CoChat
Sign Up
PreprintCited by 484Mental Health Treatment and AccessGender, Health, and Social InequalityGender Roles and Identity Studies

The gender gap in adolescent mental health: a cross-national investigation of 566,827 adolescents across 73 countries

Open access
Campbell, Olympia, Bann, David, Patalay, Praveetha·2020·SSM - Population Health
DOI 10.1101/2020.06.12.20129312PMC PMC7960541
Open this paper in CoChat — free accountOpen PDF

Abstract

Mental ill-health is a leading cause of disease burden worldwide. While women suffer from greater levels of mental health disorders, it remains unclear whether this gender gap differs systematically across regions and/or countries, or across the different dimensions of mental health. We analysed 2018 data from 566,827 adolescents across 73 countries for 4 mental health outcomes: psychological distress, life satisfaction, eudaemonia, and hedonia. We examine average gender differences and distributions for each of these outcomes as well as country-level associations between each outcome and purported determinants at the country level: wealth (GDP per capita), inequality (Gini index), and societal indicators of gender inequality (GII, GGGI, and GSNI). We report four main results: 1) The gender gap in mental health in adolescence is largely ubiquitous cross-culturally, with girls having worse average mental health; 2) There is considerable cross-national heterogeneity in the size of the gender gap, with the direction reversed in a minority of countries; 3) Higher GDP per capita is associated with worse average mental health and a larger gender gap across all mental health outcomes; and 4) more gender equal countries have larger gender gaps, with better mental health in males associated with greater gender equality and inconsistent associations with different mental health outcomes in females. Taken together, our findings suggest that while the gender gap appears largely ubiquitous, its size differs considerably by region, country, and dimension of mental health. Findings point to the hitherto unrealised complex nature of gender disparities in mental health and possible incongruence between expectations and reality in high gender equal countries.

AI snapshot

Written from the paper’s full text.

TL;DR

Using PISA 2018 data from 566,829 adolescents in 73 countries/economies, girls reported worse average mental health than boys across life satisfaction, psychological distress, hedonia, and eudaemonia, though the size and sometimes direction of the gap varied cross-nationally. Higher GDP per capita and greater gender equality were associated with larger gender gaps in adolescent mental health.

Key findings

  • Girls had worse average mental health across all four outcomes. The largest mean differences were for life satisfaction (0.41; 0.33 SD) and psychological distress (−1.1; 0.34 SD), with smaller gaps for hedonia (0.10; 0.39 SD) and eudaemonia (0.15; 0.27 SD).
  • The gender gap was heterogeneous across countries: heterogeneity was strong for every outcome (I² > 95%, p < 0.001), and the direction was reversed in a minority of countries, such as Jordan and Saudi Arabia for hedonia and eudaemonia.
  • Higher GDP per capita was associated with worse average mental health and a larger gender gap across outcomes; for example, life satisfaction β = −0.032 per 10⁴ GDP per capita, psychological distress β = 0.028, and eudaemonia β = −0.033.
  • More gender-equal countries had larger gender gaps across all outcomes; GGGI × female interactions were significant for life satisfaction (−0.081), psychological distress (0.095), hedonia (−0.16), and eudaemonia (−0.065).

Methods

  • Cross-sectional analysis of 2018 Programme for International Student Assessment (PISA) data: 73 countries/economies, 566,829 students (49.8% girls, 50.2% boys), age 15, using PISA’s two-stage sample design and weights. Countries excluded for not collecting mental health measures were Singapore, Norway, New Zealand, and Israel; non-nationally representative subsamples such as China were dropped.
  • Outcomes were life satisfaction (0–10), psychological distress (4–16), hedonia (5–20), and eudaemonia (3–12), each z-score standardised. Country-level indicators were GDP per capita, Gini, and three gender-equality measures: GII, GSNI, and GGGI.
  • Analyses included weighted country-level gender differences, meta-analytic I² tests for heterogeneity, Pearson correlations, and multilevel linear regressions with random intercepts for countries and random slopes for gender; sensitivity analyses tested quadratic associations, GSNI instead of GGGI, individual-level controls, and country outliers.

