Issues and challenges to education policies and practices in Nepal: A Descriptive and Factor Analysis

1. Introduction

Nepal’s education sector has undergone significant structural and policy reforms over the past two decades to promote quality education. The implementation of federalism in 2015 distributed the responsibility for education among the federal, provincial, and local levels. Similarly, the School Sector Development Plan (SSDP, 2016–2023) aimed to provide quality education to all children. The Digital Education Policy (2019) has encouraged technology-friendly teaching, while the National Curriculum Review (2076) envisioned a practical, skill-based, and student-centred curriculum (Ministry of Education, 2016).

However, despite these efforts, many issues and challenges, such as academic quality, curriculum relevance, teacher-student ratio, digital divide, and, most seriously, student exodus, have long plagued Nepal’s educational system.

In the context of Nepal, education policy evaluations have often focused on physical and financial indicators such as infrastructure (school buildings, water, and toilets), teacher-teacher ratio (student-teacher ratio), textbook distribution, and budget mobilization. Policymakers and donor agencies conclude that education is improving when they see progress in these indicators. However, policy analysis does not adequately consider the real perceptions, needs, and problems of students, who are the primary beneficiaries of education (Ministry of Education, Science and Technology, & Joint Financing Partners, 2022). The failure to incorporate students’ needs analysis generated a divide between policy and practice.

The significance of this study lies in the following points: It helps to identify policy weaknesses based on direct feedback from students. To date, many policies have restricted themselves to meetings involving senior officials and experts. School management receives suggestions from students themselves to enhance teacher-parent communication, utilize digital tools, and implement the curriculum. For educational research, Nepal has limited this type of PCA analysis based on the perceptions of a large number of students (Jolliffe, & Cadima, 2016). This study encourages a student-centered approach in educational research. For international comparison: Since other developing countries also face challenges similar to Nepal, the findings of this study can be useful for comparing South Asian education policies.

No education policy is complete and successful without incorporating the students’ voices. Students themselves determine the final assessment of the quality of education. Does the curriculum match their learning level? Has a digital tool really made learning easier? Is communication between teachers, school management, and local government effective? The answers to these questions can only come from the students. Against this backdrop, this study has attempted to uncover the reality of Nepal’s education policies and practices from the perspective of students.

This study seeks to answer the main research question: What are the major problems and challenges of education policies and practices from the perspective of secondary school students in Nepal? The study also includes the following sub-questions, expanding on the main question: How do students perceive the teaching-learning environment? What is the gap between the availability and use of digital tools? What is the communication between teachers, parents, school management, and local government? How has educational quality affected student dropouts?

This study is grounded in Expectancy-Value Theory (Eccles & Wigfield, 2002), which posits that students’ educational decisions and persistence are shaped by two core beliefs: (a) expectations of success (i.e., “Can I succeed?”) and (b) the value of individual work (that is “Is this worth my effort?”). In the context of secondary education in Nepal, expectations of success manifest in students’ confidence in learning through English, their perceptions of the curriculum’s relevance to future careers, and their satisfaction with overall educational quality. The value of individual work is demonstrated by students’ motivation to remain in their local schools, their attraction to foreign universities, and their inclination to migrate to urban schools. The negative loading of in our PCA results aligns with this theory—students who place a high value on local schools exhibit a reduced tendency to migrate. Ultimately, expectancy-value theory provides an appropriate framework to elucidate the three components we identified: systemic communication influences perceptions of support for success, digital tools affect task engagement, and perceptions of quality inform expectancy-value assessments—factors that ultimately determine whether students choose to stay or leave.

This study seeks to identify the objectives of analyzing students’ perceptions descriptively based on 15 indicators (learning environment, digital tools, and communication system). 2. To identify the underlying dimensions (factors) of educational challenges through Principal Component Analysis. 3. To formulate concrete and implementable policy recommendations for the federal, provincial, and local levels based on the identified factors.

2. Literature Review

3. Methods

4. Results

The study is presented in this section in a statistical format. This section includes two primary descriptive statistics and a PCA test results.

Table 1 presents the weighted mean and standard deviation of the 15 questions. Based on the mean, students agree most with regular exams (Q10, mean 4.31) and the opportunity to ask questions in class (Q9, mean 4.11). Students also consider digital tools intriguing (Q6, mean 4.09). Similarly, scholarships are perceived to help retain students in rural areas (Q14, mean 4.07), and English medium is also viewed positively (Q1, mean 4.03).

