Emerging inequality in Kenyan secondary schools: Dilemmas of educational expansion and quality improvement

This article examines the emerging unfair inequality in Kenyan secondary schools through comparative case studies of three secondary schools in western Kenya. Qualitative data were collected through fieldwork over a four-year period, with participant observation and semi- or non-structured interviews, to understand how interactions among schools, households, and communities impact the improvement of educational quality. This study demonstrates that educational inequality stems from economic background and academic performance. While establishing new schools allowed students to choose better schools in their vicinity, increasing school competition resulted in a school hierarchy restricting uniform access due to factors of affordability and academic achievement. This suggests that the unplanned establishment of new schools constrains vulnerable students from continuing their education. Expanding educational opportunities and improving quality are important facets of education; however, it is necessary to pay attention to the beneficiaries of this process, as economic inequality may translate into educational inequality.

relationships with the community, including parents/guardians (DeJaeghere et al., 2009;Eacott & Asuga, 2014;Hoadley et al., 2009;Lumadi, 2017). While these studies emphasize the importance of each of these actors in improving educational quality, they often do not look at the issue of how quality improvement reduces educational inequalities. Failure to comprehensively discuss how educational quality can be improved and what its goals are may lead only to new inequalities.
The issue of inequality in secondary education is more complex than in primary education because student enrollment in secondary school is often based on academic performance; students must satisfy stipulated academic standards to enroll in highly ranked schools. Therefore, an exploration of secondary education inequality requires a more complex analysis than an examination of primary education.
Secondary education inequality should thus be discussed in relation to equity. Equity does not merely seek to eliminate inequality among groups but is also concerned with the extent to which such gaps or other outcome distribution elements are 'fair' (Rolleston & Iyer, 2019). This study, therefore, focuses on 'unfair' inequality in analyzing secondary education in Kenya, which has expanded educational opportunities faster than most other countries in SSA. Despite the initial post-colonial euphoric confidence in education and the subsequent investment in Kenya, the system is now accused of further widening the gap between the rich and poor (Ojiambo, 2009).
This study focuses on the processes of educational expansion and school development, using comparative case studies to evaluate how inequality emerges. While previous studies have largely focused on alleviating inequality, this study identifies the factors generating such inequality in the first place.