Limitations

  • The paper states that the study relies exclusively on cross-sectional cross-country correlations, so causal pathways cannot be strongly inferred; it also notes possible cultural differences in Likert-scale responding, possible reporting bias if boys are less willing to report negative mental health, possible gender differences in school attendance, and that gender was measured as binary, excluding non-binary identities.
  • The paper states that PISA does not collect mental health data from African countries apart from Morocco, so that region is largely absent from the analysis.
  • Evident from the text: the country-level economic and gender-equality indicators are not adolescent-specific, and the ecological cross-country design cannot identify individual-level mechanisms behind the observed gender gaps.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1101/2020.06.12.20129312

Cites

69 of 69 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 51 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    More gender equal countries have larger gender gaps across all mental health outcomes.

    “more gender equal countries have larger gender gaps across all mental health outcomes”
    Quote in the paperSubstantial claim
  • Findingcitation

    Income inequality has an inconsistent and weak relationship with mental health outcomes.

    “In line with previous literature we find an inconsistent and weak relationship between income inequality and mental health outcomes (Ngamaba et al., 2018)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ngamaba et al., 2018

      Not checked in the free run.

  • Findingown data

    GGGI was associated with female hedonia (β=0.02 per 0.1 unit increase).

    “β0.18 [0.054se] in males, 0.02 in females for hedonia”
    Quote in the paperNot a substantial claim
  • Findingown data

    The association between GGGI and female life satisfaction was weakly positive.

    “apart from the association between GGGI and female life satisfaction which was weakly positive”
    Quote in the paperNot a substantial claim
  • Findingown data

    Richer European countries including Scandinavian nations such as Sweden and Finland have some of the largest gender gaps in mental health.

    “Perhaps counterintuitively, richer European countries including the Scandinavian nations, such as Sweden and Finland, have some of the largest gender gaps in mental health.”
    Quote in the paperSubstantial claim
  • May combine two claims. 50
  • The quote only partly states this claim. 3
  • Classified as probably not a substantial claim. 32

  • May combine two claims.

    Findingown data

    There is considerable cross-national heterogeneity in the size of the gender gap, with the direction reversed in a minority of countries.

    “There is considerable cross-national heterogeneity in the size of the gender gap, with the direction reversed in a minority of countries”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The gender gap in adolescent mental health is largely ubiquitous cross-culturally, with girls having worse average mental health.

    “The gender gap in mental health in adolescence is largely ubiquitous cross-culturally, with girls having worse average mental health”
    • Also stated in: § Do girls have worse average mental health than boys across all outcomes? · plain
      “On average, girls have worse mental health across all indicators (Table 1).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Higher GDP per capita is associated with worse average mental health and a larger gender gap across all mental health outcomes.

    “Higher GDP per capita is associated with worse average mental health and a larger gender gap across all mental health outcomes”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Higher income inequality was associated with marginally more psychological distress for both genders, but this association is slightly stronger for boys than girls, and thus more equal countries have larger gender gaps.

    “Higher income inequality was associated with marginally more psychological distress for both genders (β [difference in outcome per 1 unit increase in Gini] 0.0007 [0.004se], p>0.05), but this association is slightly stronger for boys than girls and thus more equal countries have larger gender gaps (Fig. 2: b2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The outcomes vary in their distributions and where in the distribution the gender gap appears, indicating that mean differences are driven by different parts of the mental health distribution for different outcomes.

    “The outcomes vary in their distributions and where in the distribution the gender gap appears, indicating that mean differences are driven by different parts of the mental health distribution for the different outcomes.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Individual-level correlations between mental health outcomes are weak-moderate; none reach 0.5.

    “The correlation matrix shows that individual-level correlations between mental health outcomes are weak-moderate - none reach 0.5 (Table 1, top half).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Eudaemonia peaks at 9 for both boys and girls in most of the countries and the gender difference looks uniform across the distribution.

