Transcript Moving to the top ranked education systems

Improving Australia’s
systems of school education*
29 October 2010
CEET 14th Annual National Conference
Ascot House, 50 Fenton st, Ascot Vale, Melbourne
*This presentation is a work in progress. It should not be quoted
without the permission of the author.
Grattan Institute
Grattan Institute is Australia's newest think-tank:
independent and evidence based
• Grattan Institute is truly non-aligned
• The independence of the Institute is underpinned by
independent resourcing
Focus
• Cities
• Energy choices
• Productivity growth
• School education
• Water
P2
Improving school education
1. Moving to the top ranked education systems
2. Teacher effectiveness and student performance
3. Student performance and economic growth
1. How do we create the best education systems in the world?
2. If teacher effectiveness is the key, what is the increase in teacher
effectiveness required to lift us into the top performing countries?
3. What would be the impact of this improvement on long-run
economic growth?
P3
Moving to the top ranked education systems
• Australia ranks 8th in the latest PISA assessment, but only statistically
significantly behind Finland, Hong-Kong China, and Canada.
• There is a 36-point gap that must be overcome if Australia is to
replace Finland as the top performing education system in the world
• There is, on average, a 20 point gap between Australia and Finland,
Hong-Kong China, and Canada
P4
Moving to the top ranked education systems
• What does a 20 PISA score point gap represent?
• 38 points is equivalent to 1 year’s curriculum
• Similar to the 36-point gap to replace Finland as the top
performing education system in the world
• But only half a year’s (53%) of a year’s curriculum to be amongst
the top performers (Finland, Hong-Kong China, and Canada)
• Assuming that curriculum gains are additive (and that on average
students have 10 years education before they sit PISA) this
implies that each grade needs to incorporate an extra 5%
(0.526/10 = 0.0526) of a year’s worth of curriculum.
P5
Moving to the top ranked education systems
• How do we best achieve the additional 5% per year?
• We have a poor record in investing to improve student performance
• 41% increase in education expenditure between 1995 and 2006.
• However, between 2000 and 2006, Australian student
performance in PISA stagnated in mathematics and significantly
declined in reading
• Between 1964 and 2003, funding in the Australian school
education sector (government funding for both government and
non-government) increased 258%, while numeracy test results
significantly fell by 1.1 points
• Expenditure largely driven by class size reductions
P6
Moving to the top ranked education systems
• Class size reductions have continually been shown not to have a
significant impact on student performance
• Recent evidence from Florida’s mandated class size reductions
• Prekindergarten – 3rd grade – 18 students (reduction of 5
students (22%))
• 4th – 8th grade – 22 students (reduction of 2 students (9%))
• 9th-12th grade – 25 students (no change from average)
• Cost $20b over 8 years + $4b. p.a.
• $1500 per student per year
• $1.1 million per school per year
• Recent research found no significant impact on cognitive or noncognitive outcomes
P7
Moving to the top ranked education systems
• The impact of high and low quality teachers outweighs the impact of any
other school education program or policy
• The research findings on the relationship between teacher effectiveness
and student performance is remarkably consistent
• In Australia, conservative estimates suggest that:
• a student with an effective teacher can achieve in three quarters of a
year what would take a full year with a less-effective teacher
• a student with a teacher in the top 10% of teachers in the country
could achieve in a half year what a student with a teacher in the
bottom 10% of effectiveness achieves in a full year
• It’s also cumulative. Students who had three ‘effective’ teachers in a row
were 49-50 percentile points higher on school assessments compared to
students assigned ‘ineffective’ teachers over a three year period
P8
Moving to the top ranked education systems
• Australia needs to increase its PISA score by 20 points if it is to be
considered amongst the best performing education systems in the world.
1. Increase the effectiveness of all Australian teachers by 10%
2. Increase the effectiveness of the 15% least effective teachers in
Australia
P9
Increasing the effectiveness of all teachers
• 20 PISA score points is equal to 20/38=0.526 of a year’s worth of curriculum.
