlation with party rose while the average score on the measure fell. The mean score on the ANES index is 0.60 (95% CI 0.59 to 0.62)0.60 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.59 to 0.62n2,036Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairsPeriod1992SourceANES Cumulative Data File in 1992, 0.56 (95% CI 0.55 to 0.57)0.56 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.55 to 0.57n3,616Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairsPeriod2016SourceANES Cumulative Data File in 2016 and 0.49 (95% CI 0.48 to 0.50)0.49 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.48 to 0.50n4,690Weightsweight (VCF0009z)Populationadults who answered the four child-rearing pairsPeriod2024SourceANES Cumulative Data File in 2024, so respondents chose the obedience answer less often over the period. The fall also appears within the face-to-face sample, where the mean is 0.610.61 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.59 to 0.63n1,907Weightsweight_ftf (face-to-face)Populationadults in the face-to-face sample who answered the four child-rearing pairsPeriod2012SourceANES Cumulative Data File in 2012, 0.580.58 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.55 to 0.60n1,050Weightsweight_ftf (face-to-face)Populationadults in the face-to-face sample who answered the four child-rearing pairsPeriod2016SourceANES Cumulative Data File in 2016 and 0.480.48 position on the 0 to 1 index (1 = obedience on every pair)Estimandweighted mean of the ANES child-rearing indexInterval95% CI 0.45 to 0.51n916Weightsweight_ftf (face-to-face)Populationadults in the face-to-face sample who answered the four child-rearing pairsPeriod2024SourceANES Cumulative Data File in 2024. Hetherington and Weiler argue that the parties came to differ on worldview [3]. The data here show the correlation and its timing, and they are consistent with that argument and with others.

The obedience side is more morally conservative and less tolerant

Figure 8 compares respondents on the autonomy side of the index with those on the obedience side, country by country and then averaged across a fixed panel of countries. Averaged across countries, 70.970.9 percent of respondentsEstimandmean across countries of the weighted share a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce, respondents on the obedience side of the full index (bootstrap over countries)Interval95% CI 67.6 to 74.2n97Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS% of respondents on the obedience side are socially conservative, meaning that they average 44 points on the 1 to 10 justifiability scaleEstimandcutoff on the mean of the three justifiability itemsWeightsnone (constant)Populationdefinition of socially conservativePeriodfieldwork 2005-2023SourceWorld Values Survey and EVS or lower on the three justifiability items of the ten-point scale, against 57.157.1 percent of respondentsEstimandmean across countries of the weighted share a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce, respondents on the autonomy side of the full index (bootstrap over countries)Interval95% CI 51.9 to 62.0n97Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS% on the autonomy side, a gap of 13.8 (95% CI 11.4 to 16.5)13.8 percentage pointsEstimandmean across countries of the share a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 11.4 to 16.5n97Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points. Naming homosexuals as unwanted neighbors is 7.7 (95% CI 6.1 to 9.4)7.7 percentage pointsEstimandmean across countries of the share naming homosexuals as unwanted neighbors, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 6.1 to 9.4n100Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points more common on the obedience side, naming immigrants or foreign workers 2.8 (95% CI 1.8 to 4.0)2.8 percentage pointsEstimandmean across countries of the share naming immigrants or foreign workers as unwanted neighbors, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 1.8 to 4.0n102Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points and naming people of a different race 2.9 (95% CI 1.8 to 3.9)2.9 percentage pointsEstimandmean across countries of the share naming people of a different race as unwanted neighbors, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 1.8 to 3.9n102Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points. Calling a strong leader who bypasses parliament a good way to run the country is 4.2 (95% CI 2.5 to 6.1)4.2 percentage pointsEstimandmean across countries of the share saying that a strong leader who does not bother with parliament and elections is a good or very good way to run the country, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 2.5 to 6.1n99Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points more common and calling army rule good is 4.3 (95% CI 3.1 to 5.4)4.3 percentage pointsEstimandmean across countries of the share saying that army rule is a good or very good way to run the country, obedience side minus autonomy side of the full index (bootstrap over countries)Interval95% CI 3.1 to 5.4n99Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points more common.

