How Far Is Progress? Gender Dimensions of Student Enrollment in Higher Education in Ghana: The Case of Kwame Nkrumah University of Science and Technology

Despite the World Declaration on Higher Education for the 21st Century that education is the ultimate support of human rights, and that access and participation remain key catalysts to accelerating equal educational opportunities for all, females continue to trail their male counterparts in educational outcomes across sub-Saharan Africa. This study focuses on the Ghanaian context and assesses the gender dimensions of student enrollment in higher education, highlighting the disparities therein. We adopt a typical case study design, and a mixed method approach, involving quantitative analysis of student enrollment data, qualitative interviews, and focus group discussions with administrators and students at the Kwame Nkrumah University of Science and Technology. We find generally low female participation across all colleges of the university, except for the College of Health Sciences. We also find low female participation across three dimensions: qualified applicants, admissions, and enrollment. The causes of disparity in the dimensions of student enrollment are rooted in four main factors: institutional, socio-cultural, economic, and awareness of STEM policy and opportunities. The present findings have implications for the implementation of STEM and other gender policies at both the pre-tertiary and tertiary levels of education across sub-Saharan Africa. Plain Language Summary How Far Is Progress? Gender Dimensions of Student Enrollment in Higher Education in Ghana: The Case of Kwame Nkrumah University of Science and Technology This paper examines the gender dimensions of higher education in Ghana between 2010-2020. It seeks to understand the extent of gender disparity in three key dimensions of student participation in higher education: qualified applicants, admissions, and actual enrolment. The paper also examines the causes of gender disparities across these three dimensions, and the impact of Science, Technology, Engineering, and Mathematics (STEM) policies intended to improve female participation in these areas of science. We use a typical case study research design, and a mixed-method approach involving quantitative analysis, interviews, and focus group discussions. We chose the Kwame Nkrumah University of Science and Technology in Kumasi as our typical case, to help examine in-depth, the trends and causal mechanisms underlying gender participation along the three dimensions. Our findings show that gender disparities continue to exist, with particularly wide disparities in some academic disciplines. Notwithstanding, the university seems to have made significant strides in bridging the gender gap in enrolment in the last five years of the study period. The full potential of STEM policies, however, remains underexplored, importantly because the targets of the policy, females, are largely unaware of the opportunities available to them to pursue higher education. We also report the socioeconomic and socio-cultural bottlenecks girls continue to face in their quest to attain higher education, and the key role female role models can play in improving gender parity in higher education. Given the limitations of single case studies, we recommend large-n or comparative designs in future research on the subject


Introduction
The pursuit of education to the highest possible level as a right that all must enjoy, regardless of gender, race, and ethnicity, has long been a fact with global consensus (Christie, 2010;McCowan, 2010).In contemporary history, Article 26 of the 1948 Universal Declaration of Human Rights enjoined nation-states to, according to what a nation's resources can bear, grant equal opportunities to both boys and girls to attain the highest level of education (Bunyi, 2008;Mikkola & Nugmanova, 2020).Nearing the turn of the 21st century, the World Conference on Higher Education in 1998 similarly reiterated the value of education as the ultimate support for human rights (Calder & Clugston, 2003).Even more explicit in recent times, the United Nations Sustainable Development Goal (SDG) 4 calls on member states to ''ensure inclusive and equitable quality education and promote lifelong learning opportunities for all.'' 1 The commitment specifically enjoins member states to eliminate gender disparity in education and ensure equal access to affordable and quality education for both men and women, up to the tertiary level by 2030 (Ferguson & Roofe, 2020).
While significant progress has been made in increasing access to higher education (tertiary level), gender disparities continue to persist.Current evidence shows that female enrollment in higher education outstrips that of males in all regions of the world except sub-Saharan Africa (SSA; Norgbey, 2018;Saadat et al., 2022;Salmi & D'Addio, 2021).The SSA region has the worst disparity against women with a Gender Parity Index (GPI) of 0.73 in 2015, and 0.8 by 2020 (World Bank, 2022).The region trails the rest of the world with 42.3% female representation at the tertiary level (Salmi & D'Addio, 2021).Countries in SSA are also significantly behind in terms of female participation in the Science, Technology, Engineering, and Mathematics (STEM) programs (Salmi, 2020;Tandrayen-Ragoobur & Gokulsing, 2021;Wurah-Norgbey, 2019).The benefits of gender parity in higher education for the SSA are, however, enormous at both the individual and systemic levels.A high quality of human capital, lower unemployment rates, increased revenue, lower dependency on social services, and increased civic, voluntary, and political participation offered by higher education (Salmi, 2017;Salmi & D'Addio, 2021), imply that the SSA region can accelerate its development transition if adequate attention is given to addressing the gender gap, especially at the tertiary level (Mkude, 2011).
The present study focuses on gender disparities in student enrollment in higher education in Ghana and the underlying causes across three components: qualified applicants, admissions, and actual enrollment.It also assesses the impact of pro-female policies aimed at improving female participation in STEM in Ghana.Our review of extant literature shows that while research on the various levels and types of inequalities at the tertiary level of education, with specific allusion to access and participation within the SSA region, exists, (see, Appiah-Castel et al., 2020;Atuahene & Owusu-Ansah, 2013;Dube, 2015), the analysis of the gender dimensions (the disaggregation of student participation by gender [male/ female]) beginning from the stage of qualified applicants remains an underexplored field of inquiry within the SSA higher education literature.Past studies have tended to examine gender disparities, using the admission stage as the starting point of the analysis.We find it more insightful, however, to answer the question, ''how many (or what proportion of) males/females qualified to be admitted'' as the starting point of empirical investigations on gender disparities than ''how many (or what proportion of) males/females were admitted or enrolled.''Also importantly, studies on pro-female policies and their effects on challenges to parity are currently limited within the SSA literature.Our study extends the existing literature on gender disparities in higher education within the SSA, by analyzing the differences right from the application stage.
We examine the gender dimension across three components of higher education: qualified applicants, admissions, and actual enrollments in Ghana's higher education from 2010 to 2020, with specific emphasis on the Kwame Nkrumah University of Science and Technology (KNUST).We also appraise pro-female specific policies and their responsiveness to the gender parity gap which has persisted over the decades, and the prevailing factors or challenges undermining such policy interventions.Our analysis shows that there is relatively low female participation across all colleges in the KNUST except for the College of Health Sciences.The study also reveals low female participation across the three dimensions of enrollment: qualified applicants, admissions, and actual enrollment, and identifies the disparity in the dimensions of student enrollment as rooted in four main factors: institutional, socio-cultural, economic, and awareness of policy and opportunities.We also found the STEM policy to have been generally less successful at increasing female admission and enrollment in the Sciences, although some progress had been made in the engineering programs.We believe these have implications for pro-female policies in education.
The paper proceeds with a brief background of Ghana's tertiary education system to set the context for the arguments within the paper.We then continue with a review of the relevant literature on gender disparity in higher education, the root causes, and policies aimed at addressing the challenge.This is followed by the data and methods, results, and discussion sections.The paper concludes with a recap of the main findings, reflections on the implications of the findings, policy recommendations, and areas for further research.

