Schools with a majority of pupils from deprived backgrounds expected new value-added data to show how well they are succeeding against the odds.
But Professor Peter Tymms of Durham University warned that results could look less impressive under a system which measures progress rather than attainment.
Value-added scores use children's test results at seven and 11, and compares their progress to average progression. The measurements will be piloted in the performance tables at the end of the year.
Professor Tymms said: "Value-added is seen as the saviour of schools in tough areas, but often it is these schools which might get the lower than average value-added. My worry is that the calculations will not take into account the very difficult circumstances these schools work in."
There are three main concerns: the quality of key stage 1 data, comparing progress between schools and pupil mobility.
Key stage 1 tests at seven are marked by pupils' own teachers. Local education authorities audit each school at least once every four years but this system means standards in different schools may not be comparable.
Deb Tyler, performance management, information and customer services manager at Newcastle city council, said: "We have had doubts about value-added for quite a long time. It is better than raw data but it does not take into account disadvantages which are compounded over time. It assumes that educational standards rise on a similar basis regardless of background."
And an Office for Standards in Education report has shown that pupil turnover is so rapid at some primaries that less than a fifth of children stay from five to 11.
The Department for Education and Skills found "fairly strong" support for the publication of value-added measures during its consultation this year.