Between-school inequalities in access to science and mathematics curricula in a marketized education system: The case of Australia

Abstract

Increasing the participation of students who study science and mathematics curricula is a priority in many countries, but between-school inequalities in the offerings of these subjects is not well understood. We examine stratified opportunities to learn science and math subjects in upper secondary schools in Australia, as a case study for examining how educational marketization reduces access to curricular subjects in comprehensive secondary education systems. Using census data from one state, we found biology and chemistry are offered in most schools, but substantial inequalities exist in access to physics and especially advanced mathematics. School size, socioeconomic composition, sector and location predict whether a school offers advanced mathematics. The findings suggest that inequalities in access to science and mathematics curricula are patterned by social background and that these inequalities are linked with educational marketization.

Keywords

upper secondary school opportunity to learn science curricula mathematics curricula school characteristics school stratification marketization educational inequalities

Introduction

Many national governments are seeking to increase the number of high school students that study science and mathematics as part of larger strategies to promote innovation and sustainable development (e.g. Committee on STEM Education, 2018; Education Council – Australia, 2018; House of Commons Science and Technology Committee, 2017). Science and mathematics subjects are promoted as important pathways for further study and stable and rewarding employment (Australian Industry Group, 2015; Morgan and Kirby, 2016). Regardless of educational or career aspirations, a solid foundation in maths and science is important for all citizens (Goos and Kaya, 2020). Thus, mathematics and science skills and knowledge provide important benefits for both individuals and the larger society. Our basic premise, therefore, is that all students should have an opportunity to learn science and mathematics subjects, regardless of where they go to school.

While improving students’ attitudes towards these subjects is important to improving participation (Kennedy et al., 2020), we agree with Smyth and Hannan (2006) that it is also crucial to identify systemic barriers that may be preventing students from enrolling in science and mathematics subjects. It is unrealistic to expect large numbers of secondary students to study physics and advanced mathematics,¹ for example, if these subjects are not widely offered in schools. The degree and magnitude of between-school inequalities in access to science and mathematics curricula in comprehensive secondary school systems is not well understood, however. These between-school inequalities have not been examined extensively in any country, nor have they been examined substantially across a range of national contexts.

Our study addresses this gap in the literature by examining unequal opportunities to learn science and mathematics curriculum in upper secondary schools in Australia. Our larger aim is to contribute to the empirical and theoretical literature about the nature and extent of between-school inequalities in access to science and math curriculum in comprehensive secondary education systems, and the reasons that drive these between-school inequalities.

In this study, we examine access to science and mathematics subjects at upper secondary level in all schools in one Australian state. Our research questions are as follows:

• RQ1: What proportions of secondary schools offer advanced mathematics, physics, chemistry and biology, and do they vary by school socio-economic composition, school sector and school location?

• RQ2: Which school characteristics predict access to the four curricula subjects?

Examining opportunities to study mathematics and science subjects in Australian secondary schools can contribute to the development of a comparative theoretical framework about the contexts and conditions that mediate access to these subjects in comprehensive education systems. The features of Australian schooling that may be theoretically relevant are as follows. It is a prosperous country with a high level of social and economic development, without the high levels of poverty and inequality that Berliner (2014) has shown are associated with diminished educational opportunities and outcomes. Its labour market consistently suffers shortages of professions in the STEM fields (National Skills Commission, 2022). It has a comprehensive education system (i.e. a system in which most schools offer both academic and vocational curricula and do not select students based on prior academic performance), which theoretically at least has the capacity to offer core academic curricula to an unlimited number of students. These features suggest that Australia has the capacity and demand to support science and mathematics curricular offerings in all secondary schools. However, Australian schooling is not immune to structural inequality. Large funding disparities exist between many public and private schools and schools are highly segregated by socioeconomic status, accompanied by large between-school inequalities in human, financial and instructional resources (OECD, 2019b). These inequalities are associated with high levels of privatization and marketization (Perry, 2024; Perry et al., 2024a).

This article is structured as follows. First, we elaborate our theoretical lenses that underpin our study, namely, educational marketization and opportunity to learn. These lenses allow us to contribute to broader understandings about the impact of educational marketization on curricular inequalities. We then review the empirical literature about opportunity to learn mathematics and science subjects in high school. We then describe the Australian educational context so that our findings can be interpreted for an international audience and, even more importantly, can be used to develop theory about the contexts, structures and policies that shape access to science and mathematics curricula. We describe our methods and present the findings, and then discuss the findings and their contribution to a larger comparative theoretical framework about access to science and mathematics curricula in comprehensive education systems.

Theoretical perspectives

Out study is informed by two theoretical lenses: opportunity to learn and educational marketization.

Opportunity to learn (OTL) has existed as a construct in educational research since the 1960s. It was originally operationalized as the amount of time in a classroom spent on learning a particular subject. Over the time the concept has broadened to include a wider range of opportunities, processes, structures and resources (Perry et al., 2024b), which are provided not just by teachers in classrooms but also in and by schools, families and the larger community (Alexander et al., 2012). It is widely considered to be one of the most important predictors of student achievement (National Research Council, 2001; Schmidt et al., 2015). Given its primary in shaping student outcomes, inequalities in opportunities to learn based on students’ socioeconomic or cultural backgrounds are inadmissible (Ladson-Billings, 2006; Lafontaine et al., 2015). As such, we agree with Darling-Hammond et al. (2016) that education authorities need to be held accountable to the opportunities to learn and resources that they provide students. From an education systems perspective, opportunities to study science and mathematics subjects in secondary schools is a crucial dimension of this accountability, especially if these curricular opportunities are inequitably linked with students’ social background.

Our second theoretical lens is educational marketization, which is the process of applying market principles into the organization and management of schooling. The two core dimensions of educational marketization are choice and competition (Chubb and Moe, 1990), facilitated by the processes of decentralization, school autonomy and accountability (Whitty and Power, 2000). The rationale for educational marketization is to incentivize innovation, efficiency and effectiveness (Whitty and Power, 2000). However, educational marketization often does not increase effectiveness or innovation (Jabbar, 2015; Lubienski, 2006) and has even been linked with increased educational inequalities in many countries (Carter et al., 2013; Fjellman and Hansen, 2025; Lubienski et al., 2022; OECD, 2019; Yang Hansen et al., 2025; Zancajo and Bonal, 2022).

Combining these two theoretical lenses enables us to see how marketization reduces opportunities to learn, which in the context of this study relates to between-school inequalities in provision of science and mathematics curriculum. Marketization has been found to increase segregation based on socioeconomic status between schools in many countries, including Australia (Chesters, 2019), Chile (Aguirre, 2020), England (Greaves, 2023), Sweden (Fjellman, 2019; Fjellman and Hansen, 2025; Yang Hansen et al., 2025) and cross-nationally (Alegre and Ferrer, 2010; OCED, 2019a). School segregation undermines educational equity and effectiveness because in most cases it is linked with an inequitable distribution of human and material resources (Benito et al., 2014; Bonal and Bellei, 2019). Socially advantaged schools typically enjoy greater opportunities and resources, which compounds the advantages already enjoyed by students from privileged backgrounds. The inequitable allocation of resources between socially segregated schools is facilitated by funding mechanisms, including school funding formulas that provide greater resources to private schools (Kenway et al., 2024) and per-pupil funding allocations that disadvantage small, residualized schools (Preston, 2025). These funding inequities make it harder for socially disadvantaged schools to offer many curriculum subjects (Perry and Lubienski, 2020), which is further compounded by socially disadvantaged schools’ limited capacity to recruit and retain specialist teachers (Blackmore et al., 2024).