Inequality and quality improvement
Quality improvement research has mainly focused on improving school conditions, such as acquiring good quality teachers, including school leaders, and gaining parental and community support. DeJaeghere et al. (2009) noted the lack of discussion of school management and principals' roles in secondary education in SSA compared to other topics, such as educationsector reform, school-to-work transitions, teacher recruitment and training, and financing and governance. However, many previous studies have discussed the role of principals and the quality of teachers as pivotal factors in improving educational quality. For example, although there is a difference in what people expect from principals as school leaders, globally, their role in educational quality improvement is commonly regarded as significant (Eacott & Asuga, 2014). The importance of school leaders for educational quality has also been analyzed qualitatively in the investigation of effective school leadership practices as perceived by school leaders in South Africa (Lumadi, 2017).
To understand the trajectory of studies on this topic, a systematic research review regarding educational leadership and management in Africa was conducted; it demonstrated that the discussion of this topic newly emerged in 2005, and was being carried out until recently (Hallinger, 2017). Many of those studies have insisted on the necessity of ensuring opportunities for school leaders and other teachers for continuous professional development to improve their educational quality in schools. For example, a South African study suggested that principals need specialized leadership preparation with continuous professional development (Naidoo, 2019). An Ethiopian study also insisted on the necessity of continuous professional development for secondary school teachers (Reta, 2017). There are also project evaluation studies on the professional development of school leaders in Ghana (Jull et al., 2014) and South Africa (Weber, 2005).
As discussed above, the principal's role in quality improvement is crucial. Another perspective on improving the educational quality of schools involves building a good relationship with the parents and the community, including the role of school leaders in such a relationship. For example, headteachers play an important role in the relationship of the school with the community, such as parents, school boards, community organizations, and others under the current policy of decentralization and school-based management system (DeJaeghere et al., 2009). Another study on South African secondary schools emphasized the importance of the school's relationship with the community for school management development (Hoadley et al., 2009). It revealed the importance of parental support and engagement, regardless of the community's poverty level. This implies that the quality of schools is not necessarily correlated to the financial level of the areas they are located in; rather, it may also depend on whether they are able to garner positive attitudes from parents regarding supporting schools (Hoadley et al., 2009).
Although these studies highlight the importance of developing school leaders and teachers and establishing a good relationship with the community, how quality improvement can lead to educational inequality is rarely questioned. Although the necessity of quality improvement should ideally be correlated with improving equality, several discussions on quality improvement focus on quality improvement for its own sake, excluding the perspective of how quality improvement can benefit students. Ironically, even with quality improvement, the lack of such a perspective may lead to the creation of new inequalities.
While the African literature has extensively discussed educational leadership and management in the cultural, institutional, social, and organizational contexts of schooling, there is limited research on how the 'sociocultural context' of African education impacts leadership, management, and school processes (Hallinger, 2017). In other words, the interactions between educational leadership and management and how they correlate to the given sociocultural context remain unclear.
Incorporating this perspective (i.e., how the given sociocultural context interacts with school development) is important because, as some studies have suggested, the rationale for school development is unclear if it lacks an understanding of the relationship between contextual background and quality improvement. For example, an anthropological study of primary schools in Nigeria stated the necessity of understanding local complexities by exploring the troubled issues of community-school relations against the policyscape of educational decentralisation (Dunne et al., 2020). The study depicts how communities are influenced by their local social hierarchies in their participation in school management, which keeps certain groups of people away, and shows that community-school relations will not work if schools are insensitive to the complexity inherent in the community. According to another study in Uganda, school leaders' confidence in obtaining support from parents was not merely determined by a school's size or where it was situated (DeJaeghere et al., 2009). Hence, the study concludes that in order to develop headteachers who build good relationships with parents and the community, there should not be a one-sizefits-all training, but rather focuses on context-specific skills which are difference depends on the school size, resources, gender, and the location of the population in that school.
It is critical to understand the contextual background of and underlying reasons for the interactions between such school leaders and parents. However, there is a shortage of discussion of their dynamics. This might be due, as Hallinger (2017) has noted, to the methodological tendency of studies on educational leadership and management to rely heavily on the quantitative method. To address this limitation would require a qualitative field survey that focuses on the contextual background and analyses its relationship with school development. Doing so can reveal how educational inequality emerges in the process of school development that aims at the improvement of the quality of education.
Furthermore, investigations of leadership in 'small schools', 'rural schools', and schools in 'challenging contexts' are needed to assess how leadership and management processes unfold in the various cultural, institutional, and organizational contexts of Africa (Hallinger, 2017). Understanding how to lead and manage these schools is essential to identifying the difference in this regard between African and Western societies (Hallinger, 2017). To explore the dynamics of school development, this study examines such schools by focusing on the process of interaction between school leadership, households, and the contextual situation.

