A visiting British professor of statistics has slammed the My School website as useless, misleading and producing perverse effects.
Bristol University statistician, Harvey Goldstein, said that test scores on My School do not reflect the school’s teaching capacity, but rather the standard of students enrolled.
“I think the website is not very useful….There’s nothing at the moment in the My School system that makes any proper attempt to adjust for initial achievement that children bring to a school,” he said.
“So, if a school is in a very wealthy area, with a lot of middle class parents, it’s going to achieve well. We know that. That’s a fact. So, to a large extent, you’re just reflecting the composition of the school and it doesn’t necessarily say anything about what the quality of the teaching is.”
Professor Goldstein said that My School reflected “poor technical advice” and he criticised its failure to report statistical confidence limits for school test results.
He also said that the “like school” comparisons on My School are “crude” and are a “naive and misleading attempt to adjust for social background”.
“Schools select students with different achievements. Some take in students with high achievements and some take in students with low achievements for various reasons. This needs to be factored in if we are to make any sensible comparisons.”
Professor Goldstein said creating league tables based on school test results had “perverse” side effects.
“There’s a lot of research now that shows one of the things that happens is you get a narrowing of the curriculum because teachers teach to the test, they teach to maximise,” he said.
“It also induces a lot of stress in the kids themselves and in the teachers, which does not promote active learning.”
“Another side-effect is that it encourages schools to play the system, so if it really matters to have high test scores, then you will go out of your way to make sure that is the case.”
“What happens is teachers in schools will do what they can to maximise the student test scores so they would consciously or unconsciously promote those students who they think will show the greatest improvement at the expense of other students,” Professor Goldstein said. “That impacts badly on disadvantaged students.”
“You may concentrate on particular children. You may do all sorts of things that are bordering on cheating in order to gain the highest possible test score.”
Professor Goldstein criticised the Australian government for ignoring extensive international research and experience which showed league tables had perverse side-effects on students and teachers.
Further articles on school league tables are available on Professor Goldstein’s personal website.