Ghana's Higher Educational System
The higher education (HE) system is the last stage of the educational system in Ghana.It is a highly diversified system comprising all post-secondary institutions including universities, training colleges, and professional institutes, among others.The history of higher education in Ghana can be traced back to the 1940s (Apusigah, 2009;Manuh et al., 2007).The system was modelled after the British colonial system with an initial intention of locally training men and women with professional qualifications to propel the social and economic development of the colony (Armah, 2022;Gyamera & Burke, 2018).Postindependence, the goal of higher education has been to hasten national development and the production of skilled human capital to meet the development needs of the nation (Apusigah, 2009).
The first university established was the University College of the Gold Coast, now the University of Ghana.It was founded in 1948 in response to a recommendation by the Bradley Commission at the instruction of the colonial governor, Alan Burns (Armah, 2022).In 1952, the Kwame Nkrumah University of Science and Technology was established to provide advanced technical education in the nation (Atuahene & Owusu-Ansah, 2013;Tsevi, 2015).In recent years, however, the number of higher education institutions has seen unprecedented growth because of a sharp increase in private higher education institutions.The sector increased from 2 private universities in 1999 to 81 private universities and colleges of education by 2019 (Kamran et al., 2019).Currently, there are 304 accredited institutions of higher education, including 15 public universities in Ghana, and 105 private institutions offering higher national diplomas and degrees (GTEC, 2022).
Notwithstanding the substantial growth in the number of higher educational institutions, several challenges remain.Predominantly, there is a lack of institutional capacity to absorb the increasing demand for higher education, and disparities in participation.Gender-based disparity is one such inequality which has characterized higher education, especially for women in Ghana.The evidence shows constant patterns of disparity across all types of tertiary institutions in Ghana as of 2018.Public universities, for example, which recorded the highest level of enrollment (58%), out of a total of 258,669, had 61% males and 39% females.In Technical and Polytechnic institutions, males comprised 66% whereas females constituted 34% out of 50,932 students enrolled.Similar disparities were evident in Public Colleges of Education (52% males and 48% females), Public Specialized Institutions (55% males, 45% females), Private Tertiary Institutions (57% males and 43% females), and Colleges of Agriculture (86% males and 14% females).Only public and private Nurses Training Colleges had female overrepresentation among the students: 26% males, 74% females and 29% males, 71% females, respectively (RPPD, 2018).Besides, gender disparities are more pronounced in STEM programs (Wurah-Norgbey, 2019).The present study is designed against this backdrop of information on Ghana's higher educational system.We now proceed to review the wider global, SSA and Ghanaian literature on gender disparities, the causal arguments, and remedial policies adopted to achieve gender parity, especially in STEM.

