Self-selection from higher education: a meta-review of resources for academic decision-making of mainstream and underrepresented students

ABSTRACT

Self-selection refers to the decision of qualified students to not pursue their highest possible educational degree, including higher education. In this systematic meta-review, we used the conceptual framework of college choice to identify resources or mechanisms for students’ self-selection from higher education in international reviews and meta-analyses. In addition, we investigate whether underrepresented student groups (i.e., first-generation and cultural minority) experience unique resources and mechanisms for self-selection. Our narrative synthesis of international reviews and meta-analyses indicated that self-selection is related to financial, informational, social, aspirational resources in all contextual layers (i.e. individual habitus, school and community, higher education, socio-political context) of the conceptual framework of college choice. Whereas the family can help prevent self-selection by providing all four types of resources, school and higher education institutions can provide important information and social support through counselling and mentoring activities. For underrepresented groups, vertical and horizontal transmissions of social capital and experienced personal fit with the higher education environment may be of special importance for preventing self-selection. We conclude by discussing the potential of self-selection as concept for future interventions and research on widening access to higher education.

KEYWORDS:

• Self-selection
• higher education
• access
• enrolment
• socioeconomic status
• cultural minority

Introduction

Western European educational research has increasingly targeted individual and contextual conditions of students’ decision to (not) transition to higher education (Heath, Fuller, and Johnston 2010; van Herpen et al. 2017; Pinto, Lopes, and Mouraz 2019). In the Netherlands, for example, the phenomenon of ‘self-selection’ has received increased political attention, which describes the process when young people decide not to follow the highest possible educational path despite having the formal qualifications to do so (NRO 2019; van den Broek et al. 2019). In other words, students who self-select do not withdraw and were not rejected by higher education institutions, but choose different educational paths than the highest possible option. In the past decade, the number of self-selecting students has been growing (Inspectie van het Onderwijs 2020). This development may be problematic insomuch as it may indicate unequal educational opportunities and unequal access into higher education for students from so-called ‘underrepresented groups’.¹ These students who would be the first in their immediate family to attend higher education (i.e., first-generation students) and/or who are descendants of migrants (i.e., cultural minority students) are less likely to enrol in higher education than their peers (Perna 2006; Whitty, Hayton, and Tang 2015). As a consequence, self-selecting students from underrepresented groups may be less likely to reach their full academic potential, to gain the economic benefits of education such as reduced poverty, and to gain social benefits such as a better health and higher civic engagement (UNESCO 2014), and widening access for these structurally disadvantaged groups in (higher) education is a necessary step towards educational equity (UNESCO 2018). Thus, by investigating self-selection, this study aims to disentangle contextual, structural conditions and individual, academic decision-making processes at the transition to higher education.

International reviews have often examined higher education access and enrolment as indicators of academic success and achievement, among other indicators such as degree attainment, academic performance, and, as negative-examples, attrition or drop-out (e.g., Mishra 2020). Furthermore, many reviews have relied on social capital theory (Bourdieu 1986; Coleman 1988) with the common understanding that underrepresented groups in higher education do not have equal access to a variety of informational, financial and/or cultural resources (Bassett et al. 2019; Heath, Fuller, and Johnston 2010; Mishra 2020; Perna 2006; Whitty, Hayton, and Tang 2015). Lacking access to these resources may explain students’ self-selection (Bourdieu and Passeron 1990). However, when employing a strengths perspective (e.g., Bottrell 2009; Elizabeth and Ulanoff 2013), we see that successful students from underrepresented groups often access social support by the family as a complementary or compensating resource (Mishra 2020; Rezai et al. 2015). Next to the family, students’ academic decision-making is embedded in other contextual layers such as school and the community, the higher education environment or the broader national policy context (Perna 2006). Thus, in this meta-review, we use the conceptual framework of college choice (Perna 2006) to create a comprehensive empirical-based understanding of (1) which types of resources are relevant for qualified students’ self-selection from higher education and which contextual layers provide these types of resources, and (2) which unique resources are accessed by underrepresented students when deciding for or against higher education.

Theoretical framework

Social capital, cultural capital, and types of resources that influence self-selection

In her review on college access and choice, Perna (2006) notes that interindividual differences in academic decision-making have often been investigated by means of social capital theory (Bourdieu 1986; Coleman 1988). This sociological perspective emphasises the role of socioeconomic background for individual success, and of resources that lie in social networks and relationships. Coleman (1988) defined social capital as networks that take collective action to secure and access different types of resources. Next to access to financial resources, social capital in the form of information (e.g., about higher education) and social support (e.g., in the application process) may be the most important resources for academic decision-making after compulsory school (Perna 2006). The low availability of these resources is considered detrimental when an individual considers the costs and benefits of any given decision (Becker 1993), including the decision to self-select from higher education (Perna 2006).