Fairness and unfairness
Despite efforts by various governments around the world, education systems have been criticized for remaining unequal or breeding inequity. Studies focusing on Kenya noted the importance of supporting marginalized groups -such as girls, slum residents, and the poor -by clarifying the factors preventing children from continuing to avail themselves of a good education (Abuya et al., 2012;Onsomu et al., 2006).
Recently, many studies have discussed the flourishing of private primary schools as a cause of inequality . In Kenya, although students are required to score high in their primary-school leaving examination (i.e., the Kenya Certificate of Primary Education [KCPE]) in order to access top-level secondary schools, most students who achieve higher scores enroll in expensive private schools rather than public schools providing free education (Nation Team, 2017). Government bursaries have also been criticized for granting scholarships based on academic achievement, rather than considering only financial circumstances (Wachiye & Nasongo, 2010). These criticisms are based on the premise that academic achievement is influenced by economic background; hence, a scholarship system that prioritizes academic achievement, which reflects not only individual talent and endeavor but also the student's greater financial resources, is problematic.
Discussing equity in secondary education is challenging; because of systemic issues, inequality cannot be avoided. While equity has received considerable attention, its definition remains unclear, though there is a common understanding that it is related to 'fairness' and 'justice' (Berne & Stiefel, 1979;Gooden, 2015). UNESCO (2014) stated, 'equity can lead to equality' and 'equity' is necessary to realize 'equality'. Equity, in this sense, is a concept that allows for differentiation in the distribution of goods, systems, and outcomes from services to generate greater fairness or justice (Chisamya et al., 2012). UNE-SCO (2018) also acknowledged that equity is the means to ensure that distribution is fair or justified and involves a normative judgment regarding distribution. It implies that perceptions of inequality vary depending on whether people consider educational circumstances 'unfair' or 'fair', and this needs to be discussed carefully to understand the given context.
Previous studies have noted three significant ways of resolving "unfair" inequality: school interventions to improve school quality (Orodho & Lawrence, 2014); household interventions to ease the disadvantage of vulnerable groups (Abuya et al., 2012); and community interventions to prevent significant inequality among communities (Abaya & Normore, 2014;UNESCO, 2015). While several studies have focused on each of these groups, there is not enough discussion of how each interacts with the others (Buchmann & Hannum, 2001). This study examines the process of expanding and improving educational opportunities from a micro-perspective in order to understand interactions between schools, households, and communities.
Some view education expansion as mainly benefiting dominant-society groups or classes (Boli et al., 1985). Increasing the length of schooling for the poor is not enough to increase their opportunities for social mobility since the new educational opportunities mainly benefit the rich (Tarabini, 2010). This is because the increase in the population's average level of education generates a parallel increase in the required years of education for high-income careers in the labor market. Even if high-quality education benefits the wealthy first and spreads to the poor later, educational inequality remains.

Secondary education expansion in Kenya
In Kenya, there has been increased research on educational leadership and management compared to other African countries (Hallinger, 2017). Its net enrollment rate (NER) for primary education reached 91% in 2017 (Republic of Kenya [RoK], 2018). When compared with the mean NER in SSA (78% in 2017), Kenya's rapid development is evident (UIS database). Secondary education is also expanding to accept primary school graduates, and its NER increased from 24% in 2007 to 51% in (RoK, 2010. This rate is also higher than the mean NER for SSA countries, which was 34% in 2017 (UIS database).
Public secondary schools in Kenya, which accept students based on their KCPE exam results, are classified as national, extra-county, county, and sub-county schools. In 2017, of the approximately 800,000 students enrolled in secondary schools, 3% were in national schools, 8% in extra-county schools, 17% in county schools, 62% in sub-county schools, and 9% in private schools (Wanzala, 2017). As most graduates were enrolled in sub-county schools, it indicates that there are fewer top-level schools, and it is difficult to enroll in those schools.
Sub-county schools, which are at the bottom level and offer education primarily to day scholars, set relatively lower school fees. These schools have been growing recently and have contributed significantly to the increase in access to secondary education. However, students' low scores in the secondary leaving examination (i.e., Kenya Certificate of Secondary Education [KCSE]) indicate an emerging issue of quality and inequality (Orodho & Lawrence, 2014). Graduates from sub-county schools, which often lack basic educational resources and qualified teachers, frequently face difficulties in proceeding to higher education and career advancement (Ngwili, 2014).
This means that while an increasing number of schools have expanded educational opportunities, the need to improve the quality of the education provided persists. Although the government has recognized the inadequate quality of new schools, it has focused on budget distribution to vulnerable children rather than investing in the schools themselves to improve quality and thereby enhance equity (RoK, 2012). For example, the government prioritizes bursaries for vulnerable children (RoK, 2015) and establishes at least two national schools in each county to offer the best quality education to rectify regional inequalities (RoK, 2012).
As mentioned earlier, the Kenyan government categorizes and ranks schools into four groups, resulting in inequality. However, there are policies in place to balance access to the best education regionally and provide extra support depending on the students' backgrounds. This system attempts to secure opportunities for students who have achieved the academic standard in KCPE required to study at the best institutions.
Another issue is the difference in the school fees for boarding and day schools in secondary education. Even if students perform well in their KCPE, they cannot enroll in toplevel schools without paying boarding fees. If students cannot afford boarding fees, they have no option but to enroll in the nearest schools, which in most cases are sub-county schools.
It might be true that establishing sub-county schools that are cheaper than top-level schools can enhance the enrollment of vulnerable groups previously excluded from the education system. However, providing opportunities in the lowest-quality schools is not sufficient for easing inequality. This wide variation among schools by category has major implications for the students' prospects (Oketch & Somerset, 2010). Although being a secondary school graduate is significant, it is important to most students that they can continue their education at a public university (Oketch & Somerset, 2010). It is thus necessary to ensure minimum quality standards to reduce inequality in education.