Participation rates and opportunity to learn science and mathematics subjects

Researchers and policymakers around the world have been concerned about low and declining participation rates of secondary students in science and mathematics subjects. These studies come from a range of national contexts, including Australia (Cooper and Berry, 2020), the UK (Smith, 2011) and the US (National Science Board, 2018). As noted by Marginson et al. (2013), increasing the number of students who study science and mathematics subjects in secondary school is a government priority in most OECD countries.

Student participation rates in science and advanced mathematics have been examined in several countries. In Australia, 52% of all students in the final year of secondary school study a science subject (Goodrum et al., 2012) and 66% study a mathematics subject (Wienk, 2022). Approximately one half of the students who are studying mathematics are enrolled in a basic course, 18% are enrolled in an intermediate course and 9% are enrolled in an advanced course (Wienk, 2022). These participation rates in advanced mathematics are lower than in some other countries. For example, the proportion of secondary students who study advanced mathematics is 14% in Finland (Marginson et al., 2013) and 19% in the US (National Science Board, 2018).

Secondary school student participation rates in science and mathematics subjects are patterned by ascriptive individual characteristics such as gender, race/ethnicity and socioeconomic status. Girls are less likely than boys to study advanced mathematics and physics (Justman and Mendez, 2018). Similarly, students from lower socio-economic backgrounds are less likely than their more advantaged peers to study advanced mathematics and the natural sciences. This has been found in many (if not most) countries around the world, including Australia (Cooper and Berry, 2020), England (Archer et al., 2017) and the US (Svoboda et al., 2016).

The factors that shape student participation in science and mathematics offerings can be divided into two categories: factors that relate to individual students and their motivations and interests, and structural factors that relate to schools and education systems. Most of the literature falls into the first category. The research literature about students’ interest and motivation to study science and mathematics subjects is extensive. For example, Wang and Degol (2013) reviewed more than 300 studies, the great majority of which concern student motivation and interest in science and mathematics subjects.

The main structural factor identified in the literature that shapes opportunities to learn science and mathematics subjects is curricular differentiation and tracking practices within schools. Studies from the past 30 years and from a range of national contexts including Australia (Lamb and Fullarton, 2002), the UK (Archer et al., 2017), the US (Eisenhart et al., 2015) and cross-nationally (Chmielewski, 2014) have consistently shown that in schools with tracked mathematics and science offerings, students from lower socioeconomic backgrounds are less likely than their more advantaged peers to be enrolled in the most rigorous courses, such as calculus and physics.

Another school-related factor that can influence participation rates is course availability. Students cannot study advanced mathematics or physics, for example, if it is not offered at their school. Somewhat surprisingly, this factor has not received much attention in the literature. Many studies have documented unequal participation rates by student socioeconomic status, rurality and ethnicity/race and low overall participation rates more generally, but few have examined the proportions of students that attend a school that offers these courses. The little research that has been conducted about between-school stratification of opportunities to learn science and mathematics subjects, however, is sobering. Students from low SES backgrounds are less likely to attend schools that offer advanced mathematics and sciences courses in the US (Adelman, 2006), Iceland (Jónasdóttir et al., 2025), Ireland (Smyth and Hannan, 2006) and the UK (Gill, 2015), as are students in rural communities in the US (Irvin et al., 2017; Saw and Agger, 2021) and Sweden (Fjellman et al., 2019). Educationally underserved racial/ethnic minorities in the US, especially Black and Latino students, are also less likely to attend a school that offers advanced coursework, including STEM subjects, compared to their white and Asian peers (Civil Rights Data Collection, 2018; Patrick et al., 2020). More generally, secondary schools that enrol larger proportions of socially advantaged students offer larger amount of subject offerings, as has been found in the US (Kolluri, 2018; Monk and Haller, 1993; Rumberger and Thomas, 2000) and the UK (Dilnot, 2018; New Schools Network, 2016).

In Australia, between-school curricular inequalities have been identified in upper secondary schools. Between-school inequalities in the number of academic courses of study have been identified in one state (Dean et al., 2023) and in one capital city (Perry and Southwell, 2014). Both studies found that low SES schools offer substantially fewer subjects than other schools, as do rural schools (Dean et al., 2023). In terms of stratified access to science and mathematics subjects, the subject of this paper, both low SES and non-metropolitan public schools are less likely to offer these subjects compared to other public schools (Dean et al., 2023; Murphy, 2019; Murphy, 2020). To date, no study has been conducted in Australia that examines between-school inequalities in upper secondary mathematics and science offerings in all schools (both public and private). This is an important gap to address as privatization and marketization have been linked with educational inequalities (Perry et al., 2024a).

The Australian educational context

Australia’s national curriculum mandates the content and achievement standards for upper secondary school learning areas, including four upper-secondary science offerings (Chemistry, Physics, Biology and Earth/Environmental Science) and four mathematics offerings of varying rigour. Specialist Mathematics, the most advanced mathematics offering, ‘contains topics in functions, calculus, probability and statistics … Specialist mathematics is designed for students with a strong interest in mathematics, including those intending to study mathematics, statistics, all sciences and associated fields, economics or engineering at university’ (ACARA, n.d).

In the final 2 years of secondary school, students choose from the range of curricular subjects that are offered at their school. Students who aspire to university education study four to six externally assessed academic subjects. Students are not required to select a mathematics or science subject. The number of externally assessed subjects that are available to be offered in schools is very large, ranging from 80 to 125 subjects (depending on the jurisdiction). Unsurprisingly, no school offers all courses of study. On average, schools offer 10–25 externally assessed subjects, with the number of offered academic courses of study positively related to the size of the school and its socioeconomic composition (Dean et al., 2023; Perry and Southwell, 2014). Schools choose which subjects to offer based on student demand, parent expectations and school resources, needs and profiles (Perry and Lubienski, 2020).

Finally, we provide an overview of the organization and structure of Australian schooling. As schools with higher socioeconomic compositions usually offer a larger number of academic subjects compared to schools with less advantaged student populations (Dean et al., 2023; Perry and Southwell, 2014), we discuss the features of Australian schooling that are related to school socioeconomic composition, namely, school sector, school choice and competition and school funding.

Australia has one of the largest private school sectors in the world, with 41% of all secondary school students attending a private school (ABS, 2020). The private sector comprises both parochial and non-sectarian schools and is divided into two categories: Catholic schools and independent schools (a few of which are also Catholic). The socioeconomic composition of schools is stratified by sector, with public schools having on average the lowest socioeconomic compositions and independent schools the highest (Connors and McMorrow, 2015). Private schools charge fees and also receive government subsidies, with fees ranging from a few thousand Australian dollars per year to $40,000 or more. Because of this funding model, most private schools have more resources to spend per pupil than do public schools (Kenway and Boden, 2025; Preston, 2025). Overall, Australian schooling is characterized by high levels of choice and competition, with schools competing for students within and between the three school sectors and many students attending a private or non-local public school (Campbell et al., 2009). Public schools have catchment zones but may also enrol students, at their discretion, from other localities.

In summary, Australian schooling has some unique characteristics that may shape between-school inequalities in opportunities to learn science and mathematics subjects. These characteristics are summarized in Table 1.

Table 1.

Features of Australian schooling that may impact between-school inequalities in curricular access.