Social capital is closely interrelated with cultural capital (Bottrell 2009). Cultural capital is often derived from parents and family (Coleman 1988) and encompasses the transmission of preferences, attitudes, habits and behaviours of their social class to their children (Bourdieu 1986). These attributes then manifest in youths’ language skills, cultural knowledge and mannerisms (Jæger 2011). High levels of cultural capital have been linked to high educational aspirations of urban youth (Cuervo, Chesters, and Aberdeen 2019) and academic achievement of students with high and low socioeconomic status (Jæger 2011). In contrast, young people who lack the required cultural capital may lower their educational aspirations, may self-select from certain situations (such as not entering higher education) because they do not know the necessary cultural norms, may overperform to compensate for their less-valued cultural resources, or may receive fewer rewards for their educational investment (Bourdieu and Passeron 1990; Lamont and Lareau 1988; Perna 2006). Based on social and cultural capital considerations, the current study will investigate the following resources as most important for the decision to (not) enrol in higher education: financial resources, informational resources, social support, and education-related values and aspirations.

Contextual layers and their resources

A successful transition into higher education depends on factors at the individual micro level all the way to the national macro level (De Clercq et al. 2021). Similarly, for studying academic decision-making at the transition to higher education, Perna (2006) proposes a conceptual framework of college choice with four contextual layers: (1) the individual habitus (including demographics and social and cultural capital provided by the family); (2) the school and community context; (3) the higher education context; and (4) the broader social, economic, and policy context. To better disentangle contextual conditions from individual decision-making (i.e., self-selection), we add to this framework by identifying which types of resources (i.e., financial, informational, social, aspirational) may be prevalent in each of the four contextual layers.

First, the individual habitus entails thoughts, beliefs and perceptions obtained in the immediate social environment (i.e., family) and is an important context for forming expectations, attitudes, and aspirations, including academic aspirations (Bourdieu and Passeron 1990). This contextual layer, and the family in particular, can be the main source for financial resources (i.e., family income), for information and for education-related values and aspirations when accessing higher education (Perna 2006; Mwangi 2015). Furthermore, the family frequently provides the social support, including emotional and motivational support, needed for academic success (Mishra 2020; Neves et al. 2019).

Second, the school and community context refers to social structures and resources that may influence the study choice of young people. For example, teachers and study advisors, as well as significant others in community centres and sports clubs, can provide access to information about studying and to networks that link parents and peers with higher education institutions (Engberg and Wolniak 2010; Stanton-Salazar 1997). However, structures within educational institutions can also make it more difficult for students from lower socio-economic environments to build a relationship of trust with lecturers and study advisors. Restrictive structures are, for example, bureaucratic processes and unsteady contacts with teachers and study advisors (Stanton-Salazar 1997).

Third, in the higher education context, the following aspects are likely to influence students’ academic decision-making: Institutional information, location, characteristics and competition (Perna 2006). Higher education institutions can be a source of information for students and their families about study options. This information can be conveyed actively through marketing and recruitment efforts (Simões and Maria Soares 2010) or passively, which includes the location of the institution or distance from the student’s home (McDonough, Lising Antonio, and Trent 1997). Importantly, regarding institutional characteristics, minority and mainstream students likely prefer an institution that they deem attractive regarding the location (van Herpen et al. 2017), and fitting regarding students’ personal and social identities and perceived comfort and acceptance (Nora 2004; Ball, Reay, and David 2002).

Finally, the broader social, economic, and policy context can influence study choice. Changes in the social context (such as demographic changes), the economic situation (such as the unemployment rate) and policy (such as a change in student finance system) can influence the study choice of young people directly and indirectly through other contextual layers (Perna 2006). Students may especially take the availability of financial resources into consideration when choosing for or against higher education (Perna 2006). In the Netherlands, for example, these financial resources come in the form of need-based grants (Dienst Uitvoering Onderwijs 2021). While these financial benefits can bridge a lack of financial means in the family in the short term, students’ may consider long-term costs (e.g., student debt) and benefits (e.g., a higher potential starting salary) before enrolling in higher education (Perna 2006).