Gender Disparity in Higher Education
Gender disparity in higher education has been a topical issue in the literature on gender, gender empowerment, and gender equality.The university system in particular has been a major discussion point for analysts.The argument has been that universities are major institutions for gender empowerment and can bridge the gender gap through policies which promote gender equality, diversity, and inclusion (Rosa & Clavero, 2022).Notwithstanding, universities continue to be gendered at all levels, including low female representation in senior professorial and decision-making/management positions (Pritchard, 2007;Rosa et al., 2020;Teelken & Deem, 2013;Winchester & Browning, 2015): a factor which, as we will soon see, is argued to limit the availability of female role models within the academic system, and consequently a source of demotivation for some females to enroll in some specific programs.The history of Western university education over the last 800 years provides significant insights as to why certain patterns may have persisted until recently.For instance, one of the oldest universities in the Anglo-Saxon world, Oxford University, admitted only male students for 800 years and only reversed its gender policy to admit females in 1920 (Wan, 2018).Females have, therefore, been historically and traditionally conceived in the literature as being the disadvantaged gender in terms of access and participation in tertiary education.
Recent reports on gender disparity in higher education, and particularly in the universities, however, challenge the conventional narrative of male overrepresentation, especially at the student level.Mounting empirical evidence from both the developed and developing worlds show that female enrollment in higher education has outstripped that of males in almost all the major regions of the world (Morris, 2011;Saadat et al., 2022;Van Bavel et al., 2018), and in nearly all fields except STEM (Holman et al., 2018;Salmi, 2020;Sax et al., 2016;Stoet & Geary, 2018;Wang & Degol, 2017).Across China (Wu et al., 2020), most countries in Asia-Pacific (Saadat et al., 2022;Wan, 2018), the West (David, 2015;DiPrete & Buchmann, 2013;Evers et al., 2006;O'Connor et al., 2015), and even in well-known patriarchal societies of the Middle-East, such as Saudi Arabia (Wan, 2018) and the United Arab Emirates (Ashour, 2020), females are overrepresented in higher education.This dominance by females is both as a share of total tertiary enrollment and as a percentage of the age cohort of women who qualify to access tertiary education (Parvazian et al., 2017).The region with exception to the overrepresentation of females in higher education, however, is SSA.
The empirical literature and data from the World Bank both suggest that the region is yet to achieve gender parity at the tertiary level.The region continues to have a GPI of less than 1 (World Bank, 2022) and has males constituting nearly 58% of student enrollments (Salmi & D'Addio, 2021).For instance, in Nigeria, Oludayo et al. (2019) found that across all accredited universities, polytechnics, and colleges in all 36 States of the Nigerian Federation, male enrollments were steadily higher than females across 6 years (2010)(2011)(2012)(2013)(2014)(2015).In Ethiopia, Mergo (2013) reports a wide gender gap in favor of males across three (3) carefully chosen universities-Addis Ababa, Haramaya, and Jimma Universities.The gender gap in enrollment was found to be greater than 50%, 80%, and 90% at the bachelors, masters, and doctoral levels, respectively.Females outstripped their male counterparts only in academic dismissal.In Rwanda, studies show that despite the conscientious efforts of the government to promote gender equality across the entire society, including the educational sector, female enrollment continues to be significantly lower at the secondary and tertiary levels (Ntawiha et al., 2018).Similar accounts of gender disparity in favor of males in higher education have been reported in the past for Ghana (Atuahene & Owusu-Ansah, 2013) and Zimbabwe (Chauraya, 2014).

Causes of Gender Disparity in Higher Education and Remedial Policies
Several factors have been identified as contributing to gender disparity in higher education in the literature.Saadat et al. (2022) conceptualize these factors around four broad themes: individual, social, financial, and educational.They argue that individual factors relate to personal attitudes and motivations for higher education, while social and financial factors are associated with the socioeconomic backgrounds of students.The educational factors on the other hand relate to the previous academic performance of students prior to the tertiary level, and the meritocratic systems universities employ in admitting students, which inadvertently exclude those with relatively weak pre-tertiary academic performances from gaining admissions (Saadat et al., 2022).
Within the SSA context, the literature suggests that some of the factors contributing to the relatively low female participation in education include structural and social barriers, such as gender-unfriendly institutional structures and leadership, sexual harassment, family preferential behavior (Mergo, 2013), and the heritage of colonialism in the sub-region (Darvas et al., 2017).Konte and Tirivayi (2020) include economic factors, harmful cultural practices and unequal gender norms, school and learning environment, crises, armed conflicts, and outbreaks as key contributors to the disparity within SSA.Notably, many of these structural factors which are argued to be deeply embedded in the organization and functioning of African societies, interplay and manifest within the educational system, leading to disparities in access and participation in education at all levels, especially, at the tertiary level (Molla & Cuthbert, 2014).Patriarchal societal perspectives on gender roles have, for instance, been argued to regulate educational opportunities for female children, such that parents place greater emphasis on the education of boys than girls: a phenomenon which also tends to guide higher education (Machira, 2013;Norgbey, 2018).It is strongly perceived that the principal roles of women are that of a sexual partner and mother, leading to low investments in the education of females (Chisamya et al., 2012).Also important to the present discussion, and as earlier intimated, is the underrepresentation of females at the faculty and senior management levels within the university system.Ntawiha et al. (2018) found that in Rwanda, women's underrepresentation in the teaching staff at higher levels of education had resulted in low levels of enrollment for females in some programs.The scholars attribute the phenomenon to a lack of motivation on the part of the female students, due to the limited number of female faculty role models who could inspire them to similar heights.Furthermore, gender-based violence coupled with rampant early and unwanted pregnancies at the primary and secondary levels have also been found to hinder females' progression into higher learning institutions (Morley, 2005;Ntawiha et al., 2018).Darvas et al. (2017) also assert that within SSA, the level of education of parents is a significant predictor of the educational outcomes of their wards.Accordingly, students from households where the household heads had at least, completed secondary education, were ten times more likely to enroll in higher education than those from households where household heads did not complete primary education.The authors trace this likelihood to the enhanced socio-economic capacities of educated parents to afford higher education for their wards.
In Ghana, Atuahene (2013), avers that the institutional patriarchal system established by the colonial universities, which favored male enrollment over females, and the lack of effective policies to correct such historical errors by first-generation African leaders and policymakers, explain the recurring gender-based disparity at the higher education level.The socio-economic background of students coupled with the admission selectivity in public universities have also been established as major hindrances to participation in higher institutions of learning within the Ghanaian context (Atuahene & Owusu-Ansah, 2013).
On the remedial side, local and international policies implemented to address gender disparity in higher education over the years, have differed in their foci and expected outcomes.While some interventions have sought to generally expand access to higher education for all groups, and as a consequence, increased female enrollment (Brock, 2010;Chang, 2018), others have specifically targeted females as part of affirmative action policies in higher education (Bagde et al., 2016;De Wit & Bekers, 2020).Even further, some policies have targeted increased female representation in specific fields such as STEM (Appiah-Castel et al., 2020).And while these interventions are context and country dependent, the literature reveals that generally, the different gender policies have typically included quota systems for women as part of general admissions, lower grade cut-off points for women in selected fields, and scholarships for socioeconomically underprivileged females (Salmi & D'Addio, 2021).These interventions have also been reported in the Ghanaian context for gender policies aimed at improving female participation in higher education (Appiah-Castel et al., 2020;Atuahene & Owusu-Ansah, 2013).