    “Eudaemonia peaks at 9 for both boys and girls in most of the countries and the gender difference looks uniform across the distribution (Fig. S4).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Higher GDP per capita was associated with lower hedonia for both boys and girls, but this association was not statistically significant (p>0.05).

    “hedonia (β−0.023 [0.012se], p>0.05)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There is considerable cross-cultural variation in the size of the average gender difference, but it appears largely ubiquitous in this global sample, particularly for life satisfaction and psychological distress.

    “Whilst there is considerable cross-cultural variation in the size of this average difference, it appears largely ubiquitous in this global sample - particularly for life satisfaction and psychological distress.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Higher income inequality was associated with slightly lower life satisfaction for boys and slightly higher life satisfaction for girls, and thus a slightly smaller gender gap in more unequal countries.

    “Higher income inequality was associated with slightly lower life satisfaction for boys and slightly higher life satisfaction for girls and thus a slightly smaller gender gap in more unequal countries (Fig. 2: a2).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Hedonia is largely left skewed and the distributional gender differences are less pronounced.

    “Hedonia is also largely left skewed (Fig. S4) and the distributional gender differences are less pronounced.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Hedonia and eudaemonia show greater cross-cultural variation with some countries exhibiting worse average outcomes for boys, such as Jordan and Saudi Arabia.

    “Hedonia and eudaemonia show greater cross-cultural variation with some countries exhibiting worse average outcomes for boys, such as Jordan and Saudi Arabia (Fig. 1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The final model explains 37.5% of the between country variance in life satisfaction, 12.33% in psychological distress, 17.8% in hedonia, and 46.4% in eudaemonia.

    “Overall, the final model explains 37.5% of the between country variance in life satisfaction, 12.33% in psychological distress, 17.8% in hedonia, and 46.4% in eudaemonia.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Hedonia and eudaemonia have smaller gender gaps, 0.10 (0.39 s.d) and 0.15 (0.27 s.d) respectively.

    “whereas hedonia and eudaemonia have smaller gender gaps, 0.10 (0.39 s.d) and 0.15 (0.27 s.d) respectively.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The newly created GSNI does not appear to measure gender equality in a qualitatively different way than the GII, as they are highly correlated.

    “The newly created gender social norms index (GSNI), despite attempting to capture the distinct attitudinal aspects of gender equality, does not appear to measure gender equality in a qualitatively different way than the GII as they are highly correlated.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Lower income inequality was associated with lower hedonia and eudaemonia and slightly larger gender gaps.

    “By contrast, lower income inequality was associated with lower hedonia (β0.009 [0.004se], p<0.05) and eudaemonia (β0.003 [0.003se], p>0.05) and slightly larger gender gaps (Fig. 3: c2 & d2).”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingnone given

    Income inequality is associated with a wider gender gap in all cases.

    “although it is associated with a wider gender gap in all cases.”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    GGGI was associated with female psychological distress (β=0.026 per 0.1 unit increase).

    “-0.069 [0.06se] in males and 0.026 in females for PD”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The proportion of total variance attributable to differences between countries was estimated to be 5.6% for life satisfaction, hedonia and eudaemonia and 7.3% for psychological distress.

    “The proportion of total variance attributable to differences between countries was estimated to be 5.6% for life satisfaction, hedonia and eudaemonia and 7.3% for psychological distress (using the variance partition coefficient from the baseline multi-level model (Table 2 Model A).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Life satisfaction and psychological distress have the largest mean gender differences between the sexes, 0.41 (0.33 s.d) and −1.1 (0.34 s.d) respectively.

    “Life satisfaction and psychological distress have the largest mean differences between the sexes, 0.41 (0.33 s.d) and −1.1 (0.34 s.d) respectively”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    More gender equal countries had larger gender gaps across all outcomes examined, consistent with previous literature in adults.

    “More gender equal countries had larger gender gaps across all outcomes examined, consistent with previous literature in adults (Zuckerman et al., 2017).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Zuckerman et al., 2017

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Quadratic country-level associations were largely similar to linear associations, with richer and more gender equal countries having larger gaps.

