Evidence that free schools drive up standards shows only modest improvement, can’t always be transferred to other school types and is contradicted in USA by latest test results from New York

Janet Downs's picture
“…evidence shows free schools are driving standards up across the board.”

Michael Gove, 31 July 2013

But what is this evidence? The Department for Education (DfE) has provided an answer.

First, the DfE rightly said more autonomous schools tend to have higher results. Greater autonomy, the DfE wrote, was an “essential feature of free schools”. But all UK schools have a high degree of autonomy*. Free schools are academies and the Academies Commission (2013) found the “greater autonomy” associated with academy status is “not nearly as substantial as is often suggested.” The Commission concluded that non-academy schools could do most things that academies can do.

So, leaving autonomy to one side because it’s a feature of all UK schools, what other evidence did the DfE supply?

1 Bohlmark and Lindahl (2012) described the positive impact of Sweden’s free schools (friskolen). But the authors said their results couldn’t be applied to other countries**. Cook (FT 2012) concluded the improved educational performance attributed to friskolen was extremely modest**.

2 CREDO (2013) described the moderate progress of US charter schools. The DfE said “good-quality charter schools” had “some clear parallels with free schools”. But most freedoms associated with charter schools have been available to English schools for over 20 years. CREDO might have to rewrite its conclusions about US charter schools: the latest results from New York show that charter schools did badly in tests aligned to Common Core State Standards particularly the Knowledge Is Power Program (KIPP) schools constantly praised by Michael Gove for their “No Excuses” ethos.

3 Dobbie, W and Fryer, R (2009). But Fryer’s 2012 project paper said “as a whole, charter schools have yielded inconsistent results.”

4 Angrist, J et al (2010) found positive results in one KIPP charter. It would be unwise to generalise from the results of one school. In any case, KIPP schools did poorly in the recent NY standardised tests.

5 Hoxby, C M et al (2009) found positive results for NY charter schools. However, more recent tests show such schools performed badly.

6 Machin, S and Vernoit, J (2011) painted a reasonably optimistic picture of Labour’s sponsored academies. However, the authors warned this could not be used to justify converter academies. This caveat can be extended to free schools.

The evidence marshaled by the DfE to show free schools raise standards “across the board” does not actually do so. Improvement, where found, was only modest. In two cases the authors warned against applying the results to other types of school. Even CREDO evidence which found disadvantaged pupils, particularly those from a Black background, benefited from attending a charter school is now contradicted by the latest NY results. New York charters tend to attract a high proportion of such pupils.

The qualities associated with successful schools have little to do with their status whether as charters, friskolen, academies or free schools. Ofsted (2010/11 Annual Report) identified qualities found in “good” and “outstanding” schools. Academy status was not one of them.

Neither is free school status.


*See faq above Is it true that schools with more autonomy tend to achieve better results?

**See faq above What does a January 2013 review of evidence say about market intervention in education in Sweden and Chile?

Share on Twitter Share on Facebook

Be notified by email of each new post.

Add new comment

Already a member? Click here to log in before you comment. Or register with us.