Curriculum and assessment	Organization and funding
• National curriculum standards with state interpretation and implementation • Subject-specific, standardized exams that are externally assessed and provide the basis for school leaving certificates and university admissions • Schools choose which externally assessed curricular subjects to offer	• High levels of school choice and competition • High levels of school social segregation • Large proportion of students that attend fee-charging private schools • Substantial funding inequalities between schools

From a comparative perspective, the Australian case is theoretically interesting because of its high level of marketization, in particular school choice and competition, but also school autonomy in terms of curriculum offerings. Schools select their curriculum offerings as a way to compete for students and to establish positional advantage in the education marketplace (Perry and Lubienski, 2020). This strategic decision-making by schools is further shaped by school funding policies that either enable or restrict curriculum offerings, and decentralized decision-making that does not mandate, beyond the absolute basics, what curriculum subjects must be offered. A similar set of school choice and school decentralization policies is occurring in Icelandic upper secondary schooling, with subsequent increase in between-school curriculum inequalities (Jónasdóttir et al., 2025). Whether these dynamics and outcomes are occurring in other countries with comprehensive secondary education systems is not well understood. As has been argued already, few studies have examined between-school inequalities in the provision of science and maths curriculum subjects in any country, and an even fewer number of studies have examined these trends with a marketization lens.

Method

We examined the extent to which four science and mathematics subjects (Chemistry, Biology, Physics and Specialist Mathematics) are offered in the final year (Year 12) of all secondary schools in one Australian state (Victoria). As described earlier, Specialist Mathematics is the most advanced offering and is therefore comparable to advanced mathematics courses in other countries. We chose these four subjects because they are considered ‘core’ science and mathematics curricular subjects (Archer et al., 2017); they provide pathways to high-status professions including medicine and engineering; and except for Biology, they receive a favourable scaling in Australia’s university admissions rank score, which can therefore influence access to prestigious universities and courses of study (majors).

Participants and variables

The unit of analysis in our study is secondary schools. Specifically, all schools in the state of Victoria that enrol students in the final year of secondary schooling were included in our study. By including all schools in one state, we can consider how curricular inequalities are patterned by school characteristics such as location and sector. This is a major contribution to the literature as previous studies from Australia have examined curricular stratification within the government sector only (e.g., Murphy, 2020) or within one capital city (e.g., Perry and Southwell, 2014). Our census examination comprises 521 schools. Four ‘special schools’ were excluded: one that uses distance learning for students who cannot attend school and three alternative schools for students with severe learning impairments or behavioural problems, none of which offer academic curricular offerings.

We chose Victoria for several reasons. First, as a populous and prosperous state, conducting our study in Victoria enhances our ability to generate solid theoretical insights because we can minimize the possibly confounding effects of high levels of social exclusion, poverty and geographic isolation that are more pronounced in some Australian jurisdictions such as the Northern Territory. Victoria is the second most populous state in Australia, with more than one quarter of Australians residing within its border. Second, the authors are familiar with the Victorian context, which enhances our ability to contextualize and interpret our findings. Third, we chose Victoria as our case because its public reporting of school-level curricular offerings enabled rigorous data collection without the need to request data from jurisdictional education authorities, who might have been reluctant to provide these data because of a desire to avoid sector comparisons.

Data

We collected data from two publicly available government websites: the federal government’s My School website for school characteristics (myschool.edu.au) and the Victorian Curriculum and Assessment Authority (VCAA) website for curricular offerings (https://vcaa.vic.edu.au/). The school characteristics include enrolment size, sector, location and socio-economic composition (i.e. school mean SES). We used school enrolment size, which is listed on My School, to estimate the number of students per year (grade). My School uses five categories to denote school location: major cities, inner regional, outer regional, remote and very remote. Our dataset includes 352 metropolitan (major cities) schools, 114 inner regional schools, 50 outer regional schools, 4 remote schools and 0 very remote schools; due to the small number of remote schools and for ease of reporting, we collapsed remote schools into the outer regional category. In addition, MySchool divides schools into two school sectors: government (public) and non-government (private). As is commonly reported in Australian research and as is reported on the VCAA website, we further divided non-government schools into ‘Catholic’ or ‘independent’.

We used the Index of Community Socio-Educational Advantage (ICSEA), a measure developed by ACARA and included on My School, as a proxy for school socioeconomic status (SES). ICSEA is based on student characteristics (parental occupation and parental educational attainment) and school characteristics (proportion of Indigenous students and school location). ICSEA scores are scaled so that the national median is 1000 and a standard deviation is 100. Scores range from a low of 500 (representing extreme disadvantage) to about 1300 (representing extreme advantage). The ICSEA values in our population range from 863 to 1203.

Table 2 shows the distribution of schools in our sample. School SES quintiles are ordered from a low SES (quintile 1) to high SES (quintile 5). The proportions of schools for each sector are shown within each school SES quintile. For example, 86% of low SES metropolitan schools are government schools, and 77% of high SES schools are independent schools. The reported proportions do not always add up to 100% due to rounding.

Table 2.

Summary statistics of sample.

Metropolitan schools
School SES quintile	Mean number of students enrolled per school	Gov N (%)	Cath N (%)	Ind N (%)	Total number of schools
1 (low)	856	61 (86%)	7 (9%)	3 (4%)	71
2	1054	53 (76%)	8 (11%)	9 (13%)	70
3	908	23 (33%)	22 (31%)	25 (36%)	70
4	1047	24 (34%)	17 (24%)	29 (41%)	70
5 (high)	1118	9 (13%)	7 (10%)	55 (77%)	71
Total	1110	170 (48%)	61 (17%)	121 (34%)	352

Inner regional schools
School SES quintile	Mean number of students	Gov N (%)	Cath N (%)	Ind N (%)	Total number of schools
1 (low)	574	22 (96%)	1 (4%)	0 (0%)	23
2	562	22 (100%)	0 (0%)	0 (0%)	22
3	537	18 (78%)	5 (22%)	0 (0%)	23
4	590	5 (22%)	13 (57%)	5 (22%)	23
5 (high)	829	1 (4%)	6 (26%)	16 (70%)	23
Total	616	68 (60%)	25 (22%)	21 (18%)	114

Outer regional/remote schools
School SES quintile	Mean number of students	Gov N (%)	Cath N (%)	Ind N (%)	Total number of schools
1 (low)	603	11 (100%)	0 (0%)	0 (0%)	11
2	592	10 (91%)	1 (9%)	0 (0%)	11
3	533	10 (100%)	0 (0%)	0 (0%)	10
4	288	10 (91%)	1 (9%)	0 (0%)	11
5 (high)	542	7 (64%)	2 (18%)	2 (18%)	11
Total	511	48 (89%)	4 (7%)	2 (4%)	54
Grand total		288 (55%)	90 (17%)	143 (27%)	521

As shown in Table 2, government schools comprise just under half (48%) of all metropolitan schools, 60% of inner regional schools and 89% of outer regional/remote schools. Enrolment size is related to location and socioeconomic composition; on average, the largest schools are high SES schools in metropolitan areas. Finally, a clear pattern can be seen between school sector and school SES. In all three location categories, low SES schools are predominantly from the government sector, and high SES schools are predominantly from the independent sector.

Analytical approach

Our analytical strategy comprises two approaches: descriptive statistics to answer Research Question 1 and a logistic regression to answer Research Question 2. The descriptive statistics show the proportions of schools that offer the four curriculum subjects, disaggregated by school SES quintile, school sector and location. This approach is not only simple but also easily accessible and powerful for uncovering inequalities. The logistic regression has the school as the unit of analysis, the dependent variable is whether the school offers Specialist Mathematics or not and the independent variables include school year size, school sector (government, Catholic, independent), school socioeconomic composition (ICSEA score) and school location (metropolitan, inner regional and outer regional/remote). We conducted the logistic regression for Specialist Mathematics since it was the only subject which was offered in less than 90% of schools in our census. Our aim with the logistic regression was to determine the relative strength of the predictor variables and their independent effects. Logistic regression is used to obtain odds ratio when there are more multiple independent variables and the dependent variable is binomial (Sperendio, 2014). The result is the impact of each variable on the odds ratio of the observed event of interest. We do not provide statistical comparisons of groups (e.g. between metropolitan and inner regional) or tests of statistical significance for any of the analyses; as our dataset is a census population rather than a sample, inferential statistics are not necessary. Moreover, applying inferential statistics to census data can lead to inaccuracies (Gibbs et al., 2015).