Figure 8

Every attitude is more common on the obedience side of the index, and the gap is largest for moral conservatism

Share holding each attitude among respondents on the autonomy side and the obedience side of the index, average of country shares (each country counted once); 95% bootstrap intervals; WVS and EVS fieldwork 2005-2023

• Autonomy side of the index
• Obedience side of the index

Show data

Data for Every attitude is more common on the obedience side of the index, and the gap is largest for moral conservatism

Attitude	Definition	Autonomy side, percent	Autonomy interval low	Autonomy interval high	Obedience side, percent	Obedience interval low	Obedience interval high	Gap, points	Countries
Socially conservative (homosexuality, abortion, divorce)	Mean of 4 or lower on the 1 to 10 justifiability of homosexuality, abortion and divorce	57.1	51.9	62.0	70.9	67.6	74.2	13.8	97
Does not want homosexuals as neighbors	Names homosexuals as people they would not want as neighbors	46.4	41.1	51.5	54.1	49.5	58.7	7.7	100
Does not want immigrants as neighbors	Names immigrants or foreign workers as people they would not want as neighbors	22.7	20.1	25.3	25.5	22.9	28.4	2.8	102
Does not want people of another race as neighbors	Names people of a different race as people they would not want as neighbors	17.9	15.5	20.4	20.8	18.5	23.3	2.9	102
Strong leader bypassing parliament is good	Says a strong leader who does not bother with parliament and elections is a good or very good way to run the country	45.0	41.0	48.8	49.2	45.6	52.7	4.2	99
Army rule is good	Says army rule is a good or very good way to run the country	25.3	22.0	28.8	29.5	26.4	32.9	4.3	99

The moral gap is smaller without the faith item (Figure 9), and the comparison needs the same countries under both indices. The index without faith leaves fewer respondents on its obedience side, so fewer countries qualify (7474 countries and territoriesEstimandcount of countries with an effective sample of 100 or more on both sides of the indexWeightsnone (count)Populationfixed panel for this outcome and indexPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS against 9797 countries and territoriesEstimandcount of countries with an effective sample of 100 or more on both sides of the indexWeightsnone (count)Populationfixed panel for this outcome and indexPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS for moral conservatism). On those 7474 countries and territoriesEstimandcount of countries with an effective sample of 100 or more on both sides of the indexWeightsnone (count)Populationfixed panel for this outcome and indexPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries the gap for moral conservatism is 12.0 (95% CI 9.1 to 15.1)12.0 percentage pointsEstimandmean across countries of the share a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce, obedience side minus autonomy side of the full index, on the countries that also qualify without the religious-faith item (bootstrap over countries)Interval95% CI 9.1 to 15.1n74Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points with the faith item and 7.2 (95% CI 4.8 to 9.9)7.2 percentage pointsEstimandmean across countries of the share a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce, obedience side minus autonomy side of the index without the religious-faith item, on the countries that qualify under both indices (bootstrap over countries)Interval95% CI 4.8 to 9.9n74Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points without it, a difference of 4.7 (95% CI 3.7 to 5.8)4.7 percentage pointsEstimandmean across countries of the gap under the full index minus the gap under the index without the religious-faith item, for a mean of 4 or lower on the justifiability of homosexuality, abortion and divorce (bootstrap over countries, one draw applied to both)Interval95% CI 3.7 to 5.8n74Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries that qualify under both indices for this outcomePeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points. The gap for naming homosexuals as neighbors falls from 6.9 (95% CI 5.1 to 8.7)6.9 percentage pointsEstimandmean across countries of the share naming homosexuals as unwanted neighbors, obedience side minus autonomy side of the full index, on the countries that also qualify without the religious-faith item (bootstrap over countries)Interval95% CI 5.1 to 8.7n77Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS to 5.0 (95% CI 3.3 to 6.6)5.0 percentage pointsEstimandmean across countries of the share naming homosexuals as unwanted neighbors, obedience side minus autonomy side of the index without the religious-faith item, on the countries that qualify under both indices (bootstrap over countries)Interval95% CI 3.3 to 6.6n77Weightsweight (S017) within each country; every country counted once across countriesPopulationrespondents in the fixed panel of countries, with an index score other than zeroPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points (a difference of 1.9 (95% CI 1.1 to 2.6)1.9 percentage pointsEstimandmean across countries of the gap under the full index minus the gap under the index without the religious-faith item, for naming homosexuals as unwanted neighbors (bootstrap over countries, one draw applied to both)Interval95% CI 1.1 to 2.6n77Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries that qualify under both indices for this outcomePeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points). For immigrants, people of a different race and army rule, the difference between the two indices has an interval that includes zero, and the gap for a strong leader is larger without the faith item (a difference of −1.2 (95% CI −2.1 to −0.2)−1.2 percentage pointsEstimandmean across countries of the gap under the full index minus the gap under the index without the religious-faith item, for saying that a strong leader who does not bother with parliament and elections is a good or very good way to run the country (bootstrap over countries, one draw applied to both)Interval95% CI −2.1 to −0.2n73Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries that qualify under both indices for this outcomePeriodfieldwork 2005-2023SourceWorld Values Survey and EVS points, taken as with the faith item minus without it).