    “we plot quadratic country-level associations (Figs. S6 and 7) and find that inferences drawn are largely similar to the linear associations, with richer and more gender equal countries having larger gaps.”
    Quote in the paperNot a substantial claim

  • May combine two claims. The quote only partly states this claim.

    Findingown data

    Higher GDP per capita was associated with lower life satisfaction, lower eudaemonia, and higher psychological distress for both boys and girls.

    “Higher GDP per capita was associated with lower life satisfaction (β [difference in outcome per 10^4 increase in GDP per capita] 0.032 [0.01se], p<0.01), hedonia (β−0.023 [0.012se], p>0.05) and eudaemonia (β0.033 [0.009se], p<0.001) and higher psychological distress (β0.028 [0.013se], p<0.05) for both boys and girls (Figs. 2 and 3 a1-d1, Table 2 model C).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For all outcomes except hedonia, the gender gap was larger for wealthier nations, mainly driven by steeper slopes for females.

    “For all outcomes (except hedonia) the gender gap was larger for wealthier nations mainly driven by steeper slopes for females.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Countries with worse society gender equality scores, such as Jordan, Saudi Arabia, and Lebanon, have some of the smallest gender gaps, and the direction of the gap is sometimes reversed with boys having worse mental health.

    “By contrast, countries with worse society gender equality scores – such as Jordan, Saudi Arabia, and Lebanon - have some of the smallest gender gaps and the direction of the gap is sometimes reversed (with boys having worse mental health).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For each outcome there was strong evidence for heterogeneity in the gender differences, both within and between regions, with I2 > 95% for all outcomes, p <0.001.

    “For each outcome there was strong evidence for heterogeneity in the gender differences - both within and between regions with I2 > 95% for all outcomes, p <0.001 (Fig. S1).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The direction and size of the gender gap and its distribution vary across mental health outcomes and across countries.

    “but the direction and size of the gender gap and distribution varies across a range of mental health outcomes and a large sample of countries.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingnone given

    The female to male ratio in secondary enrolment ranges from 0.9 to 1.1 for our whole sample, apart from Germany (0.89), the Philippines (1.19) and Qatar (1.25).

    “The female to male ratio in secondary enrolment ranges from 0.9 to 1.1 for our whole sample, apart from Germany (0.89), the Philippines (1.19) and Qatar (1.25).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    During childhood and adolescence girls tend to report substantially worse internalising mental health than boys and this gender gap increases with age during adolescence.

    “During childhood and adolescence girls tend to report substantially worse internalising mental health than boys and this gender gap increases with age during adolescence”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Bolognini et al., 1996

      Not checked in the free run.

    • cites Bradshaw et al., 2013

      Not checked in the free run.

    • cites Cavallo et al., 2006

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    We adjust the final student weights by the number of individuals in each cluster divided by the sum of the sampling weights in each cluster in order to estimate multi-level models.

    “Using weight scaling method A proposed by Asparouhov (Asparouhov, 2006) and Carle (Carle, 2009) we adjust the final student weights by the number of individuals in each cluster divided by the sum of the sampling weights in each cluster (see (Carle, 2009), Appendix B), in order to estimate multi-level models.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Asparouhov, 2006

      Not checked in the free run.

    • cites Carle, 2009

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    To test if findings are robust to the removal of country outliers, we calculate cook's distance of countries for single-level models, with countries as data points and GGGI, GDP per capita, and Gini as independent variables, and the outcome variable as the average gender gap in each mental health outcome.

    “To test if findings are robust to the removal of country outliers, we calculate cook’s distance of countries for single-level models, with countries as data points and GGGI, GDP per capita, and Gini as independent variables, and the outcome variable as the average gender gap in each mental health outcome.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Invariance testing showed that measures were invariant by gender, region and gender x region.