Findings

In this section, we present the findings of our two research questions.

Research Question 1

To answer Research Question 1, we conducted a series of descriptive analyses. The first analysis shows the proportion of schools, by school location, that offer each of the four subjects, as well as the total number of these four subjects offered. As shown in Table 3, Biology and Chemistry are almost universally offered in all schools; Physics is offered in most metropolitan and inner regional schools, but not in 15% of outer regional schools. The provision of Specialist Mathematics, however, is very limited in all school location categories; the subject is offered in 71% of metropolitan schools, 48% of inner regional schools and 22% of outer regional schools. Overall, 61% of schools in the state offer Specialist Mathematics.

Table 3.

Proportion of schools that offer the four science and math subjects, by location.

	Metropolitan schools N = 352	Inner regional schools N = 114	Outer regional/remote schools N = 54	All schools N = 520
Biology	97%	97%	100%	97%
Chemistry	96%	95%	85%	95%
Physics	95%	89%	85%	93%
Specialist mathematics	71%	48%	22%	61%
4 subjects	69%	47%	22%	60%
3 subjects	26%	39%	52%	31%
2 subjects	3%	10%	22%	6%
1 subject	1.4%	4%	4%	2.3%
0 subject	1.4%	0%	0%	1.4%

We then examined in more fine-grained detail the characteristics of schools that offer all four subjects, as shown in Table 4. Large disparities are seen between all three locations, school SES quintiles and school sectors.

Table 4.

Number and proportion of schools offering all four science and mathematics subjects, by school SES, location and sector.

Metropolitan	School SES quintile	Gov	Cath	Ind	Total
	1	25 of 61 schools (41%)	7 of 7 schools (100%)	0 of 3 schools (0%)	32 of 71 schools (45%)
	2	37 of 53 schools (70%)	5 of 8 schools (63%)	5 of 9 schools (56%)	47 of 70 schools (67%)
	3	19 of 23 schools (83%)	18 of 22 schools (82%)	8 of 25 schools (32%)	45 (46) of 70 schools (64%)
	4	20 of 24 schools (83%)	16 of 17 schools (94%)	19 of 29 schools (66%)	55 of 70 schools (79%)
	5	8 of 9 schools (89%)	5 of 7 schools (71%)	51 of 55 schools (93%)	64 of 71 schools (90%)
Total		109 of 170 schools (64%)	51 of 61 schools (84%)	83 of 121 schools (69%)	243 of 352 schools (69%)

Inner regional	School SES quintile	Gov	Cath	Ind	Total
	1	2 of 22 schools (9%)	1 of 1 school (100%)	NA	3 of 23 schools (13%)
	2	12 of 23 schools (52%)	NA	NA	12 of 22 schools (53%)
	3	9 of 18 schools (50%)	3 of 5 schools (60%)	NA	12 of 23 schools (53%)
	4	1 of 5 schools (20%)	8 of 13 schools (62%)	2 of 5 schools (40%)	11 of 23 schools (48%)
	5	0 of 1 school (0%)	5 of 6 schools (83%)	11 of 16 schools (69%)	16 of 23 schools (70%)
Total		24 of 68 schools (35%)	17 of 25 schools (68%)	13 of 21 schools (62%)	54 of 114 schools (47%)

Outer regional/remote	School SES quintile	Gov	Cath	Ind	Total
	1	1 of 11 schools (9%)	NA	NA	1 of 11 schools (9%)
	2	4 of 10 schools (40%)	1 of 1 school (100%)	NA	5 of 11 schools (45%)
	3	1 of 10 schools (10%)	NA	NA	1 of 10 schools (10%)
	4	2 of 10 schools (20%)	0 of 1 school (0%)	NA	2 of 11 schools (18%)
	5	1 of 7 schools (14%)	2 of 2 schools (100%)	0 of 2 schools (0%)	3 of 11 schools (27%)
Total		9 of 48 schools (19%)	3 of 4 schools (75%)	0 of 2 schools (0%)	12 of 54 schools (22%)

Among metropolitan schools, 45% of low SES schools offer all four subjects, compared to 90% of high SES schools. Overall, 64% of public metropolitan schools offer all four subjects. The proportion of independent metropolitan schools that offer all four subjects is similar, at only 69% of schools. Most (93%) high SES independent metropolitan schools offer all four subjects, but the proportion is much smaller among lower SES independent metropolitan schools (0% for quintile 1, 56% for quintile 2 and 32% for quintile 3). For Catholic metropolitan schools, however, a distinct pattern among school SES and access to all four subjects is not clear. On average, 84% of Catholic metropolitan schools offer all four subjects. Interestingly, the highest proportion is seen in low SES Catholic metropolitan schools; all seven of the low SES Catholic schools offer all four subjects. Among metropolitan schools, Catholic schools are the most likely of the three sectors to offer all four subjects (84%), with only 69% of independent and 64% of public schools offering all four subjects.

Among inner regional schools, only 13% of low SES schools offer all four subjects, compared to 70% of high SES schools as shown in Table 4. In terms of sector, 35% of public schools, 68% of Catholic schools and 62% of independent schools offer all four subjects. These findings suggest that access to science and mathematics subjects is far from universal in inner regional communities. It also suggests that in many communities, access to these subjects is only achieved by paying fees to attend a non-government school. This finding is even more striking in outer regional communities, where access to all science and mathematics subjects is only available in 19% of government schools and 9% of low SES schools (all of which are in the government sector). Overall, only 22% of schools in outer regional communities offer all four subjects.

Lastly, we examined in closer detail access to Specialist Mathematics (Table 5) since it is the least likely to be offered by schools. Access to this subject is far from universal, and large disparities are found between school locations, sectors and SES quintiles. For example, 95% of high SES metropolitan independent schools offer Specialist Mathematics, compared to 9% of low SES inner or outer regional government schools.

Table 5.

Schools that offer specialist mathematics.

Metropolitan
School SES quintile	Gov	Cath	Ind	Total
1	25 of 61 schools (41%)	7 of 7 schools (100%)	0 of 3 schools (0%)	32 of 71 schools (45%)
2	37 of 53 schools (70%)	6 of 8 schools (75%)	6 of 9 schools (67%)	49 of 70 schools (70%)
3	20 of 23 schools (87%)	19 of 22 schools (86%)	9 of 25 schools (36%)	48 of 70 schools (69%)
4	20 of 24 schools (83%)	16 of 17 schools (94%)	21 of 29 schools (72%)	57 of 70 schools (81%)
5	8 of 9 schools (89%)	5 of 7 schools (86%)	52 of 55 schools (95%)	65 of 71 schools (92%)
Total	110 of 170 schools (65%)	53 of 61 schools (87%)	88 of 121 schools (73%)	251 of 352 schools (71%)