Underrepresented groups in higher education

Parents and the family are often the main source for social support for all students, including mainstream, cultural and ethnic minority students at the transition into higher education (Mishra 2020; Pérez and McDonough 2008). Social support by parents may take shape in emotional support, such as being involved and showing interest in their child’s academic performance, and instrumental support, such as concrete assistance in the enrolment process (Rezai et al. 2015). Ethnic minority parents, who often have a lower socio-economic status than ethnic majority parents, further provide aspirational resources to their children, meaning that parents encourage their children to pursue their educational goals and dreams (Elizabeth and Ulanoff 2013). However, migrant parents or parents who have not themselves attended higher education may not always provide sufficient informational resources (Pascarella et al. 2004; Rezai et al. 2015). To capitalise the aspirations of their children, it is crucial that ethnic minority parents obtain and understand important information about school procedures and protocols (Elizabeth and Ulanoff 2013).

The school and community context, teachers further provide aspirational resources for ethnic minority students (Mishra 2020). Importantly, school, community and higher education institutions are more likely than parents to provide valuable informational resources for underrepresented students, such as about course work or adjusting to campus life (Mishra 2020). However, even if information is available, the access to this information (e.g., during orientation weeks) may strongly depend on marketing efforts and location of the higher education institution to reach student groups beyond the mainstream (Perna 2006) as well as a network that actively promotes university-related events and information (Mishra 2020).

In the higher education context, selection procedures or low availability of enrolment slots may influence study choice and may disproportionately affect underrepresented student groups (Perna 2006). If there is more demand for higher education slots than available, this can lead to competition for available places. Despite increased efforts in the US and Europe to broaden admission criteria for higher education (Niessen and Meijer 2017; Sinha et al. 2011), both rigorous, traditional selection procedures (e.g., GPA) and competition for available slots have often shown to disadvantage female, low-income, first-generation and cultural and ethnic minority students (Perna et al. 2005; Stegers-Jager et al. 2015).

In summary, to better understand why qualified students may decide to not enrol in their highest possible higher education degree (i.e., self-selection), the current meta-review aims to identify the range of available resources (i.e., financial, informational, social, aspirational) in different contextual layers (i.e., individual habitus, school and community, higher education, socio-political context) which are accessed by mainstream and underrepresented student groups at the transition to higher education. We critically examine existing reviews and meta-analyses on academic decision-making, access and enrolment in higher education to answer two research questions:

• RQ 1: Which types of resources and which contextual layers are accessed by qualified students during the decision to (not) enrol in higher education?

• RQ 2: Which types of resources are specifically accessed by qualified students who are part of underrepresented groups in higher education, namely first-generation and cultural minority students, compared to mainstream students?

Methods

To identify conditions for self-selection, we conducted a systematic meta-review of international research syntheses, including systematic literature reviews and meta-analyses. The literature search was guided by the PRISMA reporting guidelines (David et al. 2009), followed by a narrative synthesis (Popay et al. 2006). To reduce potential for bias during the selection and analysis of research syntheses, the first author consulted search terms, methods, and inclusion and exclusion criteria with a research team that consisted of three researchers with expertise on access into higher education and one trained research assistant.

Search strategy

In May 2020, the first author searched five electronic databases, including educational, psychological, and health databases (i.e., ERIC via Proquest, Web of Science, Scopus, PsychInfo (via OVID), and PubMed). The aim of our meta-review was to identify review studies and meta-analyses about students who were qualified for higher education and decided to not attend the highest possible level in higher education (i.e., self-selection). First, based on the expertise in the research team, we created an initial set of search terms related to self-selection such as self-select*, admission choice, college decision. Second, based on the initial set of search terms, we conducted an exploratory search to scope the topic top down in ERIC via ProQuest, ERIC Thesaurus via ProQuest, Web of Science and Scopus. We explored articles on title, abstract and key words for additional relevant search terms. We also checked if search terms resulted in very irrelevant articles and should be excluded from the search. For example, ‘college choice’ resulted in irrelevant articles about which college is chosen. Third, we conducted an exploratory search to scope the topic bottom-up via references in relevant reviews or meta-analyses. For example, ‘college readiness’ was added to the search term set. Fourth, we conducted another exploratory search to scope the topic (top down) with the new search terms. Fifth, after scoping the topic, we conducted a targeted search with a main aim to select papers for the main body. We combined the search terms set for self-selection, methodology and educational context to find most relevant publications (for the final search terms set, see Table 1).

Table 1. Search strategy.

