The pretest in evaluations of Head Start and related programs : A methodological study

The issue studied was the timing of the pretest in educational program evaluations. An attempt was made to determine whether pretests should occur on the very first day of the program, or whether they may be delayed, as has been done with some reported Head Start evaluations. The Slosson Intelligence Test was given to a random sample of children enrolling in Head Start on the opening day, followed by a posttest 3 wks. later. Control children were tested at comparable times. The E group as a whole gained slightly, but not significantly, more than the C group. However, the E boys gained significantly more than the E girls, whereas C boys gained somewhat less than C girls. The results suggested that the opening day of a program is to be preferred over later pretest dates, and that a pretest delayed beyond the second week would find subjects already modified by their contact with the program. Practice effect between pretest and posttest did not appear to be a problem, but some coaching of C subjects by their parents occurred between tests, since parents were allowed to witness the C pretests. Several possible causes for the surprising differences in IQ gains between E boys and girls were discussed in conclusion.