Data and methods
This study employs an in-depth qualitative method based on fieldwork in rural Kenya to analyze emerging educational inequality. The fieldwork was conducted five times over 4 years (2014-2017) for a total period of 7 months in a sub-county (referred to as X subcounty) within Busia County, western Kenya. Busia County has a high enrollment rate in primary education but offers limited secondary education opportunities. Hence, establishing new secondary schools is an urgent priority.
To summarize educational expansion in X sub-county, basic information was collected from each school, the education office, and the office for the constituency development fund. This study focused on three schools and conducted analysis through participant observation and semi-/non-structured interviews (which mainly comprise open-ended questions) of teachers and students. Participant observation was employed to assess daily lives in and out of schools and included collecting ideas and opinions in classes and meetings, as needed. The author spent time with students and teachers during and after school hours by visiting their homes or chatting in places such as the market in the evening after school and on weekends. This was done to understand the participants' contextual backgrounds and establish good rapport. Teachers were interviewed to gain knowledge of the establishment, management, and improvement of the schools and the relationship between the schools, community, and parents. Interviews with students focused on the effect of the quality improvement of the school on their educational opportunities and revealed their school choices and experiences. Further, households in the vicinity of the schools were interviewed to assess the neighbors' school choices. The author did not use the term quality education, which often has no clear definition, with particular indicators. Rather, it was used to evaluate how teachers, students, parents, and neighbors interpret "quality of education" and what "improvement" means in the context of school development.
Inductive analysis was used to ensure that the findings were derived from the dominant themes found in observations and interviews. Data from interviews and observations were analyzed according to the general guidelines of thematic analysis of Braun and Clarke (2006) for identifying, analyzing, and reporting patterns (themes) within data.
The data were triangulated by observations (in teachers' rooms, classes, and staff meetings), document analysis (data from public offices and schools), and interviews (with teachers, students, parents, and neighbors). Follow-up visits for observation and interviews were conducted during the four-year period of the study. Ethical approval was obtained through the ethics review board of the author's institute. To the best of the author's knowledge, participants were not harmed physically, psychologically, or socially. Additionally, a research permit was obtained from the National Commission for Science, Technology, and Innovation of Kenya. To ensure confidentiality, the names of participants, schools, and areas have not been disclosed.

Participants
Since the 2000s, the number of public secondary schools has increased rapidly in X subcounty (Figure 1), and in 2017, there were two extra-county schools, one county school, and 18 sub-county schools. This study focuses on the three sub-county schools (referred to as schools A, B, and C), established in 2006 to compare their development process and analyze differences in their development.
The high population density of X sub-county meant that with the establishment of new schools, more schools were accessible by foot, providing students several options to choose from. Thus, this area was appropriate for analyzing the impact of the increasing number of secondary schools and for understanding future educational expansion in Kenya.  1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Extra-county County Sub-county

Results
The basic information collected in the three schools, A, B, and C, represented their respective school development processes. Particularly significant were, findings on the performance of KCSE and the number of students in each school. First, the gap in the mean scores in the KCSE of the schools widened across the years (Table 1). Using a rating system, the score was divided into 12 levels, with 1 being the lowest and 12 being the highest. The KCSE results were almost the same in 2010. However, in 2015, School A achieved 8.05, which was the third-highest score in X sub-county and was followed by two extra-county schools. Meanwhile, School B scored 5.72, which was at the same level as the 5.69 mean score of X sub-county. School C obtained a score of 4.96, which was fourth from the bottom in X sub-county.
Further, the comparison of each student's score illustrated the differences across the years. For example, in 2012, the student with the lowest score from School A scored 3, while the top student scored 11. However, in 2014, all School A students scored more than 7. This implies that academic differences between students within schools decreased in just 2 years. Academic performance also correlated with the number of students; the greater the number of students (Table 2), the better the academic performance.
The correlation between the number of students and academic performance was also evident in all other public schools in X sub-county. Most of the sub-county schools, including schools A, B, and C, lack boarding facilities and provide education for day scholars only. Hence, these schools should accept students that live within walking distance. However, the current correlation between academic performance and the number of students suggests that each school may be accepting students for reasons other than where they live. The next section describes each school's development process to understand why subcounty schools differ across years.