Subject	Search terms
Method	Systematic review; comparative analysis; research review; meta$analy*; meta study; effect size
Education context	Higher education; post$compulsory; college; university; undergraduate; post$secondary; tertiary education
Self-selection	Self-select*; admission choice; college decision; study decision; admission decision; decision to attend; plan to attend; university aspiration*; educational aspiration*; college aspiration*; postsecondary planning; educational expectation*; college expectation*; college readiness; university readiness; readiness for; participation; access; admission; enrol$ment; barrier*; transition*

The final study selection was based on the following inclusion criteria: (a) peer-reviewed, (b) qualitative, quantitative, or mixed-methods research syntheses (i.e., literature review or meta-analysis), (c) published between 2009 and 2019 in English, (d) included at least one study with students who were eligible to apply for a Bachelor’s or equivalent level (i.e., level 6 in the International Standard Classification of Education; UNESCO 2012), meaning that they had completed upper secondary education or (vocational) post-secondary non-tertiary education, and (e) included at least one study on predictors of students’ academic decision-making after compulsory education.

Research syntheses were excluded if they were not peer-reviewed, were published before 2009 or after 2019, or if participants were not eligible to apply for a Bachelor’s or equivalent level. Research syntheses were also excluded if they had a sole focus on online, distance, or remote learning (incl. Massive Open Online Courses; MOOCs), on international students or access to transnational educational mobility programmes, on college choice (i.e., which university or college to choose), or on predictors of enrolment beyond the students’ own decision (e.g., admission test results).

Identification of relevant reviews

A summary of the systematic review process, from the first identification of 1077 studies through electronic databases and manual search to the final pool of studies included in the analysis is depicted in the flow diagram (Figure 1). After excluding duplicates and screening titles and abstracts, the first author completed the full-text screening of 78 studies. After discussing the preliminary selection of ten studies, the research team agreed to exclude two more studies that were comparative studies and not systematic literature reviews or meta-analyses. This led to a final selection of eight studies for analysis.

Figure 1. Summary of the systematic meta-review process.

Data analysis and critical appraisal

In preparation for the narrative synthesis (Popay et al. 2006), the first author extracted data from the final pool of studies into an Excel spreadsheet with the following criteria: Full reference, type of publication, study outcome(s), number and context of included studies, study population, geographical scope, method of data collection, main findings per contextual layer by Perna (2006), implications, and review quality. Study results were included in our narrative synthesis if they were assessed as trustworthy or generalisable by the respective study authors. If results were reported as ‘seldom’ (e.g., if only one out of 191 studies reported a particular finding; Mocca, Rojon, and Hernández 2019) or if the respective authors offered concerns about the quality of the extracted information (e.g., if certain findings were only detected in studies with the smallest sample size; See, Gorard, and Torgerson 2012), results were not included in our narrative synthesis.

To assess study quality, each study was critically appraised by the first author using the quality assessment tool for review articles (Health-evidence.org. 2010). Studies received a quality assessment rating of weak, moderate, or strong (see Table 2).

Table 2. Summary of eight included systematic reviews and meta-analyses.

Reference	Type of publication	Review outcome(s)	Number and context of studies	Study population	Geographical scope	Method of data collection	Review quality
Harvill et al. (2012)	Conference paper (peer-reviewed)	College readiness (i.e., maths achievement, language arts achievement, completed coursework, and high school graduation) and college enrolment	14 studies, including 12 school-based college access programmes	First generation, low-SES, and ethnic minority students between grades six and 12; total N not reported	US	Meta-analysis	Strong
Mocca, Rojon, and Hernández (2019)	Research article	Successful transition to higher education (incl. completion of a degree programme); access to higher education	191 studies on access to higher education	General student population; total N not reported	International; nearly half written by scholars based in the US, followed by Europe (mostly the United Kingdom and Germany) and South America (mostly Argentina, Mexico and Chile)	Systematic review	Strong
See, Gorard, and Torgerson (2012)	Research article	Post-16 participation (incl. post-secondary education enrolment)	14 studies and intervention programmes in secondary or higher education to increase post-16 participation	Ethnic minority students (English as first or additional language); total N not reported	US	Systematic review	Moderate
See et al. (2011)	Research article	Post-16 and higher education enrolment and participation	23 studies on post-16 and higher education participation	Ethnic minority students; total N not reported	UK	Systematic review	Moderate
Sneyers and De Witte (2018)	Research article	Student enrolment, retention, and graduation in higher education	25 studies on the effect of academic probation, student-faculty mentoring and need-based grants	General student population; total N not reported	US, Canada, and the Netherlands	Meta-analysis	Strong
Taggart (2018)	Research article	Academic achievement (e.g., grades, test scores, high school completion, college enrolment)	134 studies on Latina/o students’ academic achievement	Latina/o students in elementary and secondary education; total N not reported	US	Systematic review	Moderate
What Works Clearinghouse (2017)	Report	Students’ degree attainment (college), college access and enrolment, credit accumulation, completing high school, and general academic achievement (high school)	5 studies on dual enrolment programmes	77,249 students in high school	US	Systematic review	Moderate
Younger et al. (2019)	Research article	Participation in higher education (i.e., attendance at a higher education institution)	4 systematic review (results not included in this study); 12 studies and interventions for widening participation in higher education	Disadvantaged or under-represented groups in higher education (i.e., students from lower-income households, first-generation higher education, minority ethnic groups); total N not reported	UK and US	Systematic review	Strong

Post-16 = continuation of formal full-time education after compulsory schooling (age 16).

