Recently economists looked at the relationship between pupils' month of birth and their school performance. It was found that among fourth graders, the oldest children scored on math and reading somewhere between four and twelve percentile points better than the youngest children. That can be explained as a "huge effect." It means that if you take two intellectually \_21\_ fourth graders with birthdays at opposite ends of the cutoff date, the older student could score in the eightieth percentile, \_22\_ the younger one could score in the sixty-eighth percentile. That's the difference between \_23\_ for a gifted program or not. The economists said, "We do ability grouping early on in childhood. We have advanced reading groups and advanced math groups. So, early on, if we look at young kids, in kindergarten and first grade, the teachers are \_24\_ maturity with ability. And they put the old kids in the advanced stream, where they learn better skills; and the next year, because they are in the higher groups, they do even better; and in the next year, the same thing happens, and they do even better again. The only country we don't see this going on is Denmark." Denmark waits to make selection decisions until maturity differences by age have \_25\_. This is because they have a national policy where they have no ability grouping until the age of ten.