Gender and fintech adoption

What I found

• A large gender gap. Men average 4.18 on the 5-point intent scale and women 3.59 (t = 4.27, p < 0.01). Even with controls, being female roughly halves the odds of a higher intent score (odds ratio 0.48 to 0.52, p < 0.01).
• Five drivers across the whole sample. In the fullest model, technology experience (odds ratio 1.39), awareness (2.21), current fintech use (2.03), security (1.65), and convenience (1.91) all raise the odds of higher intent (p < 0.05). Older respondents are less likely (0.63, p < 0.1), and private-sector workers more likely (2.59, p < 0.05).
• Several expected drivers did not matter. Privacy, ease of use, social media influence, reliability, income, family influence, and location were not significant once the others were in the model.
• Men and women follow different paths. For men, privacy (1.77), ease of use (1.65), education (2.03), and income (1.83) matter, while social media influence points the other way (0.70). For women, convenience is the strongest driver (2.68), alongside technology experience (1.85), awareness (2.08), security (1.69), and reliability (1.59).
• Adoption is not simply higher with income. 84% of middle-income respondents (PKR 50,001 to 200,000) already use fintech, against 69.7% above PKR 1 million and 62.8% below PKR 50,000. Education matters more: the share in the top two intent categories rises from 58.2% (high school) to 70% (bachelor’s) and 83.1% (master’s or higher).

How I built the models

• Model 1: attitudes and experience (technology, awareness, current use, security, reliability, losses, privacy, ease, convenience, social media).
• Model 2: adds age, gender, occupation, and education.
• Model 3: adds income and family influence.
• Model 4: adds location.
• I compared nested models with likelihood-ratio tests and AIC. For everyone, adding personal characteristics improved the fit (LR χ²(8) = 24.99, p = 0.0016), while family and location did not (p = 0.19 and 0.18).