In the narrative synthesis, we have studied per study factors that facilitate or hinder young people to choose the highest possible education (see Mocca, Rojon, and Hernández 2019; See et al. 2011) and ordered the results according to the four contextual layers of our conceptual model as described in the introduction. Subsequently, research results related to the same factor or variable were bundled for each contextual layer. For example, in the contextual layer ‘habitus’, all research results on socio-economic status, migration background, first-generation student, gender, personal capacities and ambitions and social and cultural capital are bundled. In the contextual layer about the school and neighbourhood context, results have been bundled that relate to how experiences with peers and teachers/school staff influence the choice of whether or not to enrol, what the influence of support programmes is on school and neighbourhood level and the influence of advice from secondary and higher education to young people on whether or not to opt for the highest possible education. In the context of higher education, results have been collected about the influence of transition programmes and support from higher education institutions in the choice of study. Finally, in the broad social, economic and policy context, results have been bundled about the influence of scholarships, student loan systems and perceived job opportunities. The narrative synthesis was carried out in consultation and discussion with the entire research team.

Results

Search results

After following the steps of identification, screening, and eligibility, a total of eight international research syntheses were included in the narrative synthesis (see Table 2, for a summary). The final selection comprised systematic reviews (N = 6) and meta-analyses (N = 2), published between 2011 and 2019. Each research synthesis included between five and 191 studies and most research syntheses addressed access or enrolment of the general student population (N = 4), of ethnic minority student populations (N = 4) or students with low socioeconomic status (N = 1). Most research syntheses included study populations exclusively from the US (N = 5), whereas the remaining syntheses included a combination of studies from Europe (e.g., the Netherlands and Germany; N = 2), the UK (N = 2), South America (e.g., Argentina, Mexico and Chile; N = 1) and Canada (N = 1).

Narrative synthesis

To answer our first research question about types of resources in contextual layers that may be relevant for students’ self-selection from higher education, we present results in order of the four contextual layers by Perna (2006): (1) the individual habitus; (2) the school and community context; (3) the higher education context; and (4) the broader social, economic, and policy context. To answer the second research question about resources and contexts that are unique to underrepresented student groups, we synthesise results for students with cultural or ethnic minority status and first-generation students in higher education.

Research question 1: contextual layers and resources for students’ academic decision-making

The individual habitus includes demographics and social and cultural capital provided by the family (Perna 2006). Regarding the family’s socioeconomic status, having highly educated parents and an affluent family household made it more likely that mainstream and ethnic minority students decided to enrol in higher education and thus not to self-select (Mocca, Rojon, and Hernández 2019; See et al. 2011; Taggart 2018). Studies identified access to financial resources, to information and to social support (i.e., motivation, skills and knowledge) regarding the higher education environment by parents as possible explanations (Mocca, Rojon, and Hernández 2019; See et al. 2011).

Regarding the school and community context, a lack of social support from schools in the form of guidance, counselling or advice regarding higher education may discourage students from entering higher education (Mocca, Rojon, and Hernández 2019), and may thus facilitate self-selection. In contrast, there is some evidence from the US that allowing high school students to take college courses and earn college credits while still attending high school (i.e., dual enrolment programmes) may be positively related to students’ decision to enrol in higher education (What Works Clearinghouse 2017).

Regarding the higher education context, higher education institutions may encourage students’ enrolment if they include mentoring and counselling activities in transition programmes targeting secondary school students (Mocca, Rojon, and Hernández 2019; Younger et al. 2019). For example, college-transition counsellors who pro-actively discuss and give practical help to low-income students’ facing financial aid application barriers and emotional and aspirational barriers during the summer months leads to substantial improvements in both the rate and quality of college enrolment (Castleman, Arnold, and Lynk Wartman 2012, in Younger et al. 2019). In contrast, students’ enrolment in a higher education institution becomes less likely with missing or inaccessible information (including regarding admission procedures, the availability of grants and financial aids, the structure and type of academic curricula) and missing social support (including counselling or advice opportunities, preparatory courses, motivational support by academic staff; Mocca, Rojon, and Hernández 2019).