• This implies that each grade needs to incorporate an extra 5% (0.526/10 =
0.0526) of a year’s worth of curriculum.
• Research on the impact of teacher effectiveness on standardised test scores in
Australia shows that a 0.5 standard deviation in standardised test scores
represented advancing 1 grade.
• This implies that an extra 5% of a year’s worth of curriculum is equal to
0.0526x0.5 = 0.0263 standard deviations in test scores.
• 1 standard deviation increase in teacher effectiveness is associated with a 0.1
standard deviation increase in test scores.
• 0.0263 standard-deviations increase in test scores is equivalent to 0.26
standard-deviations increase in teacher effectiveness (10x0.0263 = 0.263).
• 0.26 standard deviations increase in test scores = 10% in teacher effectiveness
P 10
Impact on long-run economic growth
• Research connecting education with economic growth assumes that:
• the quantity of education is related to higher individual wages and
productivity
• productivity is a fundamental driver of economic growth
• More recent research indicates that the quality rather than the quantity
of education is more important
• Focus on quantity of education has been criticised
• Assumes that a year of schooling is the same in all countries
• Causality – economic growth results in greater investment in
education and years of schooling
P 11
Economic growth and years of schooling
Conditional economic growth
6
4
2
0
-2
Coef f = 0.581, s.e. = 0.095, t = 6.1
-4
-4
-2
0
2
4
6
Conditional years of schooling
Note: These results are from a regression of the average annual rate of growth (in
percent) of real GDP per capita in 1960–2000 on average years of schooling in 1960
and the initial level of real GDP per capita in 1960.
Source: Reproduced from Hanusheck and Wößmann (2007)
P 12
Impact on long-run economic growth
• Focus on education performance attempts to address these issues
• Causation
• Test scores are not correlated with school resources
• Countries with higher test scores in 1980 then had higher rates of growth
from 1980-2000
• Changes in test scores correlate strongly with changes in economic
growth rates
P 13
Student performance and economic growth
Trend in growth rate
0.15
New Zealand
0.10
Sweden Netherlands
0.05
Finland
Australia
Belgium
United Kingdom
0.00
United States
-0.05
Germany
France
-0.10
Italy
-0.15
-0.50
Japan
0.00
0.50
1.00
1.50
Trend in test scores
Note:
Only 12 countries have participated in international tests over a sufficiently long
period to look at trends over a 30 year period. In the chart, the ‘trend in growth
rate’ is simply a bivariate regression of test scores on time. Trends in test
scores are similarly derived. The plot provides the pattern of slopes from the
test regressions.
Source: OECD (2010), originally presented in Hanushek & Wössmann (2009)
P 14
Impact on long-run economic growth
• What is the relationship between student performance and economic growth?
• Hanushek and Wößmann
• One-standard deviation in test scores associated with
• 1.74% increase in GDP growth rates
• 1.47% increase in GDP growth rates for OECD countries
• 1.26% increase in GDP growth rates (including institutional
variables)
• Other research shows smaller gains
• Lee & Lee (1995)
- 1.2% increase in GDP growth rates
• Barro (2001)
- 1% increase in GDP growth rates
• Ramirez et al. (2007)
-0.6% increase in GDP growth rates
(insignificant for 1908-2000)
• Appleton et al. (2008)
-0.8% increase in GDP growth rates
• A conservative average has a one-standard deviation in test scores
associated with a 1% increase in GDP growth rates
P 15
Improving school education
• 20 PISA point gap between Australia and top performing countries
• Joining the top performers
• Increase PISA scores by 20 points
• 5% increase in the curriculum each year
• 10% increase in the effectiveness of all teachers
• 0.2% increase in GDP growth
• Make Australians 12% richer by the end of the century
P 16
Improving Australia’s
systems of school education*
29 October 2010
CEET 14th Annual National Conference
Ascot House, 50 Fenton st, Ascot Vale, Melbourne
*This presentation is a work in progress. It should not be quoted
without the permission of the author.