However, students express dissatisfaction with the overall quality of education in Nepal, as indicated by a mean score of 2.97, the lowest among all factors. Complaints that the curriculum does not match their learning level and career are also strong (Q2, mean 3.03). Students also seem to be neutral about support from local government (Q13, mean 3.15), communication between teachers and parents (Q11, mean 3.19), and communication between school management and parents (Q12, mean 3.24). These areas need further improvement.

In terms of standard deviation, local government support (Q13, SD = 1.19) shows the most polarization—that is, support is adequate in some schools and inadequate in others. Similarly, there is high polarization in overall quality of education (Q4, SD = 1.18) and school communication (Q12, SD = 1.18). The 0.86 difference between digital tools (Q6, 4.09) and teachers using them (Q7, 3.23) shows a gap between policy and practice.

Table 2 checks whether the data is suitable for principal component analysis (PCA). The first test is the KMO (Kaiser-Meyer-Olkin), which evaluates to 0.82. Generally, a value above 0.70 is considered ‘acceptable.’ A value of 0.82 confirms that the data is very suitable for PCA. This means that there is sufficient correlation between the 15 questions to identify common factors.

The second test is Bartlett’s test of sphericity. Its value is χ² = 2450.6 and p < 0.001. A p-value much less than 0.05 (highly significant) means that the variables (questions) are related to each other and there is structure between them. If the p-value were greater than 0.05, PCA would be inappropriate. Both tests confirm that the data is fully prepared for PCA analysis and that the components extracted is meaningful.

Table 3 shows the eigenvalue of each component and the percentage of variance it explains. The eigenvalue measures the strength of a component—only components with an eigenvalue greater than 1 are considered significant. The first component (PC1) has an eigenvalue of 5.12, which is the largest. It explains 34.1% of the total variance. The second component (PC2) has an eigenvalue of 2.45, which explains 16.3% of the variance. The third component (PC3) has an eigenvalue of 1.68, which explains 11.2% of the variance.

Together, these three components explain 61.6% of the total variance. This is a very positive result, because in social science research, explanatory power above 60% is considered satisfactory. PC4 has an eigenvalue of 1.02 and explains 6.8% of the variance, so it was removed. The remaining 11 components explain 31.6% of the variance, which is relatively low and can be considered ‘noise.’

Table 4 is the most important table of the PCA. It shows how the 15 questions are grouped into three components. Loadings ≥0.45 are considered significant.

The first component (PC1)—systemic communication and support: In this area, teacher-parent communication (Q11, loading 0.82) is the strongest. This is followed by principal-management-student communication (Q12, 0.79) and local government support (Q13, 0.74). Teacher feedback (Q8, 0.58) and opportunity to ask questions (Q9, 0.45) also fall into this component. This means schools where all this communication and support are beneficial have high PC1 values.

The second component (PC2)—Digital teaching and innovation: In this area, digital tools make lessons engaging (Q6, loading 0.87), and teachers use digital tools (Q7, 0.84) that are very strongly linked. The loadings for other questions are weak here. This means PC2 is mainly defined by the availability, interest, and use of digital devices. The gap between Q6 and Q7 (4.09 vs. 3.23) is the weakness of this component.

The third component (PC3)—Quality and Outmigration: It has high loadings on quality dissatisfaction (Q4, loading 0.78), city migration (Q15, 0.72), and attraction of foreign universities (Q5, 0.65). Most importantly, the loading on motivation to stay in local schools (Q3) is negative (−0.48). This means students who want to stay in local schools are less likely to out migrate. This is a crucial policy signal.

Table 5 summarizes all three components. The first component (PC1), “Systemic communication and support,” explains 34.1% of the total variance. The questions that define it are teacher feedback, the opportunity to ask questions, teacher-parent communication, school management communication, and local government support. This simply means how effective the overall communication system of the school is.

The second component (PC2), “Digital teaching and innovation,” explains 16.3% of the variance. It is defined by only two questions—the attractiveness of digital tools and the use of these tools by teachers. This means these two questions are enough to measure the state of digital teaching.

The third component (PC3), “Quality and student outflow,” explains 11.2% of the variance. It is defined by four questions—quality dissatisfaction, attraction of foreign universities, migration to cities, and local sustainability (negatively). This component confirms that low quality leads to an increase in student outflow.

Table 6 shows the relationship between the three components. The most important finding is the negative correlation between PC1 and PC3 (−0.35). The negative correlation means that when PC1 (systemic communication) increases, PC3 (emigration) decreases. In other words, schools with effective communication systems have less student emigration. This is the most important message for policymakers.

There is a weak positive correlation of 0.28 between PC1 and PC2, which means that if communication is excellent, digital is also likely to be beneficial to some extent. There is a weak negative correlation of −0.15 between PC2 and PC3, which means that there is no strong link between digital and emigration. This indicates that the use of digital tools does not directly reduce emigration, but communication does.