School A: A top school among the sub-county schools
School A is located in a densely populated area, allowing it to maintain a larger number of students on a continuous basis. This number decreased once in 2012 after two schools were established nearby the previous year; however, School A was unaffected when two more schools were established in 2014. Based on the data collected in 2016, most students referred to its KCSE as the reason for choosing School A. The principal also explained that his school was "the secondbest alternative". He mentioned extra-county schools as his competitors but not the newly established sub-county schools. Although both extra-county schools were located within a few kilometers of School A, the principal did not consider their proximity important. This was because the schools charged boarding fees, which are more than double the fees of the sub-county schools. School A was thus popular because it did not require boarding fees but still performed well academically.
To improve academic performance, School A admitted students with a standard of 250 (out of 500) points in the KCPE. This selection was possible because the school already had enough students. Further, it provided extra learning hours, such as during the 50-min lunch break and after the scheduled units are completed-efforts that may have been necessary to ensure students' quality learning.
However, it charged high school fees, and the fees structure meant that it cost almost KES 22,000 (100 KES ≒ 1 USD) per year for one student in 2017. Some teachers thought the fees were higher than those indicated in the structure, presuming they ranged from KES 23,000 to 29,000. These high fees affected the students' choice of school. For example, in 2015, one student who joined School D, a neighboring sub-county school established in 2011, said: "I wished to join School A. But I am in School D due to the school fees". In that year, School D's fee was KES 17,000 per year, and he assumed that School A's fee was KES 35,000 per year. It should be also noted that the fees were flexible and decreased for students who perform well academically. This flexibility was to prevent high-performing students from dropping out or transferring to other schools as the school expected these students to increase the school's mean score in the KCSE.
School A created a virtuous circle by improving its results in the KCSE and increasing its number of students. Meanwhile, the required academic achievements and fees excluded students who did not meet the stipulated standards, even if they resided near the school. Sub-county schools generally provide education to day scholars and are expected to accept students within walking distance. Hence, such an exclusion is unfair because it infringes on certain students' rights to access School A. The waiving of the school fees for students who performed well academically could also be understood as unfair to other students.
The principal explained the necessity to select students according to their KCPE: "The government does not care about their [the student's] life. Ironically, education would be expensive if F4 leavers [secondary school graduates] don't proceed to higher education". He argued that students who may not enroll in universities should not be in secondary schools, and instead, should pursue polytechnics because a secondary education certificate without higher education is increasingly becoming useless due to the rapid educational expansion. Overall, only a small proportion of secondary school graduates go to national universities, and School A justified the necessity of selecting students to avoid spending in vain on secondary education.

