Where Men & Women Are Most & Least Likely To be Friends

All the maps featured here come from the article: Cross-Gender Social Ties Around the World By Michael Bailey, Drew Johnston, Theresa Kuchler, Ayush Kumar, Johannes Stroebel and its annex.

Here’s a summary of the data used:

Here, we introduce, analyze, and publicly release a new global dataset on cross-gender friendship links at the sub-national level for nearly 200 countries and territories. Our measures are based on more than 1.38 trillion social ties observed between over 1.8 billion users on Facebook, a global online social networking service.

The aggregated data is available for download at the HDX (https://data.humdata.org/dataset/cross-gender-ties).

Summary of the article itself

The paper “Cross-Gender Social Ties Around the World” used a large-scale, global dataset on how men and women form social ties, using data from over 1.8 billion active Facebook users aged 18 to 65, capturing more than 1.38 trillion friendship links.

Cross-Gender Friending Ratio (CGFR): This core measure compares how frequently men befriend women compared to how frequently women befriend women. Most locations have CGFRs below 1, meaning men, on average, have fewer female friends relative to women’s share of female friends. There’s wide global and within-country variation (for example, in India versus Kenya or north versus south Italy). :
Role of Tie Strength: People’s closest friendships are somewhat less segregated by gender. Among “top 25” friendships—often family members—there is more cross-gender interaction than among looser, weaker ties.
Regional Differences: Sub-nationally, East Germany shows higher cross-gender connectedness than West Germany; in the U.S., differences are less pronounced overall but are higher in more diverse counties. In India, many regions exhibit heavily segregated networks except in some northeastern states.
Correlation with Gender-Related Indicators: Globally, higher CGFR values align with narrower gender gaps in labor force participation. Survey data (e.g., World Values Survey) also shows that localities with higher CGFR have more egalitarian beliefs on women’s education and political rights.
U.S. County-Level Findings: In the U.S., higher CGFR tends to appear in counties with lower shares of White residents, lower religious adherence, higher female labor force participation, and higher educational attainment. Counties with a higher percentage of White residents or more religious congregations tend to show more gender-segregated social networks.