The strongest effect of PC1 is seen on emigration (PC3). The lack of strong correlation between the three components confirms that they are independent dimensions.

Table 7 presents the key differences obtained from the descriptive and PCA analyses. Digital Usage Gap (DUG): Students believe that digital tools are intriguing (Q6, 4.09), but teachers do not use them much (Q7, 3.23). The gap between these two is 0.86, which is large. This indicates that the policy is not implemented as planned.

Quality Satisfaction Gap (QSG): Students are highly positive about regular exams (Q10, 4.31) but dissatisfied with the overall quality of education (Q4, 2.97). The gap between these two is 1.34, which is the largest. This means students recognize the exams but do not trust the overall system.

Outmigration Pressure Gap (OPG): The mean of students who are motivated to stay at their local school is 3.64, which is moderately positive. However, the mean number of people who want to go to school in urban areas is 3.73, which is slightly higher. Outmigration pressure is slightly higher than school sustainability.

Communication Gap (CG): The mean of all questions (Q11, Q12, Q13) falling under PC1 is between 3.15 and 3.24, which is below 3.5. This score is considered ‘neutral’, but to be positive, it needs to be at least above 3.5. Therefore, this communication is in a weak state. Finally, these four gaps illustrate the gap between policy and practice in Nepal’s education system.

5. Discussion

The main findings of this study is—three major components (PC1: Systemic Communication and Support, PC2: Digital Teaching and Innovation, and PC3: Quality and Student Attrition)—have revealed a deep gap between Nepal’s education policy and practice. Combining these findings with previous theoretical and empirical studies reveals some important similarities and some fundamental differences. We discuss each component in detail below.

PC1 explains the largest part of the total variance (34.1%). It has high loadings on questions such as teacher-parent communication (Q11, 0.82), school management transparency (Q12, 0.79), and local government support (Q13, 0.74). However, the descriptive statistics indicate that the mean score for all these factors is between 3.15 and 3.24, which reflects a ‘neutral’ level of satisfaction. Therefore, students are not satisfied with the communication and support system.

This finding is fully consistent with Cohen, Moffitt, and Goldin (2007)‘s “policy-practice dilemma.” According to them, no matter how well-designed a policymaker’s plan is, it can be thwarted by school culture, teacher trust, and resource constraints at the implementation level. After federalization in Nepal, local governments have been given responsibility for education. But the median of Q13, 3.15, shows that local governments have not been able to perform as expected. This also ties in with Verger and Moschetti’s (2017) argument that the same policy can produce different results in different contexts. Some local levels may have done well (as shown by the high SD of Q13, 1.19), but the average is poor. In the Nepali context, Sahani (2025) has presented a rights-based approach by calling it “bridging the gap.” But PC1 of this study shows that before bridging the gap, one must understand how big it is. Communication between students and parents is not regular (Q11, 3.19), there is no transparency between principals and students (Q12, 3.24), and the presence of local government is not felt (Q13, 3.15). Therefore, as Ghimire and Joshi (2025) said, despite the governance structure, accountability is weak.

Heimans (2012) analyzed the relationship between policy, practice, and power and said, “In the policy-making process, the voice of those who have power dominates, and the voice of students in rural areas is not heard.” The low mean value of PC1 in this study confirms the voice of students has not reached the center of policymaking. Dhakal (2025) looked at the challenges of principals but did not look at how students perceive the principal. This study’s Q12 addresses how students perceive the principal.

PC2 explains 16.3% of the total variance. Its most significant feature is that it is defined by only two questions: Q6 (digital tools are intriguing, 0.87) and Q7 (teachers use them, 0.84). But the gap between the two is worrying: the mean of Q6 is 4.09 (high agreement), but the mean of Q7 is 3.23 (neutral). This gap of 0.86 demonstrates Cohen’s policy-practice dilemma in practice. This gap is not unique to Nepal — it is a phenomenon seen worldwide. Burdett and O’Donnell (2016) used the expression “lost in translation” to explain how a successful policy in one country fails when implemented in another. Nepal borrowed digital policies from Western countries. Smart boards and projectors came. But teachers were not trained to use them. Technical support teams were not hired. The quality of the internet was not ensured. Consequently, the costly equipment was a futile investment. This argument also ties in with Rizvi’s (2009) challenge of global mobility—borrowing policy without considering the local context is a failure.

