Putting infants in classes of under 20 has long-term benefits - but only if they stay in them for at least three years. David Budge reports from the American Educational Research Conference in New Orleans.

INFANTS must spend at least three years in small classes to gain lasting benefits, new research suggests.

Although even one year in a class of fewer than 20 pupils provides immediate educational gains, these disappear by the age of nine, American researchers found.

Children who spend two years in small kindergarten or infant classes also lose their lead fairly quickly. By the age of 11 they are still slightly ahead of their peers in reading. But by 13 they are on a par with youngsters taught in “regular” classes (22 to 26 pupils).

Children lucky enough to be in small infant classes for three years outperform their schoolmates at least until the age of 13. Those who spend their first four years in small classes have an even greater lead at 13 in reading, maths and science. On average, they are still about nine months ahead of their schoolmates by this stage, the study suggests.

“Despite the diminution of some of the (class-size) effects ... all differences in test scores remain statistically significant ... fully five years afterwards,” the researchers say. “Few educational interventions have demonstrated this degree of longevity.”

The findings emerged from a re-analysis of statistics from the Tennessee STAR Project - te most famous class-size study ever mounted. The 12,000 children involved in the 1980s study were randomly assigned to small classes, regular classes or regular classes with a full-time teaching assistant. (In Britain, of course, an infant class of 22-26 would be regarded as relatively small.)

The new study, conducted by academics at the universities of New York State and Eastern Michigan, confirms that children in classes with teaching assistants did no better academically than those without them.

It also underlines how important it is to start early in a small class. “In general, the later the starting point, the more years of participation are required to improve students’ learning,” the researchers say.

But the study’s authors accept many questions about class sizes remain unanswered. Why are the benefits of small classes greater in some schools than others? Would small classes in middle and high schools be beneficial?

“Even more important ... how can teachers take advantage of the opportunities a small class provides, to maximise the impact on pupil learning and behaviour?” they ask. “Answers to this question are virtually non-existent at present even though over half the states and many districts across the US are undertaking class-size reduction initiatives.”

“The enduring effects of small classes”, by Jeremy D Finn and Susan B Gerber, State University of New York at Buffalo, Charles M Achilles, Eastern Michigan University, and Jayne Boyd-Zaharias, HEROS Inc.