Research Questions
The present study builds on the extant research reviewed above.We take note of the empirical findings on gender disparity in higher education, the explanations advanced for the causes of gender disparity, and the policies adopted to address such disparities, and situate our study within these theoretical and empirical framings.We, however, broaden the existing conversation to cover the wider continuum of student participation, beginning from the application stage to enrollment.We believe that commencing the analysis of the gender dimensions/disaggregation of student participation by gender (male/ female), from the stage of qualified applications rather than the admissions offers more comprehensive insights into the phenomenon of gender disparity.Our study accordingly asks the questions: RQ1: What is the extent of gender disparity in terms of qualified applicants, admissions, and actual enrollments at the university level in Ghana?RQ2: What are the main causes of gender disparity along these three dimensions?RQ3: What has been the impact of pro-female gender policies implemented to achieve gender parity?
Our expectations based on the insights gleaned from the above literature are that there would be the existence of appreciable gender disparity in our three dimensions of interest.We also expect that some of the key causes of gender parity in higher education discussed above would reflect in our case.Furthermore, we expect that the gender policies implemented so far would result in a substantial narrowing of the gender gap across all disciplines over the period of a decade of implementation.We elaborate on our research design and methodological approach in the section below.

Case Selection
Our study adopted a single case study design.We chose a typical case which is representative of our population of interest (Gerring, 2017).The reason for our choice of a typical case study design was to enable us to generate in-depth insight into the dimensions of gender disparity in higher education, to help improve our understanding of the phenomena across other higher institutions within and outside Ghana.As our population, we focused on institutions of higher education in Ghana.We then selected the university system as our unit of analysis.We preferred the universities to other institutions of higher learning, such as teacher training colleges and polytechnics, due to the wide variety of programs offered by universities.Universities also typically have larger student populations, compared to other institutions, and are typically the first choice of post-secondary education for most young people.
We selected the KNUST as our typical case, representative of the university system in Ghana.Our choice of KNUST was informed by the following reasons.Firstly, the institution offers a wide variety of programs, especially in the STEM fields where, conventionally, females have been found to be less populous in participation.Secondly, the university implements STEM policy due to its science and technology orientation to augment female participation.Thirdly, it is more accessible to students across the country due to its central location.Lastly, it is the preference of most students due to its high academic standards.Indeed, in November 2020, the institution was ranked the best university in Ghana and West Africa for the second time by the U.S. News and World Report (Appiah-Castel et al., 2020).It was also adjudged in 2022 as the best university in Africa in terms of quality education, and the 14th best in the whole world, by the Times Higher Education Impact Ranking 1.The university was legally established in 1952 and granted full university status in 1961 by an Act of Parliament.The university had a total population of 64,187 as of 2020.It consists of six ( 6

Data Collection
We employed a mixed-method approach involving quantitative and qualitative data collection and analysis for the study.Fieldwork took place between May 2021 and August 2021 on the campus of KNUST.First, we collected secondary data from the university's Quality Assurance and Planning Unit for the period 2010 to 2020, on our three dimensions of interest: applications, admission, and actual enrollment, for our gender-based quantitative analysis.The data provided information on male/female participation in programs in all three dimensions across all six colleges of the university, in response to our first research question.
Second, our qualitative data collection involved three methods: semi-structured interviews, focus group discussion (FGD), and expert interviews.The six colleges formed the sampling frame for the selection of respondents for both the semi-structured interviews and the FGD.The qualitative component of the study sought to gain insights into the interplay of factors contributing to disparity in gender participation, the colleges within which they occur, and the remedial policies implemented to address the disparities, in response to our second and third research questions.A total of 19 faculties embodied all six colleges.Two (2) faculties were simple randomly selected from each college forming 12 faculties, with their corresponding departments totaling 66. Due to the extensive number of programs and time constraints, Slovin's formula was used to derive a sample size of 39 programs.Thus, interviews were conducted with students from at most two programs from each faculty.In all, simple random sampling and snowballing were used to obtain one-on-one interviews with 78 students.A semi-structured interview guide with questions relating to knowledge about the existence or otherwise of gender disparity at the KNUST, the causes of disparity, the existence and awareness of pro-female policies, and challenges to action/policy guided the sessions with respondents.Informed verbal consent was obtained from each participant before interviews were conducted.
The FGD involved students with experience in the various dimensions of disparity across the six colleges.Participants were purposefully selected at various discussion areas in the colleges.Each focus group discussion comprised at least six (6) participants.Seven (7) FGDs were conducted after which the point of saturation was reached.We observed similar responses from previous discussions and, therefore, deemed it necessary to discontinue.Like the interview guide for the one-on-one interviews, a discussion guide with questions on the existence and cause of disparity, the availability of pro-female policies, and associated challenges/opportunities for their implementation was used to steer the FGD.All FGD participants gave their informed verbal consent before the commencement of the sessions.
We also conducted an expert interview with the Registrar of the KNUST Admissions Unit, given the knowledge, experience, and institutional memory the holders of such offices within the university possess.The Registrar has direct oversight of student admissions and is also an important stakeholder in the administration of pro-female policies, particularly at the admission stage.We found the office of the Registrar well placed to provide insights on the gender dimensions of interest, the underlying causes of disparities in participation, and the effectiveness and/or challenges of the pro-female policies in attracting females to the STEM field.The interview was recorded with a mobile phone and notepad after consent was given by the Registrar.
Our use of the three methods of qualitative data collection to elicit responses to our research questions was aimed at triangulating our findings to help improve the validity and reliability of outcomes (Flick, 2017;Heale & Forbes, 2013;Heesen et al., 2019).All our interviews and FGD were conducted in English, the official language of communication and instruction in the university.
Data Analysis.Quantitative data obtained were analyzed using descriptive statistical measures such as rates, percentages, and the gender parity index.These measures gave evidence of the level of disparity and growth investigated along the dimensions indicated.The GPI is given by the Gross Enrollment Ratio (GER) of higher education for females and males.A GPI of 1 shows parity between both sexes, a GPI lower than 1 indicates disparity favoring males, whereas a disparity greater than 1 favors females (Akbash et al., 2018;Amirtham & Kumar, 2021).Cross-tabulation was employed to ascertain relationships between variables.Frequency tables were also generated to identify patterns in the data.Descriptive statistical tools such as bar graphs, line graphs, pie charts, time series graphs and trendlines were used in the data description.Semi-structured interviews and FGD responses were written down during the sessions and exposed to thematic content analysis.The expert interview was, however, recorded, transcribed, and similarly exposed to thematic content analysis.