    “Invariance testing showed that measures were invariant by gender, region and gender x region (Table S2).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    We use a single indicator of gender equality to avoid multicollinearity with other equality measures.

    “We use a single indicator of gender equality to avoid multicollinearity with other equality measures (Table S4).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Mental health is constituted of several weakly correlated dimensions including psychological distress, life satisfaction, hedonia and eudaemonia.

    “It is constituted of several weakly correlated dimensions (Huppert & Whittington, 2003) including psychological distress, life satisfaction, hedonia (positive affect) and eudaemonia (the experience of purpose and meaning in life) (Steptoe, 2019).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Huppert & Whittington, 2003

      Not checked in the free run.

    • cites Steptoe, 2019

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Adolescence is commonly when emotional disorders and the gender gap in mental health emerges.

    “is commonly when emotional disorders and the gender gap in mental health emerges”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Wade et al., 2002

      Not checked in the free run.

    • cites WHO, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    PISA operates a two-stage sample design where schools are sampled with probability proportional to the size of their enrolment of 15-year olds, and students are sampled randomly with equal probability.

    “PISA operates a two-stage sample design where schools are sampled with probability proportional to the size of their enrolment of 15-year olds, and students are sampled randomly with equal probability.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites OECD, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Premisecitation

    Mental health is not simply the absence of mental illness but also a state of wellbeing and lies along a continuum from ill-health to positive mental health or wellbeing.

    “mental health is not simply the absence of mental illness but also a state of wellbeing and lies along a continuum from ill-health to positive mental health or wellbeing”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites WHO, 2018

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Zuckerman et al. suggest that a quadratic relationship exists between improving societal conditions (including gender equality) and the gender gap in subjective wellbeing.

    “Zuckerman et al. (Zuckerman et al., 2017) suggest that – in a sample of largely adults across 126 countries - a quadratic relationship exists between improving societal conditions (including gender equality) and the gender gap in subjective wellbeing.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Zuckerman et al., 2017

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    All questions were translated into the languages of participating countries by two independent linguists and then reconciled by a third to ensure consistent meaning in all countries.

    “All questions were translated into the languages of participating countries by two independent linguists and then reconciled by a third to ensure consistent meaning in all countries.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites OECD, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    A response rate of 80% of selected students in each school is required.

    “A response rate of 80% of selected students in each school is required.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites OECD, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    As additional and sensitivity analyses, we plot quadratic country-level associations to test for non-linear associations.

    “As additional and sensitivity analyses, we plot quadratic country-level associations to test for non-linear associations.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Income inequality is inconsistently associated with mental health, with some studies finding a correlation between higher income inequality and worse mental health whilst others find that higher income inequality correlates with better mental health.

    “However, income inequality is inconsistently associated with mental health, with some studies finding a correlation between higher income inequality and worse mental health (Oishi et al., 2011; Pickett & Wilkinson, 2010), whilst others find that higher income inequality correlates with better mental health (Rözer & Kraaykamp, 2013).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Oishi et al., 2011

      Not checked in the free run.

    • cites Pickett & Wilkinson, 2010

      Not checked in the free run.

    • cites Rözer & Kraaykamp, 2013

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing).

    Study design and conductown data

    We investigated if ecological findings were robust to adjustment for individual level controls - socioeconomic background, age and immigration status.

    “Thirdly, we investigated if ecological findings were robust to adjustment for individual level controls - socioeconomic background, age and immigration status.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    Mental health is a multidimensional concept.

    “yet it is a multidimensional concept”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Steptoe, 2019

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Students are weighted to yield a sample that is representative of the population of the country.

    “Students are then weighted to yield a sample that is representative of the population of the country.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites OECD, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Poverty is considered an established risk factor for worse mental health.

    “Poverty is considered an established risk factor for worse mental health”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Elgar et al., 2015

      Not checked in the free run.

    • cites Carol Graham & Chattopadhyay, 2013

      Not checked in the free run.