Once enrolled in higher education, high perceived social support by peers, access to cultural capital (incl. academic knowledge and skills) in extra-curricular activities and networks inside and outside of higher education, or the availability of financial aids may facilitate a positive transition experience into higher education (Mocca, Rojon, and Hernández 2019). In contrast, students may perceive the transition into higher education more negatively if a higher education institution employs cumbersome bureaucracy, outdated or inadequate teaching and learning methods for a diverse student population, underrepresentation of minority students among students and staff, or no availability of grants (Mocca, Rojon, and Hernández 2019). While these latter findings are not specific to self-selection, a lack of social support by peers and the higher education institution, appropriate cultural capital and financial aids are all considered important factors in a students’ cost-benefit consideration on the verge of higher education (Perna 2006).

In the broader social, economic, and policy context, the availability of need-based grants made it more likely for qualified students to enrol in higher education in the US (Sneyers and De Witte 2018).

In sum, students whose parents are highly educated and students in affluent family households are more likely to decide to enrol in higher education and thus not self-select. Furthermore, a lack of social support from schools and of valuable information from higher education institutions can lead to qualified students’ self-selection. If higher education institutions offer social and informational support through mentoring programmes as well as financial aids, qualified students may be more likely to enrol in higher education and thus to not self-select.

Research question 2: unique resources for underrepresented student groups

On the individual habitus level, See et al. (2011) emphasise that ethnic or cultural minority status per se may neither predict self-selection nor enrolment into higher education in the UK. In contrast, low family socioeconomic status (SES) and low educated parents were stronger predictors of self-selection for ethnic minority students in the US and UK (See et al. 2011; Taggart 2018) as well as for the general student population in the US, Europe and South America (Mocca, Rojon, and Hernández 2019).

Still, we cannot assume that a low family SES is always linked to self-selection when we look at family-level resources for underrepresented student groups. For ethnic minority students in the US and UK, family academic values and aspirations were identified as important for their higher education enrolment decision (See et al. 2011; Taggart 2018). This could be manifested in high family pressures to succeed, in trying to meet high educational expectations by parents but also in having highly educated family members (e.g., siblings) as role models or a friend group that aspires to attend higher education. Students also found motivation to enrol in higher education in the prospect to improve their own or their family’s life chances (See et al. 2011). In contrast, low parental value of education or if family members objected to students’ academic decision-making was related to self-selection of ethnic minority students in the UK (See et al. 2011). Importantly, low finances were not a big disincentive for going to higher education for African Caribbean postgraduate students, unlike for white students in the UK (See et al. 2011).

See et al. (2011) identified several gendered experiences of ethnic minority students in the UK at the transition into higher education. Female Asian students were more likely to enrol in higher education if they expected that it would have an added value for their personal development, private life (marriage prospects, motherhood) and/or career (career prospects, independence, prestige). Especially for British South Asian Muslim women, parents were keen to maximise their daughter’s social prestige by encouraging them to pursue higher education. In contrast, South Asian parents were more likely to encourage their sons to follow vocational instead of academic degrees to ensure their family’s future. Furthermore, Pakistani or Bangladeshi boys perceived higher pressures than girls to provide for extended family and to ensure upwards social mobility when deciding for post-compulsory education.

School and community programmes focusing on student success have been found to make enrolment more likely for ethnic minority students in the UK and US (See et al. 2011; Taggart 2018). Findings for the effectiveness of school career services are mixed and may differ by minority groups (e.g., services perceived as more helpful by Black students compared to other ethnic groups; See et al. 2011). Furthermore, school-based interventions that include a variety of targeted support opportunities (i.e., ‘black box’ interventions that offer, e.g., academic enrichment and counselling) may increase the enrolment numbers of underrepresented students in higher education (Harvill et al. 2012). In contrast, negative school experiences and the underrepresentation of ethnic minority or low SES students in educational routes towards higher education may make self-selection more likely (See et al. 2011).

Higher education institutions may increase enrolment numbers of underrepresented students if they offer a variety of counselling and information opportunities, including career and academic information and support, comprehensive college preparation or tailored financial advice (Younger et al. 2019). Having non-parental adults (i.e., staff or faculty members) as mentors is related to higher academic performance throughout higher education, including lower drop-out rates and lower retention of ethnic minority students in the US (See, Gorard, and Torgerson 2012). However, it remains unclear whether mentoring can directly impact underrepresented students’ decision to enrol in higher education.

Regarding the broader social, economic, and policy context, studies from the UK and the US have highlighted the importance of financial incentives for enrolment of underrepresented groups (Younger et al. 2019). Besides receiving financial scholarships, being eligible for tuition loans per se may already positively affect students’ decision to enrol in higher education (Younger et al. 2019).

