No one knows whether pupils in composite classes perform better or worse than those in single stages, Valerie Wilson, head of the Scottish Council for Research in Education centre at Glasgow University, told a seminar.
Approximately one in four primary children is in a composite class, but there is still no Scottish research on them, Dr Wilson said.
Parents were reluctant supporters of composites, as they did not want their children to learn alongside the younger ones. But researchers say this may overlook the stage individual children have reached. Some top-performing P3 pupils could be better suited to a P4 class, it is suggested.
Dr Wilson said that international evidence was conflicting. Studies in the United States contend there is a positive impact from cross-age grouping but the verdict in Holland is that composite classes are "simply no worse and simply no better".
In Scotland, the evidence from the Assessment of Achievement Programme showed that the highest performing children come from composite classes.
"But we cannot disaggregate the effect of being in a composite class from being in a small school, or even what the teachers are doing within the class in the small school. What we suspect is that a lot of it is down to the rural community, the ethos of the school and being in a small school.
The data is not strong enough," Dr Wilson said.
The highest performing children, coincidentally, were from Gaelic-medium classes, which were usually composites.
Dr Wilson said headteachers think that composites are more difficult to teach and teachers believe that they add to their workload, planning and stress.
Reports also indicated that some teachers divide children into age groups in a small class, even when it is a composite.