Figure 9

Dropping the faith item shrinks the gaps for moral conservatism and for homosexuals as neighbors, and leaves the gaps for immigrants, race, army rule and a strong leader no smaller

Gap between the obedience side and the autonomy side of the index in the share holding each attitude, with and without the religious-faith item, on the countries that qualify under both; 95% bootstrap intervals; WVS and EVS fieldwork 2005-2023

• Without the faith item
• With the faith item

Show data

Data for Dropping the faith item shrinks the gaps for moral conservatism and for homosexuals as neighbors, and leaves the gaps for immigrants, race, army rule and a strong leader no smaller

Attitude	Definition	Gap without the faith item, points	Interval low	Interval high	Gap with the faith item, points	Interval low	Interval high	Difference, with minus without, points	Difference interval low	Difference interval high	Countries
Socially conservative (homosexuality, abortion, divorce)	Mean of 4 or lower on the 1 to 10 justifiability of homosexuality, abortion and divorce	7.2	4.8	9.9	12.0	9.1	15.1	+4.7	+3.7	+5.8	74
Does not want homosexuals as neighbors	Names homosexuals as people they would not want as neighbors	5.0	3.3	6.6	6.9	5.1	8.7	+1.9	+1.1	+2.6	77
Army rule is good	Says army rule is a good or very good way to run the country	3.7	2.1	5.3	3.9	2.5	5.3	+0.2	−0.7	+1.0	73
Does not want people of another race as neighbors	Names people of a different race as people they would not want as neighbors	2.3	1.1	3.5	2.5	1.3	3.6	+0.1	−0.5	+0.8	80
Does not want immigrants as neighbors	Names immigrants or foreign workers as people they would not want as neighbors	2.3	1.1	3.7	2.2	1.1	3.3	−0.2	−1.0	+0.6	80
Strong leader bypassing parliament is good	Says a strong leader who does not bother with parliament and elections is a good or very good way to run the country	4.2	1.8	6.7	3.1	0.8	5.4	−1.2	−2.1	−0.2	73

Across countries, the median correlation of the full index with a moral-conservatism index (the reversed justifiability of homosexuality, abortion and divorce) is +0.13 (95% CI 0.10 to 0.16)+0.13 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation of the full index with a moral-conservatism index (justifiability of homosexuality, abortion and divorce, reversed)Interval95% CI 0.10 to 0.16n102Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with usable correlations for all three measures with moral attitudesPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS, and it is +0.09 (95% CI 0.07 to 0.12)+0.09 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation of the index without the religious-faith item with a moral-conservatism index (justifiability of homosexuality, abortion and divorce, reversed)Interval95% CI 0.07 to 0.12n102Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with usable correlations for all three measures with moral attitudesPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS without the faith item, against +0.07 (95% CI 0.03 to 0.10)+0.07 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation of left-right self-placement with a moral-conservatism index (justifiability of homosexuality, abortion and divorce, reversed)Interval95% CI 0.03 to 0.10n102Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with usable correlations for all three measures with moral attitudesPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS for left-right placement with the same moral index. The measure is more closely tied to moral attitudes than to left-right placement, where the median is +0.04+0.04 correlation coefficient, from −1 to +1Estimandmedian across countries of the correlation between left-right and the full index (bootstrap over countries)Interval95% CI 0.03 to 0.06n103Weightsweight (S017) within each country; every country counted once across countriesPopulationcountries with an effective sample of 300 or more on the measure and left-rightPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS.

