Fourth-graders in large city schools who reported using computers for science infrequently (once every few weeks or less, 139 and 140 respectively) had higher scores than students who reported using computers more frequently (127, 122, 112 respectively).