Do school test scores matter?

For years policy experts from free market think tanks have been arguing that charter schools and vouchers boost test scores. Last year Julie Novack’s report for the Institute of Public Affairs insisted that: "Voucher programs around the world have been shown to improve the academic performance of students" (pdf). But recent evaluation findings from Milwaukee have shifted the debate in the US. Faced with disappointing results, supporters of parental choice are now arguing that other things matter more. The Heritage Foundation’s Jason Richwine writes that sensible parents aren’t obsessing about test scores, instead what they really want are "schools that are safe, that cultivate a positive attitude about learning, and that best fit their children’s abilities and interests."

The American Enterprise Institute’s Charles Murray argues that test scores are a poor measure of school performance because schools are unable to control most of the things that drive children’s performance on standarised tests. In the New York Times he writes:

The evaluation by the School Choice Demonstration Project, a national research group that matched more than 3,000 students from the choice program and from regular public schools, found that pupils in the choice program generally had “achievement growth rates that are comparable” to similar Milwaukee public-school students. This is just one of several evaluations of school choice programs that have failed to show major improvements in test scores, but the size and age of the Milwaukee program, combined with the rigor of the study, make these results hard to explain away.

So let’s not try to explain them away. Why not instead finally acknowledge that standardized test scores are a terrible way to decide whether one school is better than another? This is true whether the reform in question is vouchers, charter schools, increased school accountability, smaller class sizes, better pay for all teachers, bonuses for good teachers, firing of bad teachers — measured by changes in test scores, each has failed to live up to its hype.

In the absense of data like Milwaukee’s, the debate in Australia is different. Here it’s right wing columnists like Janet Albrechtsen who are most excited by testing. In a 2008 blog post Albrechtsen praised New York City education chancellor Joel Klein’s approach of encouraging parents to "raise hell" if their children’s test scores came up short and suggested that public schools should face competition like they do in New York City. But back in the US the American Enterprise Institute’s Rick Hess writes that "choice enthusiasts have been overselling the miracle, restorative powers of choice for years." He acknowledges that, on its own, testing and offering parents choice probably doesn’t do much to improve children’s test scores.

8 thoughts on “Do school test scores matter?

  1. If the government really wanted to get something useful out of testing (versus all this testing which has no real effect), it would be much smarter to invest in testing young kids on really basic stuff (can you speak properly etc.). That way you could catch lots of kids before their problems snowball, and that would be worthwhile (and all you’d have to tell is the parents, and they’d no doubt be happy to hear it and happy to be given options about what could be done). Even if it was only done in some of the really poor schools it would probably be worthwhile.

    Off topic, I don’t know why you even think the opinion of Janet Albrechtson is worth mentioning. You could have a more nuanced conversation with a retarded monkey, and once Murdoch finally gives up on the Australian, I imagine that will pretty much be the end of her.

  2. Clearly school test results matter to the government which is pushing hard for them, not sure why though.

    The MySchools website doesn’t allow parents to compare the performance of schools in their immediate area. I mean why would you want to know which high school in your zone is more academic? But it allows you to compare your school with other schools around Australia of similar socio economic index. How many people care how interstate schools perform? In fact its stupider than that because year 9 is the third year of secondary school in NSW and Vic and the second year of hig school in Qld, WA and SA – so why compare them

    And why isn’t Scotch, Scots compared to Kings, Geelong Grammar, Southport, Melbourne, Xavier etc because this is the mobile parent cohort.

    I can’t imagine that state education departments are keen to embrace voucher programs with its attendant planning problems when parents flee from poorly performing schools leaving empty school buildings in one suburb while the next suburb is demanding more classrooms are built

    If schools are rigidly zoned then house prices near good schools will sky rocket

  3. “I can’t imagine that state education departments are keen to embrace voucher programs with its attendant planning problems when parents flee from poorly performing schools leaving empty school buildings in one suburb while the next suburb is demanding more classrooms are built”

    One of the weird things about this debate is that people don’t understand how much like a voucher system the current system already is. Basically, schools get at least part of their funding based on the number of students they have (see e.g., here for the mess the funding system is), so those schools without essentially unlimited resources like the rich private schools and those schools at the limit of their physical infrastructure have every reason to try and enroll as many students as they can (more money). In addition, nothing stops students enrolling wherever they want, excluding schools with special rules (which would no doubt remain even under a voucher system). Note that, at least in Vic, the obligation is on schools to take students in a certain district (e.g., here not to exclude students wanting to come in from elsewhere.