Inner Regional
School SES quintile	Gov	Cath	Ind	Total
1	2 of 22 schools (9%)	1 of 1 school (100%)	NA	3 of 23 schools (13%)
2	12 of 22 schools (55%)	NA	NA	12 of 22 schools (55%)
3	9 of 18 schools (50%)	3 of 5 schools (60%)	NA	12 of 23 schools (52%)
4	1 of 5 schools (20%)	8 of 13 schools (62%)	2 of 5 schools (40%)	11 of 23 schools (48%)
5	0 of 1 school (0%)	5 of 6 schools (83%)	12 of 16 schools (75%)	17 of 23 schools (74%)
Total	24 of 68 schools (35%)	17 of 25 schools (68%)	14 of 21 schools (67%)	55 of 114 (48%)

Outer Regional/Remote
School SES quintile	Gov	Cath	Ind	Total
1	1 of 11 schools (9%)	NA	NA	1 of 11 schools (9%)
2	4 of 10 schools (40%)	1 of 1 school (100%)	NA	5 of 11 schools (45%)
3	1 of 10 schools (10%)	NA	NA	1 of 10 schools (10%)
4	2 of 10 schools (20%)	0 of 1 school (0%)	NA	2 of 11 schools (18%)
5	1 of 7 schools (14%)	2 of 2 schools (100%)	0 of 2 schools (0%)	3 of 11 schools (27%)
Total	9 of 48 schools (19%)	3 of 4 schools (75%)	0 of 2 schools (0%)	12 of 54 (22%)

In summary, the descriptive analyses show varying degrees of access to the four science and mathematics subjects. Biology and Chemistry are almost universally offered, and Physics is widely offered. Access to Specialist Mathematics, however, is very restricted in some school contexts and is particularly limited in non-metropolitan schools.

Research Question 2

Next we report the results of the logistic regression, which we conducted to answer Research Question 2. There were no violations to the assumption of linearity in the logit (Box–Tidwell test). The likelihood ratio test indicated that the model performed significantly better than the intercept-only model (χ²(6) = 223, p < .001). The Hosmer and Lemeshow goodness of fit test showed no evidence of poor fit (χ²(8) = 6.90, p = 0.548), and McFadden’s pseudo R² indicated a good fit (R² = 0.325). We report the log odds and odds ratios for each independent variable.

From Table 6, we can see that the Catholic and independent sectors were associated with increased odds of offering Specialist Mathematics compared to government schools. The Catholic school effect was stronger than independent schools. Inner regional schools had similar odds to metropolitan schools of offering Specialist Mathematics, whilst outer regional area was associated with reduced odds. In addition, increases in ICSEA and school size were associated with increased odds of a school offering Specialist Mathematics. Controlling for all other variables in the model, the odds of an outer regional school offering Specialist Mathematics were 0.4676 of that of a metropolitan school (53.24% reduced odds). With a one-point increase in ICSEA, the odds of a school offering Specialist Mathematics increased by a factor of 1.0123 (1.23% increase in odds per point). This is a very substantial increase as the range of ICSEA scores in our population spanned approximately 350 points. To illustrate, the odds ratio for a 100-point increase in ICSEA is 3.42.

Table 6.

Log odds and odds ratios.

	Log odds	Odds ratio
(intercept)	−13.95	NA
Sector – independent	0.3480	1.4163
Sector – catholic	0.7783	2.1778
Location – inner regional	0.0212	1.0214
Location – outer regional	−0.7601	0.4676
ICSEA	0.0123	1.0123
School size	0.0170	1.0171

Discussion

Our analysis uncovered substantial between-school inequalities in access to science and mathematics subjects. While most secondary schools offer Biology and Chemistry, opportunities to learn Physics are not available in some schools and Advanced Mathematics is even less accessible. In some contexts, opportunities to learn can be severely constricted, especially for Advanced Mathematics, a core subject for any science-based university degree. These inequalities are particularly pronounced among regional schools and low SES metropolitan schools. School size, school socioeconomic composition, school location and school sector all have independent effects on the likelihood of a school offering advanced mathematics. Our findings suggest that powerful structural inequalities are shaping students’ opportunity to learn science and mathematics subjects in upper secondary school.

These structural inequalities are even more powerful because they are interrelated, compounding and cementing each other into a Gordian knot that is difficult to resolve. While school sector, socioeconomic composition and school location have independent effects, they are also interrelated. For example, 41% of low SES public metropolitan schools offer all four subjects, compared to 9% of low SES public regional schools. Similarly, high SES independent metropolitan schools are the most likely to offer all four subjects in metropolitan settings, but lower SES independent metropolitan schools are the least likely to offer all four subjects. These examples show how school sector, school location and school socioeconomic combine in complex ways. On the other hand, the provision of all four subjects in Catholic schools is moderated less by school location and/or school socioeconomic composition.

Marketization forces are the core of this knot. School choice and competition impact school size and school socioeconomic composition, leading to large schools of choice and, on the other hand, small residualized schools of last resort. As schools become more socially segregated and stratified in terms of resources and offerings, choosing the right secondary school becomes even more paramount for many parents. Per-pupil funding policies and the public subsidy of private schools further entrench this vicious cycle (Kenway and Boden, 2025; Preston, 2025). We describe these dynamics in more detail below, as well as some additional factors related to the structure and organization of curriculum offerings.

Marketization dynamics that reduce opportunities to learn

As discussed earlier, marketization includes the dynamics of choice and competition, which are facilitated by school autonomy and decentralized decision making. School funding models are part of the marketization juggernaut because they create different ‘price points’ in the educational marketplace, similar to any other markets of consumer goods (Perry et al., 2022). Many private schools have higher levels of funding (Preston, 2025), which enable them to provide enhanced resources and opportunities.

At the most basic level, science and mathematics curricular offerings vary between schools because of the decentralization of curricular decision making. As described earlier, schools in Australia are given autonomy to choose which curricular subjects to offer in the final 2 years of secondary school. This autonomy in turn is conditioned by other factors which explain why some schools offer particular curricular subjects, such as Advanced Mathematics, and other schools do not. In Australia, the factors that influence school curricular decisions are student demand, school funding and resources including teaching staff and school strategies for maintaining or increasing their profile and share in the local education market (Perry and Lubienski, 2020; Teese, 1998). These factors have also been identified in other national contexts, including student demand (Marginson et al., 2013), school marketing (Lubienski, 2006) and school resources (Darling-Hammond, 2010; Saw and Agger, 2021).

Funding models in Australia make it difficult for some schools to offer a wide range of subjects, especially in view of school size and socioeconomic composition. Schools receive funding based on the number of pupils they enrol and are given the autonomy to hire staff and make curricular decisions as they see fit and within their budget. Offering a curricular subject with low student numbers (typically less than 15 students) is economically unviable but some schools will nevertheless accept the cost, often as a strategic investment because they seek to maintain their profile and status in the educational marketplace (Perry and Lubienski, 2020). For example, well-resourced schools are particularly well placed and incentivized to make the strategic investment to offer advanced mathematics to a small number of students. Smaller schools, schools that are hard to staff and schools with lower socioeconomic compositions – features that frequently come together – face difficulties in offering a wide range of curricular subjects, especially those that are considered academically rigorous, such as advanced mathematics (Perry and Lubienski, 2020). This is because they have less overall funding and smaller staff numbers because of their smaller enrolment size, and their lower socioeconomic compositions mean they typically experience lower student demand for rigorous academic subjects, more difficulty recruiting and retaining qualified staff and additional funding demands for support services (Blackmore et al., 2024; Kelly and Fogarty, 2015; Perry and Lubienski, 2020).