    • cites Lund et al., 2010

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    We use multi-level linear regression in order to estimate the between country variation in different mental health outcomes and to formally statistically investigate the associations between each of our four mental health outcomes, gender and country-level factors.

    “We use multi-level linear regression in order to estimate the between country variation in different mental health outcomes and to formally statistically investigate the associations between each of our four mental health outcomes, gender and country-level factors – GDP per capita, Gini and GGGI.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    A meta-analysis concludes that the relationship between income inequality and mental health is weak and dependent on a country's development.

    “A meta-analysis concludes that the relationship between income inequality is weak and dependent on a countries development (Ngamaba et al., 2018).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ngamaba et al., 2018

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Tesch-Romer et al. find that the association between gender equality and the adult mental health gender gap varies with the cultural attitudes of gender equality.

    “Tesch-Romer et al. (Tesch-Römer et al., 2008) find that the association between gender equality and the adult mental health gender gap varies with the cultural attitudes of gender equality.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Tesch-Römer et al., 2008

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    The gender equality measures reflect multiple dimensions of gender equality which influence experiences through all life stages and hence provide relevant information about the societal experiences for each gender.

    “they reflect multiple dimensions of gender equality which influence experiences through all live stages in these countries and hence provide relevant information about the societal experiences for each gender.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Despite different overall distributions, the mental health gender gap remains, although where the gap appears in the distribution differs by outcome.

    “Thus, despite different overall distributions, the mental health gender gap remains, although where the gap appears in the distribution differs by outcome.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Previous research indicates that stress and educational pressure is particularly correlated with worse mental health in adolescent girls.

    “Previous research indicates that stress and educational pressure is particularly correlated with worse mental health in adolescent girls (Schraedley et al., 1999; Wiklund et al., 2012).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Schraedley et al., 1999

      Not checked in the free run.

    • cites Wiklund et al., 2012

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    Confidence intervals overlapped particularly for life satisfaction and psychological distress, so caution is required in inferring deviation from linearity.

    “However, confidence intervals overlapped particularly for life satisfaction and psychological distress; thus, caution is required in inferring deviation from linearity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    In more gender equal countries, women have entered traditionally male dominated areas of employment, but men have not entered female dominated areas of employment to the same extent, nor do they do equal amounts of domestic work.

    “While in more gender equal countries women have entered traditionally male dominated areas of employment, men have not entered female dominated areas of employment to the same extent, nor do they do equal amounts of domestic work (England & Folbre, 2005; Garcia & Tomlinson, 2020).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites England & Folbre, 2005

      Not checked in the free run.

    • cites Garcia & Tomlinson, 2020

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisereasoning

    The GGGI captures greater detail of gender equality by including more and more diverse indicators, making it more granular, whilst also separating itself from a country's level of development.

    “By contrast, the GGGI captures a greater detail of gender equality by including more and more diverse indicators (Table S3), making it more granular, whilst also separating itself from a country’s level of development.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Adolescence and puberty mark a particular period of changing identity, including developing conceptions of what it means to be a man or a woman, and while there are cross-cultural differences in experience of adolescence, identity development is common.

    “Adolescence and puberty marks a particular period of changing identity (Blakemore & Mills, 2014) including developing conceptions of what it means to be a man or a woman (Greene & Patton, 2020), and while there are cross-cultural differences in experience of adolescence, identity development is common (Gibbons & Poelker, 2019).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Blakemore & Mills, 2014

      Not checked in the free run.

    • cites Greene & Patton, 2020

      Not checked in the free run.

    • cites Gibbons & Poelker, 2019

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The gender equality measures used are not specifically designed to capture exposures directly experienced by adolescents.

    “While the gender equality measures used are not specifically designed to capture exposures directly experienced by adolescents”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    The positive association between GDP per capita and the gender gap contrasts with other findings where a positive relationship between GDP and adolescent wellbeing has been found.

    “This contrasts with other findings where a positive relationship between GDP and adolescent wellbeing has been found (Torsheim et al., 2006)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Torsheim et al., 2006

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premiseown data

    This finding contrasts with previous findings that show an equivalent positive relationship between gender equality and life satisfaction in boys and girls.

