Location, location: Student achievement and regionalityResearch 5 Aug 2014 6 minute read
The further schools are from larger cities, the more challenging it is for their students to excel, but some buck the trend. Petra Lietz explains.
Location, location: Student achievement and regionality
In Australia it is often reported that average student achievement declines the further away from capital cities one gets.
For example, results from the 2013 National Assessment Program – Literacy and Numeracy (NAPLAN) show that nationally, for each assessment domain at each year level assessed, students attending schools in metropolitan locations have the highest mean score, followed by students from provincial schools, then students from remote schools, then students from very remote schools.
At face value this pattern suggests a lack of equity in Australian education, where average student achievement appears to be strongly affected by the geographic location of the school. But is this the case internationally, or in all Australian states and territories?
The international experience
Using results from the Organisation for Economic Cooperation and Development (OECD) 2012 Programme for International Student Achievement (PISA), it is possible to compare the average mathematics, reading and science achievement of 15-year-olds based on school location in Australia, New Zealand, Canada and the United States of America. The findings are presented in Table 1, where ‘village’ represents the least populated location category and ‘large city’ represents the most populated.
Table 1: Mean performance and standard error (SE) in PISA 2012 by school location
In the 2012 PISA survey the mean score across OECD countries for mathematical literacy is 494; 496 for reading literacy; and 501 for scientific literacy. For all OECD countries, 39 score points represent about one year of schooling.
As shown in Table 1, Australian student performance in maths, reading and science consistently and significantly increases with the size of the population base in which schools are located from village to large city. Students in the least-populated areas tend to perform below the OECD average while students in the most-populated locations tend to perform above it.
In New Zealand, student performance in all three domains consistently and significantly increases from village to city, but decreases again for students in schools in large cities.
In Canada, differences in performance between students in schools in villages and small towns are not significant. However, students in schools in these locations do perform at a significantly lower level than students in schools in the highest performing locations, which for mathematics and science are towns and cities, and for reading is cities.
While results for the USA cannot be considered substantive due to the large standard errors associated with the estimate, they do show a tendency across all domains for students in schools in villages to demonstrate the lowest performance and students in towns to demonstrate the highest performance.
Together, these results show that the pattern of student achievement declining further away from large cities in Australia is not the norm around the world. They also show that the difference in achievement between the strongest and weakest performing location categories – the equivalent of more than one-and-a-half years of schooling in Australia in reading and mathematics – does not have to be so pronounced.
Case study: South Australia
Just as the international evidence shows that the relationship between a school’s geographic location and student achievement is not uniform, examination of educational achievement at the state and territory level also reveals variation in the relationship.
Under the Science, Information and Communication Technology and Mathematics Education for Rural and Regional Australia (SiMERR-SA) project, students in South Australian government schools were followed over four years of schooling to measure changes in literacy and numeracy performance.
Figure 1 records the profiles of performance on the state-wide numeracy and literacy tests that preceded NAPLAN for metropolitan and non-metropolitan regions, as well as for four non-metropolitan subregions – large provincial, small provincial, remote and very remote – in the years from 2000 to 2006.
The metropolitan group performs at a higher level in both numeracy and literacy, equivalent to almost a semester of school learning. The very remote group performs about half a semester behind in literacy learning compared with the non-metropolitan group, but is not behind the non-metropolitan group in numeracy. Interestingly, no differences emerge in either numeracy or literacy performance between the other three non-metropolitan regional groups.
The numeracy and literacy tests are formed from subtests in numeracy comprising measurement, space and number, and sub-tests in literacy comprising reading, spelling and language.
Figures 2 and 3 present the profiles of the achievement of these groups of students on the subtests on numeracy and literacy respectively.
Figure 2 clearly shows the low performance of students in large provincial towns on the spatial subtest and the high performance of students in remote areas on the measurement subtest. Figure 3 illustrates the uniformity of the language subtest scores across the subgroups and the spelling subtest scores for all groups except the very remote group. The noticeably low scores of the very remote students on the reading subtest, together with the higher scores of students in large provincial towns, is cause for concern given the Australian government’s aim for greater equity of student outcomes across the nation.
While information for teaching and learning in non-metropolitan schools in South Australia can clearly be gained from SiMERRS-SA test scores directly, it is important to keep in mind that the interrelationships between factors operating at the regional, school and student levels and the test scores are complex.
Further developmental work is required not only in SA but across Australia and beyond in order to understand and address the complex relationship between a school’s geographic location and student achievement.
Nevertheless, the further schools are from larger metropolitan centres, and the facilities, services and resources available there, the more challenging it is for their students to excel.
This article is based on the conference paper, ‘Indigenous and rural students: Double whammy or golden opportunity? Evidence from South Australia and around the world’, delivered by Dr Petra Lietz, Dr I Gusti Ngurah Darmawan and Dr Carol Aldous at ACER’s Research Conference 2014 on the theme ‘Quality and Equity: What does research tell us?’ on 5 August.
The conference proceedings, including full conference papers, are available from the ACER research repository.