Demographic Profile of Respondents
Overall, 70 interviews were successfully completed out of the expected 78 with students.The respondents comprised 34% males and 66% females.These included undergraduate students from all Colleges (14% from CoHSS, 7% from CoHS, 14% from CANR, 29% from CoE, 7% from CABE, and 29% from CoS).

Analysis of the Gender Dimensions of Student Enrollment in KNUST From 2010 to 2020
In response to research question 1 which sought to examine the extent of gender disparity within the university system in Ghana (with the KNUST as our case study), we present below the findings of the study in the three dimensions where disparities were expected to manifest, that is, qualified applicants, admissions, and actual enrollment, from 2010 to 2020.Qualified Applicants. Figure 1 shows that the number of male applicants who qualified to be admitted into the institution generally outweighed their female counterparts across all six colleges.In the 2010/2011 academic year, the total number of qualified applicants was 16,946 students: 68% were males and 32% females.About 38% of the applicants belonged to STEM fields whereas 62% of applications belonged to other fields (Art and Built Environment, Agriculture and Natural Resources, and Humanities and Social Sciences).Also, 73% of the qualified applicants in STEM fields were males whereas 27% were females.The other fields comprised 64% males and Qualified Applicants (M-F)

Deficit
Figure 1.Analysis of deficit between male and female qualified applicants to KNUST for each college between 2010 and 2020.
36% females.More females, therefore, seemed to qualify to be admitted into other fields as compared to STEM.By 2019/2020, however, out of 35, 243 qualified applicants, 49% were STEM related whereas 51% belonged to other fields.In the STEM field, males comprised 61% while females constituted 39%.Similarly, other fields comprised more males (54%) than females (46%).The number of qualified applicants for STEM has thus generally increased over the past decade.Also importantly, the percentage of female-qualified applicants has also generally increased in STEM and other fields for the same period.
The study also revealed that in the Health Sciences, more females qualified for admission from 2015 to 2020 as compared to their male counterparts.The Engineering field recorded the worst disparity against females with a deficit of 3,512, even though female-qualified applicants more than quadrupled over the decade (from 261 in 2010 to 1,066 in 2020).The interest of applicants in STEM increased for the period under consideration, especially for females.Overall, qualified applicants increased by 108% in the past 10 years, signifying an increase in demand for higher education at the KNUST.Yet, while the growth rate was generally positive, the number of qualified applicants grew at a diminishing rate (from 39% in 2010 to 25% in 2020).
Students Admitted.The analysis shows that the number of qualified applicants who gained admission into the university over the period substantially increased.Figure 2 shows the growth rate: from 2% in 2010 to 54% by 2020.There was, however, a decline in the growth rate for some years.For instance, admissions for the periods between 2013/2014 to 2014/2015, 2015/2016 to 2016/2017, and 2017/2018 to 2018/2019 academic years dropped by 240%, 210%, and 226%, respectively.Whereas a large proportion of qualified applicants in 2010 belonged to other fields (i.e., CANR, CABE, and CoHSS), we found that more applicants were admitted into STEM than the other fields (represented by 64% and 36% respectively) in the following decade, as seen in Figure 3.Although the CoE admitted the least number of females throughout the period, female admissions out of the total of qualified applicants generally outnumbered that of males by the end of the decade (from 33% in 2010 to 51% in 2020).
Actual Enrollment (New Entrant).Overall, the proportion of female students who enrolled (out of the proportion of females granted admission) as new entrants in the 2010/2011 academic year was significantly low (30% of total enrollment).However, by 2019/2020 female enrollment had overtaken that of males (represented by 53%  against 47%, respectively).STEM fields dominated actual enrollment (represented by 61% in 2020 compared to 45% in 2010).Enrollment grew by 300% over the past 10 years.Disparity against males currently exists in the Humanities and Social Sciences.Female enrollment in Engineering increased by 213% over the decade even though males still generally dominated in the field.Other fields such as CABE and CANR experienced a decline in the gap against females for the period under study.
While there was a general disparity against females for the period, exceptions were recorded in the 2017/2018 and 2019/2020 academic years: female new entrants superseded that of males.There was also an unstable pattern in growth over the decade even though the growth rate increased from 3% in 2010 to 95% in 2020.Figure 4 below shows the proportion of male/female actual enrollments for the past decade.
Table 1 below provides a summary of the participation of students by gender for the study period.It details the actual numbers of qualified applicants, admissions, and actual enrollment represented by new entrants.As can be seen from the table, the consistent decline in the overall numbers from the status of qualified applicants to new entrants is evident.Also importantly, the difference in the numbers of both sexes who are unable to transition from qualified applicants to enrolled students is also especially apparent.