In sum, previous research syntheses suggest that ethnic or cultural minority status per se may not be sufficient to explain qualified students’ self-selection from higher education. Instead, a low family SES and low educated parents are associated with self-selection from higher education for all students. Importantly, high family academic values and aspirations seem to buffer the negative association between low family SES and low educated parents and higher education enrolment for ethnic minority students. Comparable to mainstream students, the enrolment of underrepresented students in higher education may be stimulated via school and community programmes focused on student success and higher education institutions providing (academic) information and social support. In contrast to mainstream students, financial incentives through national policies and practices may be a strong determinant for underrepresented students’ enrolment in higher education.

Discussion

This study focused on understanding the contextual conditions and individual academic decision-making process behind students’ self-selection, meaning the decision of qualified students to not pursue their highest possible educational degree. In a meta-review we used the conceptual framework of college choice (Perna 2006) to identify resources in different contextual layers that influence self-selection in general (RQ 1), as well as uniquely for underrepresented student groups in higher education (RQ 2).

Our results led to three main conclusions, which we discuss in the upcoming section. First, self-selection is a nested cost-benefit consideration on fitting in the highest possible educational degree. The decision to not enrol in the highest possible educational degree in higher education is influenced by resources in and mechanisms between contextual layers apparent in students’ life. Particular resources (e.g., social support, information) can be accessed through multiple contextual layers (e.g., family, school, higher education). Second, for underrepresented student groups in higher education social capital can be transmitted vertically and horizontally through the contextual layers. Third, students’ self-selection may be strongly based on the experienced fit between resources in contextual layers (e.g. family, higher education institutions) and underrepresented student’s own preferences. We conclude by discussing self-selection as a promising concept for future research to disentangle structural inequalities in different contextual layers from individual academic decision-making during the transition into higher education. To mitigate the occurrence of self-selection and increase equity in access to higher education, interventions should focus on an integrated approach that impacts several contextual layers of young adolescents, especially for underrepresented student groups. Below, we explain our conclusions in detail.

First, as proposed in social capital theory (Coleman 1988; Bourdieu 1986), the availability of different types of social and cultural capital found in parents as well as broader social networks were a resource for deciding to attend higher education and thus not self-select. The family has the potential to provide a variety of resources, including financial, informational, social, and aspirational resources (e.g., Mocca, Rojon, and Hernández 2019). In addition, schools and higher education institutions can be important sources for information (e.g., through financial and academic counselling) and for social support (e.g., networking before higher education starts) for mainstream and underrepresented groups. Thus, a strong connection between schools, community, and higher education with families may ensure students’ access to a wide range of resources for academic decision-making. For example, aspirational support by parents depends on parents’ ability to obtain and understand important information about school procedures and protocols, which can be provided by workshops for parents in schools (Elizabeth and Ulanoff 2013). Future research may focus on connections between social capital within contextual layers (e.g., peer and faculty mentoring in higher education; See, Gorard, and Torgerson 2012) and across contextual layers (e.g., workshops for parents in schools) and how they affect students’ self-selection.

Second, for underrepresented groups in higher education, family socioeconomic status (SES) and the availability of social capital, including motivational and aspirational resources affect underrepresented students’ self-selection from higher education (See et al. 2011; Taggart 2018). Similar to research that focuses on educational successes of underrepresented groups we identified that social and cultural capital may not only be transmitted downwards, from support networks to students, but also horizontally among students and other network members (Heath, Fuller, and Johnston 2010). For example, educational experience and successes of individual students can motivate and shape perceptions of other same-aged or older community members (Heath, Fuller, and Johnston 2010; Elizabeth and Ulanoff 2013). Thus, taking a strengths perspective and identifying social capital outside of the family (e.g., through peer-to-peer mentoring, staff-student mentoring, and extra-curricular activities) may be a promising direction for future interventions to reduce self-selection and to widen access to higher education for a diverse student population.

Based on our results, we further conclude that underrepresented students’ perceptions of fitting-in with social and academic values and expectations in different contextual layers (family, school and community, higher education) may facilitate the decision to enrol in higher education. For example, having highly educated family members as role models or being among like-minded peers with high academic aspirations can facilitate the transition into higher education for mainstream and underrepresented students (Mocca, Rojon, and Hernández 2019; See et al. 2011; Taggart 2018). In contrast, mostly for underrepresented students, a lack of social support or representation in the routes towards higher education may foster the impression of not fitting-in (See et al. 2011). Importantly, the first contact with the higher education institution (e.g., through preparation programmes, counselling, mentoring) may be a key moment for building social connections and networks and for fostering feelings of belonging in higher education. Thus, transition programmes and interventions that strengthen perceptions of a good student-environment fit (see van Herpen et al. 2020, for an example) may be particularly beneficial for students from underrepresented groups in higher education, including cultural minority and first generation students.

