The purpose of this paper is to critique Stallings' 1975 evaluation of Project Follow Through (PFT) in order to examine methodological issues inherent in evaluation research. Following a summary of the objectives, methods, and conclusions of Stallings' study, the critique focuses on four elements of research evaluation: internal validity, external validity, reliability, and relevance. Criticism of the PFT evaluation includes the following assertions: (1) Internal validity may be affected by a discrepancy in the number of variables used to describe program models and the number used to assess individual programs; (2) External validity may be questionable due to nonrandom assignment of the sample to treatment conditions; and (3) Reliability may be reduced as a result of confounding observers with program models, possible inconsistency of observations, and lack of control over extraneous variables. The relevance of Stallings' evaluation is perceived as being substantial in that it produced valuable implications for the development of educational programs and contributed methodological innovations for studying large educational projects. Concluding remarks review major findings of the PFT evaluation and, emphasizing the utility of the Classroom Observation Instrument, point out implications of Stallings' work for future evaluation research. (RH)