Gender Disparity Analysis in Student Enrollment (New Entrant) Using GPI
The GER was calculated as the percentage of each sex from the total enrollment for each college in each academic year.According to Table 2 below, in the CANR, a GPI of 0.4 especially for the last 5 years resulted, representing a very low level of participation for females.The CoHS on the other hand, reached and maintained parity from 2011/2012 to 2013/2014 academic years.In 2014/ 2015, males dominated participation.Subsequently, however, female participation consistently outstripped the males, as seen in the constantly high GPI of more than 1 over the last 5 years.The CABE is also a maledominated field according to the data.Gender parity was as low as 0.2 in 2014/2015 where females accounted for only 18% of the total student population in the college.Even though the highest GPI was 0.7 in 2016/2017, female participation in the CABE was still considerably higher than in other colleges such as the CANR and CoE.The most poorly performed college in terms of participation was the CoE.Also, throughout the decade, the participation of females in the field of engineering had been less than 25%.The highest GPI (0.3) was recorded in the 2010/2011 academic year.The CoHSS on the other hand, maintained a good female participation level until it finally reached parity in the 2019/2020 academic year.It was quite a surprising finding, that males dominated the CoHSS until 2020 since the field is generally perceived to be female-dominated.In terms of equality in participation, the results show that the university was able to realize its 50/50 goal consistent with national policy in the years 2015/2016, 2017/2018, and 2019/2020.

Factors Contributing to Gender Disparity in Student Participation in Higher Education
In response to research question 2, which sought to understand the causal factors underlying gender disparity in higher education, we present the insights from the qualitative component of the study, gathered through interviews, and focus group discussions, relating to the causes of gender disparity in student enrollment at the KNUST.We address them under four main themes: Institutional, economic, socio-cultural, and awareness of STEM policy.
Institutional Factors.Respondents posited that the admission selection process (cut-off points and programs offered) contributes to the disparity.Furthermore, respondents established that the present higher educational structure also accounts for the low female participation.Focus group discussions revealed that the mixed orientation of higher education does not encourage female participation in male dominating fields amidst cultural and societal barriers.An only female university was suggested by some respondents as a remedy to the challenge.
Socio-Cultural Factors.The study revealed that the culture and training given to students, especially females, at the senior high school level, influence the programs they pursue in higher education.Respondents from focus groups argued that single-sex senior high schools performed better at encouraging females to pursue STEM courses than mixed senior high schools.The respondents, who incidentally came from both single-sex and mixed senior high school backgrounds, observed that mixed institutions do not educate students on STEM programs they can pursue at the Tertiary level nor encourage especially females to pursue the field.Knowledge about STEM is, therefore, virtually absent in discussions at the senior high school level.This, it was observed, ends up influencing the types of programs females read at the tertiary level.It was also emphasized by respondents that the relevance of female role models in increasing female participation in higher education, especially in STEM has been underestimated and neglected.They argued that all other things being equal, the type of role models available to girls determine their level of participation in higher education.
Economic Factors.Poor socio-economic background of students was established as the major contributor to gender disparity in higher education participation.This factor was seen to affect all the dimensions of student enrollment discussed above.It was also discovered that the lack of adequate financial preparation ensuing from the lack of accurate information on the cost of participation prior to admission accounts for low enrollment and completion.
Awareness of Policy and Opportunities.It emerged that the knowledge about the STEM policy at the pre-tertiary education level was very low.Most of the students in the university interviewed as part of the study had no knowledge about the policy.Also, information about scholarship opportunities seemed unavailable to the public.According to respondents, this has led to discrimination in the awarding of scholarships.It turned out that, the targets of STEM policy are not taking optimum advantage of the opportunity due to lack of information.STEM to this knowledge gap, despite the implementation of the policy.

Assessment of the Performance of Pro-Female Policies/ Initiatives to Increase Female Participation in KNUST
In response to research question 3 which sought to assess the effectiveness of pro-female policies implemented to boost female participation, the study found that the institution had implemented the STEM policy (aimed at increasing female enrollment in STEM) and the WiSTEM initiative since 2010.The performance of the STEM policy was, therefore, assessed in terms of the growth in admission and enrollment before and after policy implementation.Growth years considered for the ''before'' and ''after'' policy implementation included 2010/2011 to 2011/2012, 2011/2012 to 2012/2013, and 2017/2018 to 2018/2019 and 2018/2019 to 22019/2020, respectively.Pertaining to admission, the results indicated that, before the implementation of STEM, there was a reduction in the growth of female admission by 12 applicants and 7 applicants correspondently for both growth years.After the implementation of STEM, the trendline depicted an upward trend implying growth in female admission into STEM fields for the initial growth year (2017/2018-2018/2019).In the second growth year (2018/2019-2019/2020), the trend indicated a general decline in female admission by 194 applicants on average.However, it seems that the impact of the STEM policy on female admission was only evident in the Health Sciences and Engineering but failed to improve admissions in the Sciences.Figure 5 shows the trend in female admission during the STEM implementation years 2010 to 2020.Regarding actual enrollment, a similar pattern was observed, as seen in Figure 6.Before the implementation  Quarshie et al.
of the policy, the trendline indicated an increasing trend in the number of females enrolled, although at a diminutive number of one on average in the first growth year.
In the second growth year, there was a downward trend in the growth of females enrolled on STEM generally.While female enrollment increased by 454% in the Health Sciences, and 22% in Engineering, the College of Science experienced a decline from 30.1% to 3.5% from 2017 to 2020.Again, the STEM policy tended to improve female enrollment in Health Science and Engineering but not in the Sciences.