    “This is in contrast to previous findings that show an equivalent positive relationship between gender equality and life satisfaction in boys and girls (Looze et al., 2018).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Looze et al., 2018

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    While more equal countries have larger gender gaps across all outcomes, the direction of association between Gini and mental health differs by outcome.

    “Thus, while more equal countries have larger gender gaps across all outcomes the direction of association between Gini and mental health differs by outcome.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    In countries with lower gender equality, women's roles are more fixed, whereas in more gender equal countries they are less prescribed, leading to potential conflict between roles, which may affect mental health.

    “In countries with lower gender equality women’s roles are more fixed, whereas in more gender equal countries they are less prescribed, leading to potential conflict between roles, which may affect mental health (Hopcroft & Bradley, 2007).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Hopcroft & Bradley, 2007

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisenone given

    The GGGI includes five indicators for economic participation, such as ratio of female earned income to male, and ratio of female professional and technical workers to males, compared to the GII's one measure of female and male labour force participation rates.

    “For example, the GGGI includes five indicators for economic participation, such as ratio of female earned income to male, and ratio of female professional and technical workers to males, compared to the GII’s one measure of female and male labour force participation rates.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Findings point to the complex nature of gender disparities in mental health and possible incongruence between expectations and reality in more gender equal countries.

    “Findings point to the complex nature of gender disparities in mental health and possible incongruence between expectations and reality in more gender equal countries.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    Cross-country comparisons are necessary to elucidate risk factors that operate at the population level, such as indicators of gender and income inequality.

    “However, cross-country comparisons are necessary to elucidate risk factors that operate at the population level (Pearce, 2000), such as indicators of gender and income inequality.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Pearce, 2000

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Overall, adolescence marks a period of emerging new stressors which may negatively affect girls' mental health to a greater degree than boys, and in more gender equal countries there may be more of these stressors.

    “Overall, adolescence marks a period of emerging new stressors which may negatively affect girl’s mental health to a greater degree than boys, and in more gender equal countries there may be more of these stressors.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Our measure of gender was binary in nature and does not allow investigation of non-binary gender identities and mental health.

    “Lastly, our measure of gender was binary in nature and does not allow investigation of non-binary gender identities and mental health.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The gender gap itself may partly be a product of reporting bias, with boys being less willing to report negative mental health than girls.

    “Thirdly, the gender gap itself may partly be a product of reporting bias – with boys being less willing to report negative mental health than girls.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    There could be systematic differences across genders in school attendance amongst the countries in our sample that could potentially bias comparison of gender gaps across countries.

    “Fourthly, there could be systematic differences across genders in school attendance amongst the countries in our sample that could potentially bias comparison of gender gaps across countries.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The study relies exclusively on cross-sectional cross-country correlations; thus, we cannot make any strong conclusions regarding the causal pathways involved.

    “Firstly, our study relies exclusively on cross-sectional cross-country correlations; thus, we cannot make any strong conclusions regarding the causal pathways involved.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Evidence suggests that individuals who violate gender stereotypes may receive backlash, which may have negative consequences for mental health.

    “Additionally, evidence suggests that individuals who violate gender stereotypes may receive backlash (Rudman et al., 2012), which may have negative consequences for mental health.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Rudman et al., 2012

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Cultural differences in Likert scale responses, such as positivity biases, may confound cross-country differences.

    “Secondly, whilst we cannot exclude cultural differences on likert scale responses, such as positivity biases, that may confound cross-country differences (Oishi, 2010)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Oishi, 2010

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Changing norms of female education and economic participation can increase educational stress and psychological distress for girls whilst they are still burdened with traditional anxieties related to maintaining a female identity and appearance.

    “Indeed, changing norms of female education and economic participation can increase educational stress and psychological distress for girls whilst they are still burdened with traditional anxieties related to maintaining a female identity and appearance (West & Sweeting, 2003)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites West & Sweeting, 2003

      Not checked in the free run.