As access to science and mathematics subjects is patterned by school social composition, it makes sense to consider the systemic features that may exacerbate social segregation between schools. As described earlier, Australian schooling has several features that are linked with its high level of social segregation of schools, namely, high levels of privatization and school choice and competition (Perry et al., 2024a). Access to science and mathematics curricula is likely to be both a cause and a consequence of school segregation. Schools that do not offer advanced mathematics and physics may be less attractive for families that aspire to university study and science and mathematics careers. Consequently, such families may choose a different school in the educational marketplace, thereby contributing to the residualization of the local public school (Campbell et al., 2009; Lamb, 2007). At the same time, residualized schools are less likely to be able to offer these science and mathematics subjects due to low student demand and interest, thereby contributing to a vicious cycle of stratification and segregation.

In many rural/regional communities and urban working-class communities, access to science and mathematics curricula requires families to pay for private schooling. For rural students, this may even mean boarding away from home, a practice that has been common in Australia among affluent farming families for decades (Hodges et al., 2013). Limited curricular choices in upper secondary school drive some educationally aspirational families to leave the local school or even the rural community, with negative consequences for the students and community members left behind, as has been found in Australia (Beswick et al., 2022; Hodges et al., 2013; Stokes et al., 2000) and Sweden (Fjellman et al., 2019).

Our finding that Catholic schools, regardless of their socioeconomic composition, provide widespread access to science and mathematics subjects is consistent with Xu and Kelly (2020) from the US. It may be that Catholic schools are committed to offering ‘core’ curricular subjects as part of their Catholic ethos and mission, as has been suggested for Catholic schools in the US by Lee et al. (1998). It is also possible that low SES Catholic schools are better placed to offer these subjects than public schools with similar compositions because of parental expectations and student demand. Given the limited access to advanced mathematics in low SES public schools, for example, it is plausible that Catholic schools provide the only opportunity to study science and mathematics subjects in many working-class communities. Families with aspirations for STEM careers and perhaps university study more generally may choose a Catholic school over the local public school in these working-class communities for the curricular advantages that the former provides (Teese, 1998). This in turn creates sufficient demand for science and mathematics (and other traditional academic) curricular subjects, which makes it more sustainable for a school to resource and staff them. From a marketization perspective, it is plausible that Catholic schools occupy a unique position that is due to their traditional ethos of providing scholastic education to their constituents, many of whom have traditionally come from the working-class.

Organization of curriculum

Marketization’s impact on opportunities to learn math and science subjects is compounded by another element, namely, the impact of curriculum structures and practices on student demand and selection of curriculum subjects to study. In Australia (and the UK and many other countries with comprehensive education systems), the practice of subject specialization in upper secondary school may reduce widespread participation in science and mathematics subjects. Rather than study a broad range of curricular subjects, as is common in the US, Australian students are encouraged to choose subjects based on their interest and aptitude, with no requirement to study a mathematics or science subject. Second, Australia’s use of externally assessed, subject-specific examinations as a basis for secondary school completion and university admittance may create additional barriers for universal access to science and mathematics subjects. External examinations are linked with decreased student interest in science and mathematics careers, and by association, interest in studying science and mathematics subjects in secondary school (Han, 2016). Third, the large range of externally assessed curricular subjects makes it inevitable that some subjects will not be offered. It also increases opportunities for students to be drawn to other subjects (Kennedy et al., 2014).

Limitations and future research

Our study is limited in that we did not examine curricular offerings beyond science and mathematics subjects. Examining the breadth of curricular offerings and enrolment numbers in other subject areas would have extended our ability to explain the availability of science and mathematics offerings. Future research could examine whether the availability of science and mathematics offerings is linked with the offerings of other subject areas. For example, is the availability of science and mathematics offerings greater in schools that offer fewer arts and humanities subjects, when school characteristics such as socioeconomic composition, location and size are controlled? A further future research opportunity would be to investigate if the issues revealed in this study have been stable over-time by examining mathematics and science offerings data over the past 20 years and considering if there have been changes in response to various policy initiatives.

We did not include in our analysis access to curriculum provided by distance education providers, such as the Victorian Virtual Learning Network (VVLN). The VVLN gives students access to curricular subjects that are not offered at their home school. It is likely that some schools in our sample are providing virtual access to the science and math subjects examined in our study via the VVLN. Virtual/distance education is an important mechanism for increasing access to educational opportunities, but it comes with many challenges and is not as effective as face-to-face teaching for many students (Akpen et al., 2024).

Finally, we recommend further comparative research about stratified access to curricular subjects in a range of national contexts. Comparative research is particularly useful for generating theory about the conditions and contexts that explain a particular educational phenomenon or outcome. As research about curricular access is conducted in other countries, theoretical understanding of the contexts, characteristics, practices and policies of schools and education systems that explain stratified curricular access between schools will be further developed.

Conclusion

While not all students have the interest or motivation to study science and mathematics subjects, we argue that all students should have the opportunity to study them. It is neither fair nor efficient to base an innovation agenda on an education system that provides access to science and mathematics curricula based on one’s ability to pay for private schooling or reside within the catchment zone of a metropolitan middle-class public school. Ensuring this educational right does not mean that every school must offer every science and mathematics curricular subject, but it does mean that every student should be guaranteed free access to it.

Researchers have examined the reasons why student participation in science and mathematics offerings in secondary school and beyond is stagnant or even declining in many countries. Most researchers have focused on student-related factors that shape science and mathematics participation in secondary school. A main contribution of our paper is to show that structural inequalities, many of which relate to marketization dynamics, also underlie participation rates in science and mathematics curricula. It is difficult to increase science and mathematics participation rates if schools do not offer these subjects. Thus, efforts to increase science and mathematics participation should address structural inequalities in students’ opportunities to study science and mathematics subjects, as well as attempt to increase students’ motivation and interest. Our findings from Australia may offer insights for other countries with similar education policies and contexts, as well as for countries that are embracing educational marketization policies. Comparative cross-national research is essential for developing robust theory about the factors that shape stratified curricular access in comprehensive education systems. As theory is developed, researchers and policymakers will have a better understanding of the policies that can be leveraged to reduce unequal curricular access.

Footnotes

ORCID iDs

Laura B. Perry

Philip Roberts

Steve Murphy

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Note

Author biographies

Laura B. Perry, PhD, is Professor of comparative education and education policy at Murdoch University. She conducts research about educational disadvantage and inequalities, especially as they appear between schools, and the systems, structures and policies that shape them. The aim of her research is to inform policy and practice for improving equity of educational opportunities, experiences and outcomes. Specific research interests include educational marketization, school segregation and stratification and school funding.

Philip Roberts, PhD, is Professor of curriculum inquiry and rural education at the University of Canberra. His major ongoing research focuses on the role of knowledge in curriculum, rural knowledge and the sustainability of rural communities.

Steve Murphy, PhD, is Senior Lecturer of teacher education and assessment at La Trobe University, Australia. He conducts research about STEM education and rural education.

Christopher Lubienski, PhD, is Professor of education policy and Director of the Center for Evaluation and Education Policy at Indiana University. His research focuses on education policy, reform and the political economy of education, with a particular concern for issues of equity and access. His current work examines (1) organizational responses to competitive conditions in local education markets, including geo-spatial analyses education opportunities, and research on innovation in education markets and (2) policymakers’ use of research evidence as influenced by advocacy organizations.

References

ACARA (n.d.) Specialist mathematics: Rationale. https://www.australiancurriculum.edu.au/senior-secondary-curriculum/mathematics/specialist-mathematics/rationale/ (Accessed 17 May 2023).

Adelman

(2006) The Toolbox Revisited: Paths to Degree Completion from High School Through College. US Department of Education.

Aguirre

(2020) How can progressive vouchers help the poor benefit from school choice? Evidence from the Chilean voucher system. Journal of Human Resources 57: 956–997.

