The Tribulations of Reusing and Repackaging Data - A Review of “How has School Productivity Changed in Australia?”

Oct 15, 2026

Uploaded by Jasmine

Argues that Leigh and Ryan’s 2008 study on Australian school productivity relies on an inappropriate statistical decomposition and draws conclusions that contradict earlier positive assessments of student standards. Contests the research in a shifting social policy context by challenging core assumptions about demographics, real per-child spending, schooling and economic change, class-size effects, and data (mis)use. Highlights how funding growth and expenditure allocation may be overstated, and questions whether declining teacher quality and teaching approaches contribute to observed outcomes.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Abstract

Introduction

Changes in “Real” Per Child Spending

Conclusions and Further Arguments

The Tribulations of Reusing and Repackaging Data: A Review of “How has School Productivity Changed in Australia?” David Zyngier Monash University Abstract In their paper, How has school productivity changed in Australia? social economists Andrew Leigh and Chris Ryan (2008)1 attempt to show that Australian (government) schools and their teachers’ productivity has declined by some 73 per cent between 1964 and 2003 compared to an overall increase of 64 per cent across the Australian economy. In response I contend that Leigh and Ryan use an inappropriate statistical device to “decompose” student demographics to arrive at conclusions that contradict the original positive assessment of student standards by independent authorities (Rothman, 2007) . Their work, which for many years seemed to be the foundation stone for Howard government education policy has by their own admission (A. Leigh & C. Ryan, personal communication, February 14, 2008), never been put to the test of peer review prior to publication. This paper critically analyses and contests their research by placing their claims within a social policy context that has subtly changed from one where teachers can make a difference to teachers are the difference. Introduction Despite significant publicity2, Leigh and Ryan (2008) reach radically different conclusions to the original research from which they drew their data. The 37 page report makes the following claims: • A small but statistically significant fall in numeracy has taken place over the period 1964-2003. • A small but statistically significant fall in both literacy and numeracy has taken place over the period 1975-1998. • The decline in both is in the order of one-tenth to one-fifth of a standard deviation. The Australian Educational Researcher, Volume 36, Number 2, August 2009 •73 DAVID ZYNGIER • Factors such as demographic or societal changes are not sufficient to explain theses changes; • During the relevant periods per child expenditure on education increased by 10 percent over the period 1975-1998, and by 258 percent over the period 1964-2003. • The “productivity” as measured by output per dollar has therefore fallen over the past 3-4 decades, specifically by 13 per cent with regard to numeracy and 73 per cent with regards to literacy. • Most of the increased expenditure can be explained by a decrease in class sizes and student teacher ratios. Therefore decreased class sizes have not increased educational “productivity”. • Another possible cause of the decrease may be “falling teacher quality”. This may be related to low teaching salaries, the implication (not specifically stated) being that increased salaries could lead to increased“teacher quality”. • The shift to a “whole-language” approach to teaching reading in the 1970s may have also contributed to the decline. The claims are based on four Longitudinal Surveys of Australian Youth (LSAY) cohorts, the Youth in Transition 1961 and 1975 birth cohorts (YIT 61 and YIT 75), the Longitudinal Surveys of Australian Youth 1995 and 1998 Grade 9 cohorts (LSAY 95 and LSAY 98) . The authors also examine numeracy data from the International Association for the Evaluation of Educational Achievement (IEA)3. Leigh and Ryan’s (2008) argument that teacher and school productivity has declined since 1964 relies on accepting their tendentious data in relation to: (i) changes in “real”per child spending,(ii) changes in schooling and the economy,(iii) constancy in student demographics,(iv) more teachers and lower class sizes, and,(v) data (mis)use. Changes in “Real” Per Child Spending Crucial to Leigh and Ryan’s (2008) productivity argument are the increases attributed to government funding per pupil since 1964, which they claim is largely driven by increased teacher numbers. Nevertheless, they overstate the increase in funding. They claim that government funding per pupil has increased from $165 (in today’s dollar terms) in 1964 to $71

FAQs

What is the main critique of Leigh and Ryan’s 2008 paper?

The review argues that Leigh and Ryan use an inappropriate statistical device to decompose student demographics, leading to conclusions that contradict earlier independent assessments of student standards.

Which core claims does the 37-page report (Leigh and Ryan, 2008) make?

It claims statistically significant falls in numeracy over 1964–2003, and in both literacy and numeracy over 1975–1998, with declines attributed to measured productivity falling as spending rises.

How does the review challenge the spending figures used to support productivity decline?

It disputes the magnitude of government funding per pupil by contrasting the authors’ claimed increase with Australian Bureau of Statistics Year Book figures, and discusses differences between public and private school funding allocations.

Oct 15, 2026

Uploaded by Jasmine