School B: The dilemma of quality improvement
School B is 20 kilometers from the central area of the sub-county. It is approximately 7 kilometers from the main road but is situated close to a market that is 15 min away on foot. Its number of students has increased steadily since its establishment, as there are few other secondary schools in the area. However, it has had to confront sluggish growth in its mean score in the KCSE (Table 3). Although its mean score gradually increased, with the school ranking fourth in X sub-county in 2012, it has ranked in the middle since 2013. Furthermore, from 2013 to 2015, three additional secondary schools were established in the neighborhood, and School B had to compete with these schools for students.
Subsequently, the principal decided to improve educational quality to attract more students. The main improvements were increasing the number of classrooms, employing qualified teachers, and increasing the school fees (Table 4).
First, in her efforts to increase the number of classrooms, the principal requested assistance from the local government. As a result, the school no longer needed to conduct classes outside and could divide its students into two smaller classes. However, this increased school expenditure as more teachers had to be employed for the divided classes and a technician was needed to establish the laboratories. Second, the school attempted to employ only qualified teachers from 2016 onward and pay them better to prevent instability. The monthly salary in 2014 was KES 4,000 and KES 5,000 for non-qualified and qualified teachers, respectively. In 2015, these figures increased to KES 8,000 and KES 10,000, respectively. Third, to compensate for the increase in expenditures, the school fees Table 3 Performance of students in KCSE (2009KCSE ( -2015 Source: Created by the author, based on research by Ogawa (2017b) Year 2009   increased from KES 18,000 in 2015 to KES 23,000 in 2016. As a result, School B went from being the area's cheapest day school in 2014 to the most expensive in 2016. Some students were proud of being in a "developing" school, and the main reasons for enrolling in School B changed between 2014 and 2016. In 2014, most students said they joined because it was the nearest or cheapest school, while first-year students in 2016 said they chose School B because of its quality. These students expressed pride in being enrolled in School B and mentioned the superiority of the teachers' qualifications and school facilities compared to other new sub-county schools in the neighborhood.
The quality improvement, however, negatively impacted the earlier enrolled students. The increased fees placed an extra financial burden on them, with some students complaining that "the teachers don't understand our parents' situation". The number of students dropped dramatically in 2016 because of the increased fees. For example, the number of new third-year students fell to less than half the number of second-year students the previous year, with most transferring to other low-fee sub-county schools. The students who transferred needed only to purchase a new school uniform, which was cheaper than paying School B's increased fees.
The large balance of unpaid school fees could also be why students transferred to other schools. This balance is recorded by the schools; usually, students can obtain their secondary education certificate only after they have paid their school-fee debt. Transferred students, however, can obtain a certificate from their new school, and hence, students sometimes transfer to elude pending fees.
School B, which tried to improve its quality, faced a financial crisis because of unpaid balances and decreased number of students. Although parents and students understood the importance of quality education as well as the corresponding higher fees, they still faced the reality of insufficient budgets.

School C: A vicious cycle
School C is 6 km farther than School B from the main road. The narrow road to School C is too rocky and hilly to pass by car, and people use motorbike taxis that charge KES 50 (one way) to reach the nearest market. The school is located between two rivers, and during the rainy season, crossing the bridges can be dangerous. Hence, two schools were established on opposite sides of the rivers in 2013 and 2014.
The school established in 2013 (referred to as School E) presented another disadvantage to School C. There was a rumor that School E had benefited from being established near the home of a local politician. Students could be supported with KES 5,000 per year through bursary funds provided by the local government. Although these bursaries should be distributed according to financial status, people thought they were distributed unfairly, and the students in the school near the politician's home would benefit more. The deputy principal of School C thought that this was one of the reasons why his students were absorbed by School E. He also mentioned that "education is purely affected by the politics". Teachers in other schools had a similar viewpoint.
In 2016, School C was forced to stop serving lunch because of the build-up of unpaid school fees, prompting further transfers to other schools. In its worst period, it had only 45 students and began to accept students without any conditions. A part of the school fees, set as low as KES 16,000 per year, could be paid in crops instead of cash.
Teachers employed by School C were all qualified, but their salary, KES 9,000 per month, was lower than that offered by other schools. Most of the teachers were not from the area around the school and were required to rent rooms near the market. This meant that teachers needed approximately KES 100 per day to commute. Further, it was common for them not to receive their salary on time, due to the unstable payment of fees by students. Hence, School C was not a popular place to work, and the teachers often transferred to other schools with better working conditions. Consequently, the school struggled to provide regularly taught classes.
It should be noted that, initially, the neighborhood wanted the school to be built and residents contributed to its establishment. The deputy principal explained that "the school was wanted by the community. But the community has gone". A decade later, and with the establishment of new schools, School C simply became one of the options for education. People could compare the characteristics of the schools, such as the distance from their homes, school fees, and quality of education. Comparing School C with School B, the deputy principal of School C stated: He implied that there was a difference between the residents in the surrounding area and those who enrolled in the schools. The difference between Schools B and C was caused by the latter; students enrolled in School C consisted of those who did not have the option to enroll in other schools.
When asked, "Why did you choose this school?" some students replied, "I did not choose", emphasizing "choose". Their explanations for being sent to the school included a) "lack of money", b) "parents decided [apart from my desire]", and c) "near my home". There was little possibility for School C to increase its number of students unless it became a more attractive option. It faced the risk of closure, but for nearby students with vulnerable backgrounds, it was the only opportunity to continue their education.