Akpen

Asaolu

Atobatele

, et al. (2024) Impact of online learning on student’s performance and engagement: a systematic review. Discover Education 3(1): 205.

Alegre

MÀ

Ferrer

(2010) School regimes and education equity: some insights based on PISA 2006. British Educational Research Journal 36(3): 433–461.

Alexander

Johnson

Kelley

(2012) Longitudinal analysis of the relations between opportunities to learn about science and the development of interests related to science. Science Education 96(5): 763–786.

Archer

Moote

Francis

, et al. (2017) Stratifying science: a Bourdieusian analysis of student views and experiences of school selective practices in relation to ‘Triple Science’ at KS4 in England. Research Papers in Education 32(3): 296–315.

Australian Bureau of Statistics (2020) Schools, Australia, 2019 (Cat. No. 4221.0).

Australian Industry Group (2015) Progressing STEM Skills in Australia. Australian Industry Group. https://cdn.aigroup.com.au/Reports/2015/14571_STEM_Skills_Report_Final_-.pdf

10.

Benito

Alegre

MÀ

Gonzàlez-Balletbò

(2014) School segregation and its effects on educational equality and efficiency in 16 OECD comprehensive school systems. Comparative Education Review 58(1): 104–134.

11.

Berliner

(2014) Effects of inequality and poverty vs. teachers and schooling on America’s youth. Teachers College Record 116(1): 1–26.

12.

Beswick

Roberts

Eacott

, et al. (2022) Rural and Regional Research Education Project: Final Report. UNSW Gonski Institute for Education and School of Education.

13.

Blackmore

Hobbs

Rowlands

(2024) Aspiring teachers, financial incentives, and principals’ recruitment practices in hard-to-staff schools. Journal of Education Policy 39(2): 233–252.

14.

Bonal

Bellei

(2019) Introduction: the Renaissance of school segregation in a context of globalization. In: Bonal

Bellei

(eds) Understanding School Segregation: Patterns, Causes and Consequences of Spatial Inequalities in Education. Bloomsbury, pp. 1–26.

15.

Campbell

Proctor

Sherington

(2009) School Choice: How Parents Negotiate the New School Market in Australia. Allen & Unwin.

16.

Carter

Welner

Ladson-Billings

(eds) (2013) Closing the Opportunity Gap: What America Must Do to Give Every Child an Even Chance. Oxford University Press.

17.

Chesters

(2019) Alleviating or exacerbating disadvantage: does school attended mediate the association between family background and educational attainment? Journal of Education Policy 34(3): 331–350.

18.

Chmielewski

(2014) An international comparison of achievement inequality in within-and between-school tracking systems. American Journal of Education 120(3): 293–324.

19.

Chubb

Moe

(1990) Politics, Markets, and America’s Schools. Brookings Institution Press.

20.

Civil Rights Data Collection (CRDC) (2018) 2015–16 Civil Rights Data Collection: STEM Course Taking. U.S. Department of Education Office for Civil Rights. https://www2.ed.gov/about/offices/list/ocr/docs/stem-course-taking.pdf

21.

Committee on STEM Education (2018) Charting a Course for Success: America’s Strategy for STEM Education. National Science & Technology Council. https://www.whitehouse.gov/wp-content/uploads/2018/12/STEM-Education-Strategic-Plan-2018.pdf

22.

Connors

McMorrow

(2015) Imperatives in Schools Funding: Equity, Sustainability and Achievement. Australian Council for Educational Research.

23.

Cooper

Berry

(2020) Demographic predictors of senior secondary participation in biology, physics, chemistry and earth/space sciences: students’ access to cultural, social and science capital. International Journal of Science Education 42(1): 151–166.

24.

Darling-Hammond

(2010) The Flat World and Education: How America’s Commitment to Equity will Determine our Future. Teachers College Press.

25.

Darling-Hammond

Bae

Cook-Harvey

, et al. (2016) Pathways to New Accountability Through the Every Student Succeeds Act. Stanford University.

26.

Dean

Roberts

Perry

(2023) School equity, marketisation and access to the Australian senior secondary curriculum. Educational Review 75(2): 243–263.

27.

Dilnot

(2018) The relationship between A-level subject choice and league table score of university attended: the ‘Facilitating’, the ‘Less Suitable’, and the counter-intuitive. Oxford Review of Education 44(1): 118–137.

28.

Education Council (2018) Optimising STEM Industry-School Partnerships: Inspiring Australia’s Next Generation. Education Council. https://www.educationcouncil.edu.au/site/DefaultSite/filesystem/documents/Reports/and/publications/Publications/Optimising/STEM/Industry-School/Partnerships/-/Final/Report.pdf

29.

Eisenhart

Weis

Allen

, et al. (2015) High school opportunities for STEM: comparing inclusive STEM-focused and comprehensive high schools in two US cities. Journal of Research in Science Teaching 52(6): 763–789.

30.

Fjellman

(2019) School choice, private providers and differentiated mobilities in Swedish metropolitan school markets: exploring through a counterfactual approach. International Journal of Educational Research 98: 171–191.

31.

Fjellman

A-M

Hansen

(2025) Examining trends in school segregation in compulsory and upper secondary education in Sweden: the showcase of an urban school market. Scandinavian Journal of Educational Research 69(7): 1359–1379.

32.

Fjellman

A-M

Hansen

Beach

(2019) School choice and implications for equity: the new political geography of the Swedish upper secondary school market. Educational Review 71(4): 518–539.

33.

Gibbs

Shafer

Miles

(2015) Inferential statistics and the use of administrative data in US educational research. International Journal of Research and Method in Education 40(2): 214–220.

34.

Gill

(2015) Provision of GCE A Level Subjects 2014: Statistics Report Series no. 83. Cambridge Assessment.

35.

Goodrum

Druhan

Abbs

(2012) The Status and Quality of Year 11 and 12 Science in Australian Schools. Australian Academy of Science.

36.

Goos

Kaya

(2020) Understanding and promoting students’ mathematical thinking: a review of research published in ESM. Educational Studies in Mathematics 103(1): 7–25.

37.

Greaves

(2023) Segregation by choice? School choice and segregation in England. Education Economics 32(1): 1–19.

38.

Hodges

Sheffield

Ralph

(2013) Home away from home? Boarding in Australian schools. Australian Journal of Education 57(1): 32–47.

39.

House of Common Science and Technology Committee (2017) Industrial Strategy: Science and STEM Skills. House of Commons. https://publications.parliament.uk/pa/cm201617/cmselect/cmsctech/991/991.pdf

40.

Irvin

Byun

Smiley

, et al. (2017) Relation of opportunity to learn advanced math to the educational attainment of rural youth. American Journal of Education 123(3): 475–510.

41.

Jabbar

(2015) Every kid is money’: market-like competition and school leader strategies in New Orleans. Educational Evaluation and Policy Analysis 37(4): 638–659.

42.

Jónasdóttir

Eiríksdóttir

Ragnarsdóttir

(2025) Policy confluences, curricular changes and access to academic knowledge: mapping the structure and content of the Icelandic upper secondary academic curriculum. Nordic Journal of Studies in Educational Policy 11(3): 1–16.

43.

Justman

Méndez

(2018) Gendered choices of STEM subjects for matriculation are not driven by prior differences in mathematical achievement. Economics of Education Review 64: 282–297.

44.

Kelly

Fogarty

(2015) An integrated approach to attracting and retaining teachers in rural and remote parts of Australia. Journal of Economic and Social Policy 17(2): 1–19.

45.