Every gap stays above zero under both indices. The moral outcome is more common on the obedience side in 8787 countries and territoriesEstimandcount of countries where the outcome is more common on the obedience sideWeightsnone (count)Populationfixed panel for this outcome and indexPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS of 9797 countries and territoriesEstimandcount of countries with an effective sample of 100 or more on both sides of the indexWeightsnone (count)Populationfixed panel for this outcome and indexPeriodfieldwork 2005-2023SourceWorld Values Survey and EVS countries. These are differences between groups of respondents, averaged across countries, and show what occurs together. Chapter 6.7, Support for democracy takes up support for strong rule directly, and Chapter 8.5, Order, security and threat (planned) asks whether security values and threat perceptions lean right.

What these data cannot show

The correlations come from one interview per person, so they do not show whether valuing obedience shapes where people place themselves, whether placement or party shapes the child-rearing answers, or whether a third influence shapes both. The rise in the American correlation with party is a change in how the measure lines up with party, and the ANES and GSS are repeated cross-sections that cannot say whether people changed parties, the parties changed or the composition of each party changed.

The measures are related but different. The WVS checklist, the ANES forced choices, the GSS ranking and the ESS values battery use different formats, and the ESS has no question about children. The agreement between the WVS and ESS country patterns shows that both rank countries alike, and the comparison with age, education, sex and religiosity shows that this agreement is not specific to the obedience measure. This chapter does not use party manifestos, expert placements of parties or vote choices, so it cannot say why the label attaches to conservation values in one country and not in another.

Whether the ANES child-rearing items mean the same thing to Black and white respondents is disputed [10] [11], and this chapter reports the groups separately without testing it. The Hispanic and other group is a residual category whose makeup changes across studies. The intervals are model based and ignore the sampling design, and the countries are those the surveys reached, not a sample of the world. The cohort profile cannot separate age from birth decade, the country groups by income are a single classification from the WVS context file, and waves are pooled inside each country, so a correlation blends years. The link between the disposition and politics is thought to depend on perceived threat [1]. This chapter does not test whether threat conditions the link; Chapter 8.5, Order, security and threat (planned) takes up the threat measures in the WVS and ESS.

---

Data sources

• World Values Survey and EVSWorld Values Survey and European Values Study, fieldwork 2005-2023: child-quality checklist, left-right, moral, neighbor and political-system itemsYears: 1981-2023 (7 unified waves) · Weights: weight (S017) within each country; every country counted once across countries · License: Free download with registration; users must cite the release.
• European Social SurveyEuropean Social Survey rounds 3 to 10: Schwartz conservation values and left-rightYears: 2002-2023 (rounds 1 to 11; round 11 was fielded in 2023 and 2024) · Weights: analysis weight within each country · License: Open access after free registration in the ESS Data Portal; follow the ESS data citation requirements.
• ANES Cumulative Data FileANES Cumulative Data File 1992-2024, presidential-year studies: forced-choice child-rearing pairs, party identification, ideologyYears: 1948-2024 (33 study years; even years only after 2004; no 2006, 2010 or 2014 study) · Weights: weight (VCF0009z); weight_ftf and weight_web for the mode views · License: Free download after accepting the ANES terms of use; cite the release.
• General Social SurveyGeneral Social Survey 1986-2024: obey and think-for-oneself rankings, party identificationYears: 1972-2024 (35 study years) · Weights: weight_use · License: See the provider's terms of use.

