The theory underlying the measurement of intellectual growth by the Peabody Picture Vocabulary Test (PPVT) and its congruence with the objectives of the Appalachia Educational Laboratory (AEL) Early Childhood Education Program is explored. The PPVT was administered to a sample of 160 3- and 4-year-old children in three treatment groups: (1) Package (Mobile Classroom, TV, and Home Visitor), (2) TV-Home Visitor (HV), (3) TV Only, and a control group. Data are analyzed by a three-way analysis of variance and an analysis of covariance procedure. Because of the highly specific nature of the test items on the PPVT, it is not likely that it reflects general program effects as well as the more broadly based instrument in a test battery. Two groups of children (Package and TV-HV) scored near the national mean (50th percentile) in IQ and two groups (TV Only and Control) scored near the 40th percentile when compared to the national sample. The lack of overall deficit indicates that many of the children have an adequate vocabulary level. Raw score analysis suggests the probability of a treatment effect in the verbal area which is reflected by the PPVT and which favors the Package and TV-HV groups. A summary of the AEL Early Childhood Program is available as PS 004 889. (Author/NH)