72 of 72

  • The paper relies on this without support.

    It is unknown how wealth or income inequality are associated with the gender gap in mental health, and whether this differs by dimension of mental health.

    “It is unknown how wealth or income inequality are associated with the gender gap in mental health, and whether this differs by dimension of mental health — life satisfaction questions for example typically correlate more strongly with economic factors than affect-related questions (Graham et al., 2010).”

  • The paper relies on this without support.

    Few studies have explicitly examined the relationship between gender equality and the mental health gap in adolescents, investigated the adolescent gender gap in a broad sample of countries including low- and middle-income countries, and focused on multiple indicators of mental health.

    “Few studies, to our knowledge, have 1) explicitly examined the relationship between gender equality and the mental health gap in adolescents, 2) investigated the adolescent gender gap in a broad sample of countries including low- and middle-income countries and, 3) focused on multiple indicators of mental health.”
2 of 2

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Female life satisfaction drops earlier in adolescence than male life satisfaction.

    “Further, female life satisfaction drops earlier in adolescence25,26”

    From Windows of developmental sensitivity to social media (2022) · cites this paper as [25]

    The Results report that girls had lower life satisfaction than boys in most countries. But the Methods (Participants) describe a cross-sectional survey of students aged 15, and the paper does not establish that girls’ life satisfaction drops earlier in adolescence than boys’. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Discussion reverses the stated income-inequality relationship

    Major

    The results say more income-equal countries have larger gender gaps across all outcomes. The discussion instead says income inequality is associated with a wider gap in all cases.

  • Abstract overstates the GDP association with the gender gap

    Major

    The abstract says higher GDP is associated with a larger gender gap for every outcome, but the results explicitly exclude hedonia.

  • Two different female hedonia coefficients are given

    Minor

    The same sentence labels both 0.02 and −0.057 as the female hedonia coefficient. The latter appears to be intended for eudaemonia, but as printed the values conflict.

  • GDP coefficients have the wrong signs in the results text

    Minor

    The text reports positive GDP coefficients for life satisfaction and eudaemonia while saying that both outcomes decrease with GDP. Model C gives negative coefficients for both: −0.032 and −0.033, respectively.

4 of 4

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. D. Bann , E. Fitzsimons , W. Johnson , Determinants of the population health distribution, or why are risk factor-body mass index associations larger at the upper end of the BMI distribution? bioRxiv, 537829 (2019). · DOI
  2. K. Pickett , R. Wilkinson , The Spirit Level: Why Equality is Better for Everyone (Penguin UK, 2010).

Cited by

Sorted by citations

Related

Authors
Campbell, O., Bann, D., & Patalay, P.
Affiliations
University College London (GB)
Corresponding author
Campbell, Olympia
ORCID
Venue
SSM - Population Health
Publisher
openRxiv
Published
2020-06-13
Volume
13
Pages
100742-100742
ISSN
2352-8273 (ISSN-L)
DOI
10.1101/2020.06.12.20129312
PMC
PMC7960541
OpenAlex
W3125299664
Cited by
484
References
69
Citations by year
2026: 97 · 2025: 110 · 2024: 121 · 2023: 88 · 2022: 47 · 2021: 17 · 2020: 2 · 2014: 1
FWCI
67.37
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • SSM - Population Health (publishedVersion, cc-by)
  • medRxiv (acceptedVersion, cc-by-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
Funders
  • Academy of Medical Sciences (HOP001/1025)
  • Economic and Social Research Council
  • European Research Council (834597)
Topics
  • Social Sciences › Psychology › Social Psychology › Mental Health Treatment and Access
  • Social Sciences › Sociology and Political Science › Gender, Health, and Social Inequality
  • Social Sciences › Gender Studies › Gender Roles and Identity Studies
Keywords
adolescent mental health, psychological distress, life satisfaction, eudaimonia, hedonia, gender inequality, GDP per capita, Gini coefficient, cross-national comparison, gender equality indicators, country-level determinants

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review