Discussion
The three cases revealed the interactive process of school development between schools, households, and communities. This process was strongly affected by the households' ability to pursue better education. School A created a virtuous cycle of enhancement by improving its educational quality and increasing its students, although it increased school fees. Meanwhile, School B struggled to compete for students because of the students' increased choices due to the new schools in its neighborhood. School C strived to improve quality because of insufficient income caused by a shortage of students and their unstable financial condition.
A similar correlation is revealed in a case study in Uganda; there, larger schools have more resources than smaller schools, which are often newly established, particularly in rural areas (DeJaeghere et al., 2009). This interrelationship between school performance, number of enrollments, and school income has been illustrated in the case of special schools in Kenya (Nderitu & Ngunju, 2014). The higher the performance of the school, the more students it has, and the more income it generates, and vice versa.
As noted in previous studies, having a management strategy is important for school development (Uchendu et al., 2015). Although an increase in enrollment leads to increased school income and improved quality of education, the current study implies that the school management strategy can also give rise to inequality. Inequality among sub-county schools does not refer only to the inequality between newly established and older schools but also to the inequality created by, and inherent in, the process of school development. When schools fall into a vicious cycle and are surrounded by unequal conditions, their upward mobility becomes difficult.
Further, while improving educational quality, it is important and necessary to achieve "fair" equality. This study demonstrates that the requisite increase in school fees created another "unfair" inequality that pushed out vulnerable students from the school.
As noted, In Kenya, each secondary school sets its rate for school fees and decides how to use them. Students can also choose their own schools as long as they meet the school's requirements. This flexibility might enhance competition between the schools and result in educational improvement as a whole. A liberalized education system could encourage schools to compete for students and find creative ways to satisfy students' and parents' needs and preferences (Uchendu et al., 2015). Further, merely expanding educational opportunities may be important even if the quality of education is insufficient, as students require only a secondary school certificate. In this sense, the unequal quality among schools does not necessarily imply "unfairness" and might even be regarded as "fair", as an education that meets some students' needs is nevertheless provided.
However, like the students in top-level schools, most students in sub-county schools desire to attend university (Oketch & Somerset, 2010). They understand that getting a decent job with only a secondary education certificate is not realistic. Therefore, it is not fair to provide only low-quality, cheap, sub-county schools without ensuring minimum quality standards.
The current differences in the amounts of money required within sub-county schools under the same category restrict students according to their academic performance and financial background. The three schools discussed here have developed differently because educational expansion creates competition among schools. Previous studies have demonstrated the necessity to improve educational quality on the assumption that quality improvement would benefit students. It is important, however, to consider students' fluidity in the process of quality improvement; this study found that students and their families used dropping out and transferring as an educational strategy.
Currently, more than 60% of secondary school students are enrolled in sub-county schools. This ratio is likely to increase, given the current NER of 51%. The unequal school categories were implemented by the government as part of the development of the secondary education system. However, inequality within the same category (i.e., sub-county schools), occurred unintentionally due to school choices. Although establishing schools in rural areas is regarded as a step toward expanding educational opportunities, options are limited because of economic status, suggesting that each sub-county school is a cardinal point for stratifying the students' financial backgrounds. Previous studies have demonstrated the existence of educational inequality based on gender, ethnicity, financial background, location, and other factors (Abuya et al., 2012;Rolleston & Iyer, 2019;Sarmiento et al., 2019). This study analyses educational inequality by delving into the process of quality improvement through the stratification of sub-county schools.

Conclusion
Expanding educational opportunities and improving quality are important facets of education. However, because economic inequality may be reproduced in the form of educational inequality, it is also necessary to pay attention to the beneficiaries, to ensure that the UN 2030 Agenda for Sustainable Development, including its focus on "equity", is maintained. This study's findings suggest that the unplanned establishment of new schools restricts vulnerable students from continuing their education by enhancing competition among schools for quality improvement. To achieve more equitable quality education for all children, it is important to not only consider schools as a whole but to also focus on students as individuals.
This study has certain limitations in that it did not discuss the issue of inequality among different school categories or within other school categories; it focused only on inequality among and within sub-county schools. To further expand our understanding of this topic, it is necessary to consider a long-term perspective and conduct subsequent research on the impact of inequality on students' futures.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.