Online friendships vary sharply across Nepal
Nepal ranks among South Asia’s most gender-mixed countries on Facebook, yet its national figure hides a striking divide between places. Men and women are more likely to appear in one another’s close friend networks in parts of the eastern hills and Karnali, while cross-gender ties are less common in Parsa and across Madhesh Province.
Contents
A study by researchers at Meta and New York University puts Nepal’s cross-gender friending ratio at 0.58. In practical terms, women make up a smaller share of men’s closest Facebook friends than women do of women’s closest friends: the share in men’s networks is about 60 percent of the corresponding share in women’s networks.
Among South Asian countries, only Bhutan scores higher. India and Pakistan have lower rates. Within Nepal, Bhojpur has the highest district-level ratio, above even Bhutan’s national figure, while Parsa is near India’s national level. Karnali leads Nepal’s provinces, and Madhesh records the lowest ratio.
The figures come from an unusually large analysis of Facebook connections: about 1.38 trillion ties among 1.8 billion users worldwide. For the Nepal comparison, the analysis focuses on the 200 closest friends of active Facebook users aged 18 to 65. Those numbers provide a detailed view of online networks, though they do not amount to a direct count of people’s friendships offline.
What does the ratio measure?
The cross-gender friending ratio compares the share of women among men’s close friends with the share of women among women’s close friends. A score of 1 would mean that women account for an equal share in both groups’ networks. Nepal’s 0.58 score indicates a substantial gap, not that 58 percent of all Nepali friendships are between men and women.
The measure also helps account for differences in the number of women and men in users’ networks. Researchers rank ties by strength, using Facebook’s measure of connection strength, and examine close connections rather than treating every added contact as an equally meaningful friendship.
Gender homophily, the tendency for people to form ties with others of their own gender, appears in almost every country studied. The pattern is often stronger among less close connections than among the strongest ties. Family relationships and other close bonds may cross gender lines even where social circles farther from home remain divided.
National averages conceal local differences
Nepal’s range, from Bhojpur’s comparatively high rate to Parsa’s lower one, shows why a single national score cannot describe how people interact across the country. The same point appears at the provincial level: Karnali has the highest rate, while Madhesh has the lowest. The available figures establish these rankings but do not, by themselves, explain the causes of each local pattern.
Geography, access to social networks, age composition, employment, education, local customs and the way people use Facebook could all affect what appears in the data. The research does not prove that any one factor explains why a district scores higher or lower. Interpreting local results requires care, especially when differences in internet access and platform use may shape who is represented.
One clue comes from the broader study’s international findings. Across countries, cross-gender friendship rates track established measures of gender disparity, including differences in labour-force participation. Within countries, researchers also find an association between more mixed-gender friendship networks and greater support for gender equality. These are correlations: they show that patterns tend to occur together, not that one necessarily causes the other.
Why friendship networks matter beyond Facebook
Friendship ties can offer a window into who has routine contact across gender lines. Such contact may reflect whether women and men study, work, travel and participate in public life together. Where women have fewer opportunities to enter workplaces and other shared settings, their chances to build friendships with men outside family networks may also be more limited.
The relationship can run in more than one direction. Regular contact can make people more familiar with one another’s experiences, while social rules and economic conditions can shape who meets in the first place. The study’s connection between friending patterns and gender equality is therefore useful as a broad indicator, but it cannot establish whether mixed friendships change attitudes, whether more equal settings create mixed networks, or whether both reflect other conditions.
Researchers’ global results add context. Cross-gender friendships are comparatively uncommon across much of South Asia and parts of the Middle East and North Africa, while several areas in Latin America, sub-Saharan Africa and Southeast Asia show higher rates. The pattern is not uniform within any broad region, and national or regional labels should not be mistaken for descriptions of every community.
Facebook data has limits
The scale of the dataset is a major strength, but the platform does not represent every person or every relationship. The analysis covers active Facebook users aged 18 to 65, and it measures online connections rather than asking users directly about closeness, trust or time spent together. A Facebook friend may be a relative, colleague or acquaintance, while some close offline friends may not be connected on the platform at all.
Age also matters. Facebook use is less common among many younger people than it was in earlier years, so the networks observed on the platform may not reflect how younger Nepalis form friendships. Differences in device ownership, internet access and social expectations can shape who uses Facebook and whose connections appear in the dataset.
Digital access may be particularly important when studying gender. If women are less likely than men to have reliable access to mobile internet or Facebook, the women who do appear may not represent women in the wider population. Research on social media friendships should therefore be read alongside surveys and evidence on internet access, work and everyday social life.
Patterns are not fixed
The global analysis describes an association between cross-gender ties and measures of gender equality, not an unchanging cultural trait. Social networks can shift as people’s opportunities and habits change. Education, urbanisation, migration, employment and new communication technologies may influence who meets and stays in contact, though the Facebook figures alone cannot measure the effect of each factor.
That distinction matters when comparing Nepal’s districts. A higher ratio in Bhojpur or Karnali does not establish that residents there hold more egalitarian views than people elsewhere. Nor does a lower score in Parsa or Madhesh explain the attitudes of individuals. The metric summarises group-level networks, and it should not be used to stereotype communities or infer a person’s beliefs from where they live.
Researchers and policymakers can use these patterns to identify questions for further study. Local research could examine whether employment, school and college settings, migration, household expectations or internet access help account for differences across districts. Comparing Facebook data with in-person surveys would also show how closely online friendship networks match relationships beyond the platform.
What to Know
- Nepal’s cross-gender Facebook friending ratio is 0.58, meaning women’s share of men’s close networks is about 60 percent of women’s share in women’s networks.
- Bhutan is the only South Asian country with a higher ratio; India and Pakistan are lower.
- Bhojpur has Nepal’s highest district-level ratio, while Parsa is near India’s national figure.
- Karnali leads Nepal’s provinces, and Madhesh has the lowest rate.
- The study analyses close Facebook ties among active users aged 18 to 65, not all friendships in the population.
- Cross-gender friendship rates are associated with measures of gender equality, but the data do not establish cause and effect.
