A controversial league table of Scottish education authorities has been rubbished by one of the country's leading academics.
It is based on findings by market research company CACI, which suggest that, once social deprivation is taken into account, 10 Scottish authorities are doing less well than they should.
But Lindsay Paterson, professor of education policy at Edinburgh University, criticised the study for its absence of information on individual pupils' attainment and social circumstances. "You can't infer anything about the effectiveness of a local school system from highly- aggregated data of this sort," he said.
"The fallacy of doing so was first identified by Harvey Goldstein of the London Institute of Education more than a quarter of a century ago. It's dismaying that people are still trying to do this kind of thing. But they do it because it is, compared to proper research, quick and cheap."
The study examined the proportion of pupils gaining five or more Standard grade Credit or Intermediate 2 passes, and what social deprivation suggested should be achieved. Dundee was farthest below its expected pass rate (23 per cent instead of 29.6), but Aberdeen was farthest from its predicted place in the league table, down 15 places to 22nd. East Renfrewshire was farthest above its projected pass rate (62 per cent instead of 43), but Highland made the biggest jump in the table, up 11 places to 7th.
Professor Paterson said that, in moving from percentages to rank order, the study made no attempt to distinguish between large and small differences from expectation: Highland was only 4 per cent better than predicted; Edinburgh was 10 places lower on the league table than expected, yet only 1 per cent off target.
Authorities which fared badly have hit back after a report in a Sunday newspaper claimed the study challenged the link between deprivation and low achievement.
Craig Munro, senior education manager at Fife Council, said: "We are concerned that some of the information which has been reported is factually inaccurate and presents a distorted view of the very good progress in raising attainment within education in Fife.
"The way we benchmark progress against local authority comparators is rigorous, using the Scottish Index of Multiple Deprivation factors. This incorporates six criteria relating to deprivation, which appears to go unnoticed in this limited study."
A Glasgow City Council spokesman said Scotland had a "statistically robust" system for evaluating secondary schools in socio-demographic terms. This could not be used to draw comparisons with other authorities, as Glasgow was statistically too different, he said.
The research was defended by CACI's director of business planning, John Rae. He said the company's EducationAcorn analysis tool allowed greater precision than deprivation indices used by government and academics. The study's purpose, he added, was to identify points of interest, not provide explanations.