Pupils at faith schools are about one year ahead of those at general state schools, a study suggests.
Children at faith schools were ahead of their peers
Analysis of national tests scores of pupils in one London borough back up the findings of other recent studies.
Academics from the London-based National Institute of Economic and Social Research studied results for 11-year-olds in Barking and Dagenham.
They said possible reasons for the differences included pupils' family backgrounds and smaller year groups.
Pupils who were of the lowest ability gained the greatest relative advantage, they said.
For the study, researchers looked at the maths national test results of 11-year-olds in Barking and Dagenham for 2003.
A total of 303 children were in their final year at the seven religious foundation schools in the borough, while 2,076 children were at the borough's 28 local authority schools.
Researchers concluded that children in religious schools were about one year ahead of those in general schools, on average.
The advantage was greater for children in the lowest-performing band, they said.
The bottom 25% of pupils at church schools were about a year and a half ahead of those in bottom 25% of the general schools.
Study author Professor Sig Prais wrote: "It could be that the home backgrounds are more educationally and emotionally supportive - for example, having a greater general concern with the ideals of education."
He also suggested there could be a cumulative "infectious" effect from "a few disturbed or disturbing children", and said the generally smaller yearly intakes of church schools might foster a calmer atmosphere.
Other possible causes of the difference, he said, could include church schools being more academically selective, for example by suggesting to parents of children showing signs of disability that their child might be better catered for by a bigger local authority school.