All data sources

Methods

Sample. World Values Survey and European Values Study (WVS/EVS), fieldwork in 2005 or later (waves 5 to 7 and the EVS 2008 and 2017 rounds): 112 countries and territories with fieldwork in 2005 or later (118 in the full 1981 to 2023 file, after restoring the ten unlabelled EVS countries from the country code); 107 had enough respondents to test and 103 have a usable correlation, on 281,678 respondents in those 103 countries with a left-right answer (1 to 10, 10 = right), a positive weight and all four child-quality answers. The cleaned file carries ten of the child qualities. The cleaned WVS and EVS file leaves the country name blank for Austria, Belgium, Denmark, Iceland, Ireland, Luxembourg, Malta, Portugal, Kosovo and Northern Cyprus; the country code restores those 10 labels, and 8 of them have a usable correlation. European Social Survey (ESS), rounds 3 to 10 (2006-2022; round 11 carries no values battery in the cleaned file): 38 countries with a usable correlation. American National Election Studies (ANES) Cumulative File, the eight presidential-year studies 1992-2024 that field the four forced-choice child-rearing pairs (the file carries none of them for 1996), adults with all four answers. General Social Survey (GSS), 1986-2024, adults who ranked both obeying and thinking for oneself. ANES race groups are the file's three categories: white non-Hispanic, Black non-Hispanic and other (Hispanic, Asian and every other group). Measures. WVS/EVS: obedience and religious faith minus independence and tolerance and respect for other people (each 1 if mentioned as an important child quality, 0 if not; A042, A040, A029, A035), range −2 to +2, complete cases; robustness versions drop the faith item, swap tolerance for imagination (A034), use pairs of items or single items. ANES: one minus the mean of four forced choices coded 0, 0.5 and 1 (VCF9246 to VCF9249), 0 to 1. GSS: the rank of thinking for oneself minus the rank of obeying. ESS: the mean-rating-centred Schwartz conservation composite (conformity, tradition, security). The sign of every item is fixed from the wording, not from a correlation with the outcome. Estimand. A correlation is the survey-weighted Pearson correlation between the measure and left-right placement (WVS/EVS 1 to 10, ESS 0 to 10), party identification (ANES 1 to 7, GSS 0 to 6, leaners included, other-party respondents dropped) or liberal-conservative self-placement (ANES 1 to 7), computed inside a country or a study year with waves pooled. Net-of-covariate versions correlate the residuals from weighted least squares on age, sex and three education levels (WVS/EVS and ESS), or on a college degree, age and whether the respondent is white non-Hispanic (ANES). Weights. WVS/EVS weight (S017) within a country and every country counted once across countries; ESS analysis weight; ANES weight (VCF0009z), with weight_ftf and weight_web for the mode views; GSS weight_use. Intervals. 95% Fisher z intervals on the Kish effective sample size of the complete cases, which are model based and ignore strata and clusters; partial correlations use the effective sample minus the number of covariates; medians and means across countries and the country gaps have percentile intervals from a bootstrap over countries (2,000 replicates, seed 12345), that is, the list of countries is redrawn with replacement and the statistic recomputed; a difference between two medians for the same countries (index definitions) uses one draw of countries for both, and a difference between two income groups redraws the countries of each group on its own; correlations across countries use Fisher z (a transformation that makes the sampling error of a correlation nearly constant) on the number of countries; gaps between two GSS groups combine standard errors in quadrature. Suppression. A WVS/EVS or ESS country cell needs an effective sample of at least 300; an ANES or GSS year cell needs at least 100 respondents and an effective sample of 100; shares by index side need an effective sample of 100 on each side; the cohort panels need 100 in every birth decade. Country groups and cohort panels. Income groups are the World Bank groups in the WVS context file, with lower middle income and low income combined as lower income. Birth decade is the survey year minus age. The cohort panels are fixed sets of countries with enough respondents in every birth decade: 36 high-income, 22 upper-middle-income and 10 lower-income countries. Across the country-by-decade cells of the panels, the same birth decade measured in 2014 to 2023 rather than 2005 to 2013 differs by 3.8 points on average in absolute value. Outcome gaps. Respondents on the obedience side score above zero on the index and those on the autonomy side below zero; each country's share is computed first and countries are averaged with equal weight over a fixed panel of countries with enough respondents on both sides. The full index is shown on every country that qualifies under it and, beside the index without the faith item, on the countries that qualify under both indices, because the index without faith leaves fewer respondents on its obedience side. The difference between the two gaps has a bootstrap interval that redraws the same countries for both. Socially conservative means a mean of 4 or lower on the three 1-to-10 justifiability items (homosexuality, abortion, divorce) with all three answered. Mode. ANES 2008 is a single-mode face-to-face study, 2012, 2016 and 2024 have face-to-face and web weights, and 2020 has none; in 2024 the panel and paper cases have neither weight. The GSS 2021 wave is mostly web and 2022 and 2024 mix in-person and web interviews. The mode views report both modes where a cell has 100 respondents. Post-communist countries. Post-communist ESS countries are those governed by a communist party before 1990 in eastern Europe, the former Soviet Union and the former Yugoslavia (ESS codes AL, BG, BY, CZ, EE, GE, HR, HU, LT, LV, ME, MK, PL, RO, RS, RU, SI, SK, UA, XK); Germany is not classed as post-communist because the ESS reports it as one country.

