Normal curve equivalent achievement gains estimates were compared with RMC Title I evaluation Models A1 and B1. The comparison focused upon the amount of bias introduced by Model A1 when its underlying assumptions were violated. The model assumes, first, that the local school population is accurately represented by the national norm group; and secondly, that the percentile standing of the treatment group on the pretest remained unchanged on the posttest in the abscence of treatment effect. Data were from a Model B1 evaluation--kindergarten children were pre- and posttested with the Aural Comprehension subtest of the Stanford Early School Achievement Test, while the first and second grade children were pre- and posttested with the Woodcock Reading Mastery Test. Comparison of the three model B1 gain estimates (unadjusted, covariance adjusted, and principal-axis adjusted), illustrated the wide variation between estimates. For kindergarten groups, the adjustments were relatively modest, but for grades one and two, the adjustments caused a dramatic shift from negative to positive estimates. Model B1 (adjusted) provided much lower estimates for kindergarten and much higher estimates for grades one and two than did Model A1. Questions were raised concerning representativeness of the norm group and validity of the equipercentile assumption. (Author/CP)