Mobility of students in NSW government schools
This report was originally published 29 February 2016.
Student mobility has been an ongoing concern to educators due to the perceived negative effects that changing schools can have on students’ educational outcomes, on schools and on teaching. Mobility has been identified in studies in many countries including the United States, United Kingdom and Australia as being a contributing factor to student disadvantage, particularly in the areas of school engagement and lower achievement (Dobson, Henthorne & Lynas 2000; Navin, Hill & Doyle 2012; Reynolds, Chen & Herbers 2009).
In addition, significant student turnover may also impede teachers’ and administrators’ efforts to implement a reform agenda in schools thus hampering schools’ efforts to turn around low performance. Therefore, the need to understand mobility and its effect on schools’ ability to improve performance is critical, especially as greater emphasis is placed on school accountability.
Despite the widely held belief that mobility is damaging to student learning, little effort has been made to measure the extent and consequences of student mobility in the New South Wales school system. One major barrier has been a lack of access to high quality data bearing on student movements across schools and school systems. As a result, there has been insufficient evidence to comment on the nature, pattern and size of student mobility in NSW schools, let alone to discuss the policy significance of student transitions and mobility across different student groups, school structures and geographies.
Like a number of other Australian jurisdictions, the NSW Department of Education currently offers a small amount of funding for primary schools that believe they have high student turnover . However, this funding originated from a resourcing perspective, rather than an educational one, and is designed to compensate schools requiring additional resources such as textbooks and equipment. If mobility has as much impact on NSW student outcomes as demonstrated in other countries, there may be much greater need to support mobile students than has traditionally been provided.
The introduction of a unique Student Registration Number (SRN) in 2008 has provided the means to track students within the state government education system over time. Student mobility, approximated by student movements within the government sector, can now be calculated and its impact on student achievement can be estimated.
This report presents the results of a preliminary investigation into student mobility undertaken by the Centre for Education Statistics and Evaluation (CESE). The following section provides a brief overview of the international and national literature on student mobility, identifying ways in which mobility is defined and measured, and the effects that mobility has on educational outcomes. Section 3 presents the goals of the current study. Sections 4, 5 and 6 present the investigations and preliminary analyses undertaken by CESE to define and measure student and school mobility in NSW government schools and the effect mobility can have on students’ educational outcomes. Section 7 concludes with an overview of the findings, and suggestions for future directions.
Goals of the study
As the first large scale mobility study in NSW, using longitudinal administrative data to analyse mobility, we aim to:
- construct measures of student and school mobility that are specific to the NSW context
- understand the size of our mobile student population, their demographic characteristics,
- any persistent movement patterns and the geographical distribution of the mobile population
- examine the impact of mobility on student and school performance, and how this impact is moderated by other known background factors.
The following section presents the definition of student mobility developed for the current study, along with descriptive analyses of the extent of mobility in the NSW government student population, patterns of mobility, and the characteristics of students who move.
Student mobility is a significant issue for NSW government schools, with more than 30 per cent of students making at least one school move that is not required by the system structure during their years of schooling, and at least 5 per cent of all students, or 1 in 20 students, experiencing three or more non- structural school moves in their primary or secondary schooling career. In addition, 13 per cent of NSW government schools experience high levels of enrolment churn as benchmarked against international studies (Dobson, Henthorne & Lynas 2000). These estimates are conservative, as there is still insufficient system data to measure student mobility over the full 13 years of schooling, or to factor in student movements between government and non-government schools, or those that involve interstate or overseas schools.
Consistent with the literature, our analysis confirms that student mobility overlaps with other indicators of educational disadvantage, with mobile students more likely to be of Aboriginal descent or from low socio-economic family backgrounds. In fact, the lower achievement of these student groups may be partly explained by higher levels of disruption to schooling as a result of more frequent school changes, and higher rates of absence from school as suggested by the analysis of test participation.
The analysis also supports previous findings that mobility has a detrimental impact on student outcomes, even after other disadvantage background factors and prior achievement have been taken into account. Not only are mobile students more likely to be absent from school on national testing days, those that do attend achieve lower reading and numeracy results than stable students with similar backgrounds and prior achievement, especially in the early primary years. Further, the greater the number of moves, the greater the impact on outcomes, with test results for highly mobile students (those who move three or more times over a measured period for a given level of schooling) around 0.10 to 0.15 standard deviations below those for stable students with similar backgrounds and prior achievement.
In addition to the impact on achievement, mobility is also associated with a higher likelihood of leaving school prior to completing Year 12. Compared to stable students with similar backgrounds and prior achievement at the start of secondary education, changing school once during the junior secondary school years is associated with a 12 percentage point increase in the probability of students leaving school before completing Year 12. This effect increases to 21 percentage points for students who move twice and nearly 30 percentage points for those who move at least three times.
Mobility not only impacts on the educational outcomes of mobile student population, it also potentially causes disruption at the classroom level and interrupts the learning of stable students enrolled in schools with high enrolment churn. While our analysis of spillover effects did not detect an effect on stable students in the primary years and only found a small negative effect for students in the junior secondary years, it is worth noting that the school mobility rates as currently constructed and as previously noted underrepresent the true extent of the enrolment churn schools experience. The effects might be more notable if these measures are further improved.
Nonetheless, even based on the conservative estimates of school mobility using the current measures, 13 per cent of government schools are experiencing high rates of enrolment churn. For these schools, mobility could be a significant barrier to educators’ ability to provide coherent learning and support and has the potential to negatively impact on mobile students’ learning progress. As fewer students stay in the same school over a sustained period of time, it can be difficult for schools to implement an effective reform agenda to lift student and school performance.
Overall, the findings from this study indicate that student and school mobility is an additional indicator of educational disadvantage that the NSW government school system should monitor and for which specific policy responses may need to be developed. However, before policies and strategies can be developed, we need a deeper understanding of the causes of mobility, and the mechanisms by which mobility impacts on student achievement. In this regard, qualitative case studies of highly mobile students and schools would be useful to shed light on these other important aspects of student mobility.
It is noted that policies and strategies to reduce the impact of mobility on students’ educational outcomes have been suggested and/or examined in a number of the studies reviewed for this report. While beyond the scope of the current report, a review of the literature and evidence for such policies and strategies would also be useful to identify ‘what works’ to improve educational outcomes for mobile students.
Finally, before the Department starts monitoring mobility on a regular basis, further development and refinement of the measures of student and school mobility is required. For example, the types of school moves to be included in the measures (such as planned moves to or from a specialist support setting) need to be clarified and agreed with relevant Departmental stakeholders. Further, the limitations with the current methodology, that result in levels of student and school mobility being underrepresented, need to be overcome to ensure that the measures of mobility are as accurate and robust as possible.