References

1. Stenner, Karen (2005). The Authoritarian Dynamic. Cambridge University Press. doi:10.1017/CBO9780511614712
2. Feldman, Stanley (2003). Enforcing Social Conformity: A Theory of Authoritarianism. Political Psychology 24(1): 41-74. doi:10.1111/0162-895X.00316
3. Hetherington, Marc J.; Weiler, Jonathan D. (2009). Authoritarianism and Polarization in American Politics. Cambridge University Press. www.cambridge.org/core/books/authoritarianism-and-polarization-in-american-politics/C0831CCE83C7C2D506A0954A8956BAE8
4. Engelhardt, Andrew M.; Feldman, Stanley; Hetherington, Marc J. (2023). Advancing the Measurement of Authoritarianism. Political Behavior 45: 537-560. doi:10.1007/s11109-021-09718-6
5. Inglehart, Ronald; Baker, Wayne E. (2000). Modernization, Cultural Change, and the Persistence of Traditional Values. American Sociological Review 65(1): 19-51. doi:10.1177/000312240006500103
6. Kish, Leslie (1965). Survey Sampling. John Wiley and Sons.
7. Fisher, Ronald A. (1921). On the Probable Error of a Coefficient of Correlation Deduced from a Small Sample. Metron 1(4): 3-32.
8. Schwartz, Shalom H. (1992). Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries. Advances in Experimental Social Psychology 25: 1-65. doi:10.1016/S0065-2601(08)60281-6
9. Pop-Eleches, Grigore; Tucker, Joshua A. (2017). Communism's Shadow: Historical Legacies and Contemporary Political Attitudes. Princeton University Press. doi:10.23943/princeton/9780691175591.001.0001
10. Pérez, Efrén O.; Hetherington, Marc J. (2014). Authoritarianism in Black and White: Testing the Cross-Racial Validity of the Child Rearing Scale. Political Analysis 22(3): 398-412. doi:10.1093/pan/mpu002
11. Fialho, Fabrício M. (2026). Reassessing Racial Differences in Authoritarianism in the United States: Measurement Invariance and Group Comparisons Using the ANES Child-Rearing Values Scale. Public Opinion Quarterly. doi:10.1093/poq/nfag020

How to cite

@misc{singh2026theobedienceautonomydisposition,
  author = {Singh, Vishal},
  title = {The obedience-autonomy disposition},
  year = {2026},
  url = {https://vishalsingh.org/politics/the-obedience-autonomy-disposition},
  note = {In The Public Mind. Published 30 September 2026.}
}

Reading paths

• Outside the United States