Tuesday, December 07, 2010

New evidence on class size effects

Class size effects: evidence using a new estimation technique

Kevin Denny, Veruska Oppedisano
This paper estimates the marginal effect of class size on educational attainment of high school students. We control for the potential endogeneity of class size in two ways using a conventional instrumental variable approach, based on changes in cohort size, and an alternative method where identification is based on restriction on higher moments. The data is drawn from the Program for International Student Assessment (PISA) collected in 2003 for the United States and the United Kingdom. Using either method or the two in conjunction leads to the conclusion that increases in class size lead to improvements in student’s mathematics scores. Only the results for the United Kingdom are statistically significant.

8 comments:

Martin Ryan said...

Interesting results Kevin. Was it not possible to include Ireland in the analysis? I am not too familiar with the availability of PISA data or existing studies using PISA data.

A story in today's Irish Times reports that "In 2007, Ireland was ranked 17th in the OECD with regard to maths literacy; we have now fallen to 26th".

The findings come from the 2009 Programme for International Student Assessment (PISA) report from the Organisation for Economic Co-operation and Development (OECD).

Irish Times article:
Irish students 'lag in maths'

Kevin Denny said...

No because you need countries where there are significant numbers in two adjacent years (say 8th & 9th grade) so you get within school variation (with between school sorting dealt with by school dummies). Only the US & UK had that.
You could apply one of out identification strategies (Lewbel approach) without that but we wanted to be able to do two.

Martin Ryan said...

If you tried the Lewbel approach on the Irish data would you be willing to divulge the results?

And... are you and Veruska the first authors to compare the Lewbel approach with the more familiar IV method?

Kevin Denny said...

We're the first to look at class size I think: there are about 5 or 6 papers who use it.
I think getting people to accept the approach may be tricky when you don't have a conventional instrument also.If you want to apply it to Irish data I am happy to help though.
The raw correlation (between class size and test score) is positive for most countries including Ireland. My conjecture is that the Lewbel method won't change that.

Liam Delaney said...

You will be happy to here that one of the few things that have not been touched in today's budget is the class size!

Martin Ryan said...

Thanks for the extra information Kevin.

Kevin Denny said...

Martin, re the PISA data: so our ranking has gone down.But does that mean that Irish students are doing worse or just that other countries are doing better?
I don't know if you can discern that. If each year the scores are normed (mean 500 I think) then you can't tell. Of course educationalists are all against league tables so they will surely disregard this particular one?
Liam:I suspect there will be upward pressure on class sizes one way or another. The drastic cut in the Educational Disadvantage Budget is a terrible sign.

Martin Ryan said...

Another good point Kevin. Though one can still say that we are further away from being the best.