Limitations

Despite having synthesised important research findings for self-selection, we wish to acknowledge some study limitations. First, reviews and results were mainly based on US- and UK-based research. We acknowledge that educational experiences of underrepresented groups are unique to national, local, and community conditions (e.g., Cuervo, Chesters, and Aberdeen 2019) and that our findings need to be enriched by knowledge from regions and contexts outside the US and UK. Second, the method of a meta-review allowed for a synthesis of existing findings regarding academic decision-making but did not allow for an in-depth exploration of the complexity and diversity of students’ transition experiences and thought processes, which could be better addressed through person-centred or qualitative approaches (Tett, Cree, and Christie 2017; Hope 2017; Holmegaard, Møller Madsen, and Ulriksen 2017). Third, none of the included studies in this review has explicitly focused on self-selection but rather on various indicators of academic success before, during and after the transition into higher education. Thus, we encourage future studies to focus on the mechanisms behind self-selection (including resources and contextual layers) and under which circumstances self-selection may become problematic. Fourth, gendered experiences in underrepresented student groups emphasise the need for future educational research to move beyond established dichotomous categories (e.g., cultural minority, gender) and to focus on populations that share several marginalised identities (Vietze et al. 2021). By employing the paradigm of intersectionality (e.g., Bowleg and Bauer 2016), future research may better illuminate processes that drive self-selection and, as a result, maintain existing social-structural educational inequalities and the underrepresentation of marginalised groups in higher education. Finally, by focusing on the transition to higher education, we did not grasp the extent to which selection already happens at an earlier timepoint in the life-course (Gorard and Smith 2007) and particularly in the school career (e.g., by early tracking; de Werfhorst and Herman 2019). In the future, we wish to encourage researchers to investigate (self-) selection at different educational transition moments (e.g., from primary to secondary school) and the contexts that influence this.

Conclusion

In conclusion, self-selection is a promising concept for studying academic decision-making of diverse students at the transition to higher education. This meta-review gives a more comprehensive understanding of types of resources provided in different contextual layers (Perna 2006) that may influence self-selection from higher education for diverse groups of students. Important resources lie in the family context and in educational institutions, like financial resources, informational resources, social support and education-related values and aspirations. Importantly, for students to have equal opportunities to choose the education that is most appropriate for their future career, students need access to the necessary social and cultural capital in family and school, to a higher education context with effective educational guidance, essential social support and information, within a policy context that aims to reduce structural barriers.

Disclosure statement

No potential competing interest was reported by the authors.

Additional information

Funding

This research project was funded by the Netherlands Initiative for Education Research (NRO) under Grant 405-00-860-009.

Notes on contributors

Jana Vietze

Jana Vietze is an Assistant Professor at the Department of Psychology, Education and Child Studies of the Erasmus University Rotterdam, the Netherlands. Her research interests include the belonging, inclusion, well-being, and academic trajectories of cultural minority and first-generation students in secondary and higher education.

Sanne G. A. van Herpen

Sanne G. A. van Herpen is a Senior Researcher at the Research Group of Student Success of Inholland University of Applied Sciences, Haarlem, the Netherlands. Her research interests include accessibility of higher education, students’ motivation, self-regulated learning, well-being and academic success.

Aike Dias-Broens

Aike Dias-Broens is a Ph.D. candidate at the Department of Psychology, Education and Child Studies of the Erasmus University Rotterdam, the Netherlands. Her research interests include belonging and inclusion of students in higher education.

Sabine E. Severiens

Sabine E. Severiens is a Full Professor at the Department of Psychology, Education and Child Studies of the Erasmus University Rotterdam, the Netherlands. Her research interests include diversity and educational inequality, from the perspective of the teacher (culturally responsive teaching) as well as the student (motivation, integration and the learning environment).

Marieke Meeuwisse

Marieke Meeuwisse is a Associate Professor at the Department of Psychology, Education and Child Studies of the Erasmus University Rotterdam, the Netherlands. Her research interests include diversity, belongingness, inclusion, academic success, peer interaction and student-staff interaction in higher education.

Notes

1. In this study, we use ‘underrepresented groups’ an umbrella term to include first-generation and ethnic or cultural minority students, as is common in studies on higher education enrolment and access (e.g., Mishra 2020; Perna 2006). We acknowledge that many other societal groups are structurally disadvantaged and underrepresented in higher education (e.g., students with special needs; Lindsay et al. 2019). However, including groups beyond first-generation and ethnic or cultural minority students was beyond the scope of this study.