Discussion
The present study reveals that gender disparity continues to exist at the KNUST, despite considerable gains made over the last decade.While female participation outstrips males in the Health Sciences, they generally trail their male counterparts in all other fields, particularly Engineering and the Sciences.As an explanation, the study found the socio-economic background of students as the most cited barrier to gender parity.We found that this barrier cuts across two of the dimensions of student participation: admission and enrollment.Also importantly, the study found that the lack of adequate financial preparation on the part of students toward their university education, ensuing from the lack of accurate information on the cost of participation prior to admission, also contributed appreciably to the low enrollment.As already discussed in the theoretical section of the paper, a plethora of studies have emphasized students' inability to participate in higher education due to their low socio-economic status (Appiah-Castel et al., 2020;Atuahene, 2013;Atuahene & Owusu-Ansah, 2013;Konte & Tirivayi, 2020;Saadat et al., 2022).It seems, however, that females are disproportionately affected in this regard at the KNUST.Hence, despite the appreciable gains made in female participation, they continue to come behind the males in two key dimensions-admissions and enrollment.Admittedly, low socioeconomic status genuinely limits access to entry and continued stay in school.However, this can be argued for both sexes and does not convincingly explain why fewer females persist in the various dimensions.We think, therefore, that familial preferential treatment for males over females in the pursuit of higher education may be of special relevance here (Machira, 2013;Norgbey, 2018;Wan, 2018).Given the limited financial resources of most families, it sounds plausible that when faced with the decision on whom to invest scarce family resources in for higher education, females would be considered less profitable within Ghana's socio-cultural context.Thus, while poverty is undoubtedly a barrier to access to higher education, it is not unitary in its expression toward a particular sex.We are more inclined to believe that, underpinning the persistent gender disparity within Ghana's higher educational system is the socio-culturally biased choices against females in the allocation of family resources for education.This in turn may be aggravated by the lack of information on the actual cost of participation, prior to admission to selected programs.Should families find out much later at the admission stage that the cost of participation in a program is too high, the likelihood of support for the participation of females diminishing can be expected to be reasonably high.
The study also discovered that the university's admission selection process (grade cut-off points) was another major factor contributing to gender disparity and low female participation in higher education.This is congruent with previous findings in Ghana that admission selectivity to public institutions of higher learning is the second main hindrance to participation ( (Atuahene & Owusu-Ansah, 2013).We found that despite the university's goal of achieving 50:50 enrollment, the general selection process and one bonus cut-off point for females in STEM do not yield equal shares in admission quotas since both sexes are competing for the same space.We believe that while upholding the standard, the institution can realize its goal by setting separate admission quotas for both males and females, especially in STEM fields, to accelerate the achievement of parity (See Appiah-Castel et al., 2020;Bagde et al., 2016).
Another factor which was reported to militate against equal access to higher education was the mixed-oriented higher education structure.Konte and Tirivayi (2020) have opined that single-sex school orientation contributes to girls' participation in science and technology education and careers.In some cases, girls in single-sex schools have been found to participate and perform better in mathematics, compared to their peers in mixed schools: single-sex schools seemed to play an overarching role in promoting females into science and technology jobs (Morrell, 2000).They are argued to eliminate the major challenges faced by females in mixed institutions, such as the lack of role models, sexual harassment, choice of programs, unintended pregnancies, drop-outs, etc (Morrell, 2000).The foregoing argument seems to suggest, therefore, that the establishment of single-sex higher education institutions can encourage female participation, especially in STEM.We are, however, cautious in admitting this line of argumentation, given that among other reasons, a core focus of the university system is the promotion of social integration between and among different social groupings in the pursuit of higher education.
Also important, the role of females as role models generally, and especially in STEM came up strongly as a key factor militating against female participation in higher education.As argued by Konte and Tirivayi (2020), female teachers across the SSA region are deemed chief factors in addressing issues about female enrollment and retention.The presence of female teachers/role models contribute significantly to the removal of socio-cultural blockades to female participation in education, as they serve as immediate points of reference and inspiration for young girls (Carrell et al., 2010;Porter & Serra, 2020).As earlier intimated, women's underrepresentation in teaching staff in higher education can result in low levels of enrollment for females, due to a lack of role models to motivate them (Ntawiha et al., 2018).The same challenge has been identified recently in the KNUST in the enrollment of females in STEM programs (Appiah-Castel et al., 2020).The present study confirms the need for the promotion of female role models in STEM as part of deliberate measures to increase female participation.
Additionally, low awareness of policy and opportunities were identified as contributors to gender disparity in the KNUST, especially in STEM fields.We found that although the university had the STEM and WiSTEM policies in place, together with scholarship opportunities to help boost female enrollment in the institution, awareness about their existence was limited, particularly among the policy targets.There is little or no sensitization and awareness at the pre-tertiary and the tertiary level as well.This dearth of information is seen to have partly contributed to the persistently low female participation, despite the existence of such favorable policies and scholarships.We found that the institution had implemented the WiSTEM and STEM policies (aimed at increasing female enrollment in STEM) since 2010 and 2017, respectively.Notwithstanding, the impact of the policies has been mixed.For instance, female admission into STEM increased initially and then started to dwindle.Also, the policy increased female admission in the Health Sciences and Engineering but not in the Biosciences, Physical, and Computational Sciences.A similar trend was seen in new entrants (actual enrollment): female enrollment increased in Health Science and Engineering but not in the Sciences.These confirmed earlier findings on female enrollment in STEM in KNUST (Appiah-Castel et al., 2020).It remains to be known, however, why this has persistently been the case over the past decade.

