Walking to School Linked to Lower Grades in Nepal, Study Finds

Every morning in the lakeside city of Pokhara, thousands of children set out for school on foot, weaving through narrow lanes, crossing busy roads and climbing over uneven terrain before the first bell rings. Others arrive by bus, motorcycle or private car, dropped at the gate rested and ready. For years, public health campaigns in wealthy countries have celebrated the walking child as the ideal one, praising active commuting for its benefits to fitness, mood and even cognition. But a new study from Nepal turns that familiar narrative on its head. Researchers at Tribhuvan University report that among eighth-grade students in Pokhara’s community schools, those who walked or cycled to school scored significantly lower on academic tests than those who traveled by motorized transport, and the gap persisted even after accounting for family wealth, parental education and a host of other factors.

The research, published in the journal Discover Education, drew on data from 918 students across 12 randomly selected community schools in Pokhara Metropolitan City. The team, led by Sudesh Pandit with colleagues including Ramesh Adhikari, Om Chandra Thasineku and Sushil Sharma, chose grade VIII deliberately: in Nepal’s school system, this is the year students sit a standardized examination administered by the Metropolitan Department of Education, marking their transition from basic to secondary education. Educational achievement was measured with a self-administered test built from multiple-choice items drawn from the National Assessment of Student Achievement and model questions from the Curriculum Development Centre. The test covered four core subjects, English, Mathematics, Science and Social Studies, with 20 items per subject. Items were reviewed by subject experts and pre-tested on 48 students from schools outside the final sample to check clarity, difficulty and content validity before the main study began.

The headline numbers are striking. Roughly three quarters of the students, 75.1 percent, were active commuters, meaning they walked or cycled to school, while the remaining 24.9 percent used motorized transport such as school buses, public vehicles or private cars. Passive commuters averaged 13.54 points on the composite achievement score, with a standard deviation of 3.19, while active commuters averaged 11.49 points, with a standard deviation of 2.62. That difference of just over two points on a 20-point scale represents a meaningful academic gap. By contrast, differences tied to commuting time and distance were small. Students who traveled more than 20 minutes scored only marginally higher than those with shorter trips, and distance from home to school showed essentially no relationship with achievement. What mattered, the data suggested, was not how far or how long students traveled, but how they traveled.

To test whether that pattern would survive statistical scrutiny, the researchers built a series of multiple linear regression models, adding blocks of control variables step by step. The first model adjusted only for commuting time and distance; the second added demographic characteristics such as age, sex and ethnicity; the third introduced socio-economic factors including parental education, parental occupation, household income and housing ownership; and the final model incorporated measures of the supportive environment, such as parental support at home, parental engagement in school activities and teacher support, along with indicators of student proactiveness like homework regularity, extracurricular participation, absenteeism and peer relationships. Active commuting remained significantly and negatively associated with educational achievement in every model. The standardized coefficient started at minus 0.319 in the first model, fell to minus 0.268 after demographic controls, minus 0.244 after socio-economic controls, and settled at minus 0.121 in the fully adjusted model, still statistically significant at p less than 0.001.

That shrinking but stubborn coefficient tells a nuanced story. Each layer of controls absorbed part of the raw gap, confirming that active commuters in Pokhara tend to come from families with fewer educational and economic resources. Maternal education, in particular, emerged as a strong socio-economic predictor of achievement, consistent with a large international literature linking mothers’ schooling to children’s academic outcomes. Yet even after fifteen control variables were accounted for, commuting mode retained an independent association with test scores. The authors checked the assumptions of their regression models carefully, examining residual plots for normality and homoscedasticity, confirming low autocorrelation with a Durbin-Watson statistic of 1.642, and verifying that multicollinearity was not a problem using variance inflation factors. Some variables, including father’s education and certain occupation categories, had to be dropped from the final model because of perfect multicollinearity.

Why would walking to school, so often promoted as a cognitive booster, be linked to poorer grades in this setting? The authors point to the starkly different meaning of active commuting in Nepal compared with high-income countries. In cities like Pokhara, walking or cycling to school is rarely a lifestyle choice; it is frequently an involuntary consequence of financial constraints, with narrow lanes, unsafe roads and difficult terrain making the daily journey a burden rather than a pleasure. Long, physically demanding trips can produce fatigue that erodes the energy available for homework and classroom concentration. Carrying heavy school bags over long distances adds a further psychological and physical toll, and prior research cited by the team suggests students forced into such commutes may feel less motivated toward their studies. Passive commuters, by contrast, arrive feeling safe, comfortable and rested, and more than 81 percent of them commute for more than 20 minutes each day, a duration that in the fully adjusted model showed a weakly positive association with achievement for this group alone.

The study’s demographic findings add further texture. The sample was 55.9 percent female and 44.1 percent male, with the largest ethnic group being Janajati at 38 percent, followed by Brahmin or Chhetri at 31.8 percent and Dalit at 27.6 percent. Sex showed no significant association with achievement, echoing international evidence that the gender gap in education is closing. Age and ethnicity did matter, with older students and certain ethnic groups scoring lower, but neither explained away the commuting effect. Household economics painted a familiar picture of urban Nepal: more than half of families earned between 21,000 and 40,000 Nepalese rupees per month, over half lived in rental housing, and a large share of fathers worked in foreign employment or daily wage labor while most mothers reported household work as their primary occupation. Parental support at home was reported by 59.7 percent of students, teacher support by 61.5 percent, and 78.6 percent participated in extracurricular activities, though the latter showed no significant link to achievement.

The findings carry real weight for policy. Nepal’s government has pursued a policy of ensuring schools lie within a 20-minute walking distance of children’s homes, and recent measures such as customs duty exemptions for school buses have expanded motorized options. Yet the School Sector Development Plan and the constitutional guarantee of free and compulsory basic education, the authors note, pay little attention to the environmental and geographical barriers that shape how children actually get to school. The study does not argue that walking is inherently harmful; rather, it argues that when active commuting is unsafe, exhausting and imposed by poverty, its cognitive benefits are unlikely to outweigh its structural and psychosocial burdens. The authors call for safe, comfortable, convenient and efficient commuting arrangements, suggesting that well-designed transport systems and safer routes could transform the daily journey from a drain on learning into a neutral or even positive experience.

As with any cross-sectional study, caution is warranted. The design captures associations at a single point in time and cannot establish that commuting mode causes differences in achievement; unmeasured factors such as neighborhood safety or school quality could still play a role. The authors themselves emphasize that the results should be interpreted as associations, not causal effects, and they call for longitudinal and mixed-method research to unpack why active commuting appears to work against academic success in this context despite its well-documented health benefits. Still, the study is a vivid reminder that the same behavior can mean entirely different things in different places. In Copenhagen or Portland, a child walking to school is exercising a healthy choice. In Pokhara, that same walk may signal a family’s lack of options, and the weight of that daily journey, the researchers suggest, follows the child into the classroom.

Subject of Research: The association between active and passive commuting modes to school and educational achievement among students in Pokhara, Nepal

Article Title: Association between commuting modes to school and educational achievement among students in Pokhara, Nepal

Article References: Pandit, S., Adhikari, R., Thasineku, O. C., & Sharma, S. (2026). Association between commuting modes to school and educational achievement among students in Pokhara, Nepal. Discover Education, 5(1), Article 1108. https://doi.org/10.1007/s44217-026-02252-9

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02252-9

Keywords: active commuting, educational achievement, Nepal, Pokhara, school transport, students, public health, socio-economic status, regression analysis, education policy, commuting time, Discover Education