    An even more voucher-like system occurs in the universities, where money is given to universities based on the number of students they take (modified by the type of degree), but people still claim vouchers would help increase performance in universities. Now, as it happens, almost every university I know of competes for students and the money attached to them, excluding for courses where places are limited due to other factors like the course costing too much (and vouchers won’t fix that problem anyway if universities are not allowed to charge more). Thus, it’s hard to see how attaching funding to students as is the case now and getting rid of maximum enrollment numbers to try and make the voucher-like system function like a more pure voucher system will make any difference (indeed, the only university I know that complained about this in Victoria is the Aus. Catholic University, which basically means all the others think it will make no difference).

  4. That’s very interesting about the free-marketeers’ change of heart, but let’s be careful not to confuse the issues of (1) what a test score tells us about the abilities of the child and (2) what it tells us about the ‘performance’ of the score.

    I’ve been dismayed by Julia Gillard’s persistent blurring of these two issues in the last few months. The teachers’ unions made it absolutely clear that they agree with testing, and centralised testing, as a diagnostic tool, but believed that the publication of school average scores as putative performance indicators was harmful. Yet every time Gillard was asked about the teachers’ objections, she would babble on sanctimoniously about the importance of measuring childrens’ skills against national standards.

    I’ve written a couple of posts on the league table issue in the last few months, and no one has yet mounted a coherent defense of the publication of school averages. As usual it’s only the Greens who have a principled position on it. Joel Klein is the last wave of bully managerialism, spilling over from corporate into public sector culture.

  5. Conrad at (3) highlights an important aspect of what is wrong with the Australian voucher debate. There has been too much focus on parental choice, of which there is already a lot by world standards in Australia, and too little on supply-side reforms. I did not support Julie Novak’s paper at the time because it proposed huge expense in shifting more students to private schools, without dealing with the long-term structures and incentives of what are now government schools.

    On test scores, I think there are good reasons to be cautious about believing that there can be system-wide significant improvements over any short period of time. While better school management can probably secure some gains, if we believe the early chilhood research the socio-economic differences are pretty entrenched before the kids arrive at school. We can’t replace parents. The biggest single factor in school performance is usually teacher quality, and again there is not very much we can do about this in the short term (though recent efforts to get the worst teachers out of the classroom are sensible).

    The Australian research on school choice shows that academic matters are only one factor among many in why parents choose private schools. The debate here has been quite different to the debate in the US. There the abysmal academic performance of many public schools has driven the choice movement. Here it has been driven by demand for largely religious private schools, so values issues have been more important.

  6. sensible parents aren’t obsessing about test scores, instead what they really want are “schools that are safe, that cultivate a positive attitude about learning, and that best fit their children’s abilities and interests.”
    It looks like there’s a 2-factor thing at work.
    A parent won’t care about the test scores if they’re child isn’t safe, or if the school is failing to motivate them to learn anything at all.

  7. By God Murray is a foul hypocrite. He was all for standardised single test scores as a measure of everything in The Bell Curve.

    More broadly, there’s lots of intellectual dishonesty on display here. We all know that if the evaluation had found that the charter schools improved test scores then the same people now suddenly downplaying them would have been noisily promulgating the test scores’ rigour, importance and necessity. Just as if the Lancet study of casualties in Iraq had found a low number then they would have been praising its rigour, importance and necessity.

  8. A parent won’t care about the test scores if they’re child isn’t safe, or if the school is failing to motivate them to learn anything at all.

    I find it hard to believe that unsafe and unmotivated kids will consistently deliver high test scores, so the problems you describe should be easily detectable.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.