Conclusion
The objective of this study was to assess the gender dimensions of student enrollment in higher education in Ghana, highlighting the disparities embedded therein.We also sought to understand the causes of the disparity within the Ghanaian context, and the effectiveness of gender policies implemented to address the disparities, using the KNUST as a typical case for our study.The study discovered a generally low female participation across all Colleges in KNUST except for the Health Sciences, across three dimensions: qualified applicants, admissions, and actual enrollment.The causes of disparity in the dimensions of student enrollment were rooted in four main factors: (a) institutional (admission selection process and mixed orientation structure of higher education), (b) socio-cultural (inadequate female role models, orientation, and culture of SHS schools toward STEM), (c) Economic (poor socio-economic background of students), and (d) awareness of policy and opportunities (inadequate sensitization of STEM policy and opportunities at pre-tertiary and tertiary levels).The study also discovered that although the KNUST had implemented the WiSTEM and STEM initiatives to augment female participation in STEM, the policy had been less successful at increasing female admission and enrollment into the Sciences, albeit with some progress in Engineering.
The findings of this study hold important implications for the present discourse on reducing gender disparity in higher education across the SSA region.In particular, we note the seeming disconnect between the existence of meaningful policies and scholarships, aimed at reducing these disparities, and the surprising lack of awareness of these opportunities among even the policy targets themselves-females.Our findings highlight the need to develop and implement targeted dissemination and sensitization plans in congruence with the policies and scholarships at the inception stage for such policy interventions.We find this action crucial since the STEM policy and its corresponding scholarship for females hold the capacity to address most of the bottlenecks to higher education identified in this study.Also importantly, we find it crucial that sensitization about STEM, the policy and scholarship opportunities are initiated at the pre-tertiary level.Additionally, the assignment of admission quotas to female-qualified applicants in STEM programs could serve as an extra motivation.
Furthermore, we deem it imperative that higher educational institutions would prioritize policy solutions that promote mentorship and role models, to increase females' participation in specific programs which hold good promise of returns to both female students and society.An example could be the promotion of conscientious female role models in STEM.It is of great importance that females who have excelled in academia and STEM-related careers make themselves known and available as role models to encourage female participation.Universities should also put in place deliberate capacity development programs to enhance the leadership skills of all staff, especially female faculty members and managers.These leaders can become important role models and an attraction for female students to the various programs.Additionally, higher educational institutions should develop gendersensitive curricula, especially in STEM, to help bridge the gender gap and improve the teaching and learning environment for female students.Such policy interventions hold the potential of creating more inclusive learning environments which promote gender equality.
Also, higher educational institutions should upscale their research on gender issues and increase enrollment for females through affirmative action.The institutions should ensure intersectionality by investigating the relationships and connections among different kinds of identities and how they influence experiences of privilege and oppression.The different experiences of female students with different intersecting identities should be theorized, monitored, and evaluated.Additionally, the institutions should promote critical pedagogy in teaching and learning, as this could enhance critical thinking and transformative change.Such an approach can help challenge gender norms and expectations and promote genderinclusive teaching practices and curricula.

Limitations of Study and Areas for Future Research
We note some limitations of our study.First, our choice of a single case research design for such a study may have limited the variety of our qualitative and quantitative insights on the subject, generated through the study.Our focus, nonetheless, was to leverage the advantages provided by single cases, such as depth and convenience of time and resources.This was especially important, because the study took place in the midst of the COVID-19 pandemic in Ghana, when preventive protocols limited in-person engagements and travel to a significant degree.Second, the choice of a top-class university like the KNUST for our case study is also not without concerns, given that top universities have much stricter admission requirements, which effectively exclude many prospective students.The location, disciplinary focus, popularity, and the existence of STEM policies, however, made KNUST the best-placed institution within the Ghanaian context for our type of study.Last, we were not able to examine why the STEM policy appears to have increased female enrollment in the Health Sciences more than the other branches of science.Future research could, therefore, investigate this question, in a large-n or comparative design to enable us to understand how widespread or otherwise the present finding is across different spatial and temporal dimensions.These limitations notwithstanding, we believe the present study is an important addition to the growing literature on gender disparity within higher education in sub-Saharan Africa.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
) colleges-College of Science (CoS), College of Engineering (CoE), College of Art and Built Environment (CABE), College of Health Sciences (CoHS), College of Agriculture and Natural Resources (CANR), and College of Humanities and Social Sciences (CoHSS).The university offers courses at the bachelors, masters, and doctoral levels.

Figure 5 .
Figure 5. Assessment of STEM policy performance on female admission and enrollment.Source.Authors' construct, 2021.
Figure 2. Growth rate (in percentages) for general admissions.

Table 1 .
Summary Total for All Colleges for Students' Participation at the KNUST.

Table 2 .
Gender Parity Analysis in Enrollment.Source.Authors' construct based on university's official admission data, 2021.
Assessment of STEM policy performance on female actual enrollment.