In the Nepali context, Joshi (2017) said 8 years ago, “The policy is there, but the implementation is limited”. However, PC2 of this study indicates that the same gap continues to exist in 2025. It indicates that the policy, implementation, and students’ fate are unchanged. This finding is a serious warning for policymakers. Kaatrakoski et al. (2017) analyzed the same paradox seen in open educational practice—even though the technology is available, the implementation is limited by the lack of digital literacy of teachers, institutional support, and incentives.

Murphy (2013) also concluded the same—adopting open educational practices in higher education requires a change in culture, leadership commitment, and long-term investment. Nepali community schools do not exhibit any of these characteristics. Therefore, despite the high loadings of PC2, the descriptive data show a gap. PC2 of this study not only confirms Joshi’s (2017) findings but also adds student voice to them. Now, policymakers know students are ready for technology, but teachers are not.

PC3 explains 11.2% of the total variance. But its significance is not measured by the amount of variance—it indicates a long-term threat. PC3 consists of Q4 (Quality Dissatisfaction, 0.78), Q15 (City Outmigration, 0.72), Q5 (Foreign University, 0.65), and Q3 (Local Sustainability, −0.48). Descriptive statistics show Q4 means 2.97, which is the lowest of all 15 questions. This is consistent with the argument of Adams, Acedo, and Popa (2012)—that the definition of quality education should be multidimensional. But Nepalese students do not show satisfaction in a single dimension. This finding is in line with Shields and Rappleye’s (2008) study of the rural-urban gap. Therefore, the gap between policy and practice demotivates learners to pursue education.

6. Conclusion

This study attempted to identify the problems and challenges in Nepal’s secondary education system from the perspective of students. After conducting descriptive and principal component analyses of the responses of 390 students from 73 schools, three main challenges: lack of systematic communication, underutilization of digital tools, and quality dissatisfaction emerged. These challenges are interrelated, but each has its own specificity and solution path.

The first challenge—the weakness of systemic communication—is the most powerful finding in this study. PC1 explains 34.1% of the total variance and includes teacher-parent, school management, and local government communication, indicating that this aspect is highly significant. Unfortunately, the mean of all of these is below 3.25—meaning that students clearly feel a lack of communication. Policymakers often focus on infrastructure and teacher numbers, but there is a significant gap in listening to student voices, informing parents, and holding local governments accountable.

The second challenge—the underutilization of digital tools—shows the largest policy-practice gap. Students are enthusiastic about digital tools, but they do not agree with teachers using them. This gap of 0.86 indicates that digital policies have been made, budgets have been allocated, and equipment has been purchased, but due to a lack of teacher training, technical support, internet quality, and regular monitoring, those tools are gathering dust. The issue is not a problem of technology but a problem of the lack of use of technology.

The third challenge—quality dissatisfaction and student exodus—is the most serious long-term threat. Student dissatisfaction with the overall quality of education in Nepal is the lowest on average. This means students have lost faith in the system they are studying in. The result of this dissatisfaction is their tendency to move to the city or abroad. The result is a series of brain drains — from the village to the city, from the city to Kathmandu, from Kathmandu to abroad.

However, the negative loading of on PC3 presents an opportunity. Motivated students who wish to remain in local schools exhibit a reduced likelihood of dropping out. The direct message is clear: improving the quality of local schools keeps students there and prevents dropouts. Moreover, the negative correlation between PC1 and PC3 provides further evidence that dropouts are lower in schools with excellent communication. In other words, strengthening communication can reduce dropouts instead of waiting for quality improvements.

Finally, the biggest contribution from this study is that the students’ voices should be at the center of policy evaluations. Infrastructure, teacher numbers, and budget—all these are important, but the most important indicator is student perception. They are expressing dissatisfaction with the quality. They are experiencing a lack of communication. They are complaining about the lack of use of digital tools in classroom teaching and learning. And most worryingly, they are losing faith in their country’s education and preparing to move to cities and abroad. If we do not initiate improvements today, we are likely to confront an even larger brain drain tomorrow.

7. Policy Implications

The following policy actions are urgently needed to address the three main findings of this study—weak systemic communication (PC1), underuse of digital tools (PC2), and dissatisfaction with quality (PC3).

Ethics Statement

Ethics Approval and Consent to Participate

Ethical approval for this study was obtained from the Research Ethics Committee (REC) of the Research Management Cell (RMC) of Tribhuvan University, Saraswati Multiple Campus, Nepal (968/082/083). The research adhered to the ethical principles of the Ministry of Education (2017). Before participation, written informed consent was obtained from the parents or legal guardians of all children.

Consent for Publication

All the parents or legal guardians of all children provided written informed consent for publication of anonymized data and findings.