Kennedy

Lyons

Quinn

(2014) The continuing decline of science and mathematics enrolments in Australian high schools. Teaching Science 60(2): 34–46.

46.

Kennedy

Quinn

Lyons

(2020) The keys to STEM: Australian year 7 students’ attitudes and intentions towards science, mathematics and technology courses. Research in Science Education 50: 1805–1832.

47.

Kenway

Boden

(2025) The ‘better and fairer schools agreement’; fairer funding but not fair. Australian Educational Researcher 52: 2999–3019.

48.

Kenway

Boden

James

(2024) The accumulation economy of private schools: extraction, mystification and eepletion. Discourse: Studies in the Cultural Politics of Education 45(3): 346–362.

49.

Kolluri

(2018) Advanced placement: the dual challenge of equal access and effectiveness. Review of Educational Research 88(5): 671–711.

50.

Ladson-Billings

(2006) From the achievement gap to the education bebt: understanding achievement in U.S. schools. Educational Researcher 35(7): 3–12.

51.

Lafontaine

Baye

Vileuf

, et al. (2015) Equity in opportunity-to-learn and achievement in reading: a secondary analysis of PISA 2009 data. Studies in Educational Evaluation 47: 1–11.

52.

Lamb

(2007) School reform and inequality in urban Australia: a case of residualising the poor. In: Teese

Lamb

Duru-Belat

(eds) International Studies in Educational Inequality, Theory and Policy. Springer, pp. 1–38.

53.

Lamb

Fullarton

(2002) Classroom and school factors affecting mathematics achievement: a comparative study of Australia and the United States using TIMSS. Australian Journal of Education 46(2): 154–173.

54.

Lee

Chow-Hoy

Burkam

, et al. (1998) Sector differences in high school course taking: a private school or catholic school effect? Sociology of Education 71(4): 314–335.

55.

Lubienski

(2006) Incentives for school diversification: Competition and promotional patterns in local education markets. Journal of School Choice 1(2): 1–31.

56.

Lubienski

Perry

Kim

Canbolat

(2022) Market models and segregation: Examining mechanisms of student sorting. Comparative Education 58(1): 16–36.

57.

Marginson

Tytler

Freeman

, et al. (2013) STEM: Country Comparisons: Final Report. Australian Council of Learned Academies.

58.

Monk

Haller

(1993) Predictors of high school academic course offerings: the role of school size. American Educational Research Journal 30(1): 3–21.

59.

Morgan

Kirby

(2016) The UK STEM Education Landscape. A Report for the Lloyd’s Register Foundation from the Royal Academy of Engineering Education and Skills Committee. https://www.raeng.org.uk/publications/reports/uk-stem-education-landscape

60.

Murphy

(2019) School location and socioeconomic status and patterns of participation and achievement in senior secondary mathematics. Mathematics Education Research Journal 31(3): 219–235.

61.

Murphy

(2020) The impact of school disadvantage on senior secondary science: a study of patterns of participation and achievement in government secondary schools in Victoria, Australia. Research in Science Education 50: 1603–1618.

62.

National Research Council (2001) Adding It Up: Helping Children Learn Mathematics. National Academy Press.

63.

National Science Board (2018) Science and Engineering Indicators 2018. National Science Foundation.

64.

National Skills Commission (2022) 2022 Skills Priority List: Key Findings Report. Australian Government.

65.

New Schools Network (2016) Glass Ceilings: Access to A-Levels in England. Author.

66.

OECD (2019a) Balancing School Choice and Equity: An International Perspective Based on PISA. OECD.

67.

OECD (2019b) PISA 2018 Results: Combined Executive Summaries (Volume I, II and III). OECD.

68.

Patrick

Socol

Morgan

(2020) Inequities in Advanced Coursework: What’s Driving Them and What Leaders Can Do. The Education Trust. https://edtrust.org/wp-content/uploads/2014/09/Inequities-in-Advanced-Coursework-Whats-Driving-Them-and-What-Leaders-Can-Do-January-2019.pdf

69.

Perry

(2024) Tensions undermining equitable school funding: insights from Australia. Journal of Educational Administration and History 57(1): 56–73.

70.

Perry

Lubienski

(2020) Between-school stratification of academic curricular offerings in upper secondary education: school decision-making, curriculum policy context, and the educational marketplace. Oxford Review of Education 46(5): 582–600.

71.

Perry

Rowe

Lubienski

(2022) School segregation: theoretical insights and future directions. Comparative Education 58(1): 1–15.

72.

Perry

Southwell

(2014) Access to academic curriculum in Australian secondary schools: a case study of a highly marketised education system. Journal of Education Policy 29(4): 467–485.

73.

Perry

Thier

Beach

Anderson

Thoenessen

N-M

Roberts

(2024a) Opportunities and Conditions to Learn (OCL): A conceptual framework. Prospects 54(1): 55–72.

74.

Perry

Yoon

E-S

Sciffer

Lubienski

(2024b) The impact of marketisation on school segregation and educational equity and effectiveness: Evidence from Australia and Canada. International Journal of Comparative Sociology.

75.

Preston

(2025) The Australian private school ascendancy: origins, development and future. The Australian Educational Researcher. Advance online publication.

76.

Rumberger

Thomas

(2000) The distribution of dropout and turnover rates among urban and suburban high schools. Sociology of Education 73(January): 39–67.

77.

Saw

Agger

(2021) STEM pathways of rural and small-town students: opportunities to learn, aspirations, preparation, and college enrollment. Educational Researcher 50(9): 595–606.

78.

Schmidt

Burroughs

Zoido

, et al. (2015) The role of schooling in perpetuating educational inequality: an international perspective. Educational Researcher 44(7): 371–386.

79.

Smith

(2011) Staying in the science stream: patterns of participation in A-level science subjects in the UK. Educational Studies 37(1): 59–71.

80.

Smyth

Hannan

(2006) School effects and subject choice: the uptake of scientific subjects in Ireland. School Effectiveness and School Improvement 17(3): 303–327.

81.

Sperandei

(2014) Understanding logistic regression analysis. Biochemia Medica 24(1): 12–18.

82.

Stokes

Stafford

Holdsworth

(2000) Rural and Remote School Education: A Survey for the Human Rights and Equal Opportunity Commission. Youth Research Centre, University of Melbourne.

83.

Svoboda

Rozek

Hyde

, et al. (2016) Understanding the relationship between parental education and STEM course taking through identity-based and expectancy-value theories of motivation. AERA Open 2(3): 2332858416664875.

84.

Teese

(1998) Curriculum hierarchy, private schooling, and the segmentation of Australian secondary education, 1947-1985. British Journal of Sociology of Education 19(3): 401–417.

85.

Wang

M-T

Degol

(2013) Motivational pathways to STEM career choices: using expectancy–value perspective to understand individual and gender differences in STEM fields. Developmental Review 33(4): 304–340.

86.

Whitty

Power

(2000) Marketization and privatization in mass education systems. International Journal of Educational Development 20(2): 93–107.

87.

Wienk

(2022) Year 12 Mathematics Participation Report Card. Australian Mathematical Sciences Institute.

88.

Kelly

(2020) Re-Examining the public–catholic school gap in STEM opportunity to learn: new evidence from HSLS. Social Sciences 9(8): 137.

89.

Yang Hansen

Patsis

Gustafsson

J-E

(2025) How does school composition mitigate socioeconomic and ethnic gaps in students’ achievement in Sweden: a long-term trend between 1988 and 2020. Educational Review December: 1–26.

90.

Zancajo

Bonal

(2022) Education markets and school segregation: a mechanism-based explanation. Compare: A Journal of Comparative and International Education 52(8): 1241–1258.