The effects of use and non-use of a particular electronic formative evaluation tool, the Program Evaluation Analysis Computer (PEAC) system on the cognitive, affective and evaluative aspects of college students' perception of an educational television (ETV) program emphasizing cognitive learning were investigated. A specific real-time evaluation question was used for Study 1 and a more global question for Study 2. Scores for theshort answer test assessing content recall in Study 2 increased from pretest to posttest for non-users and decreased for users, suggesting that PEAC usage hindered recall involving comprehension. No other PEAC effects were found. The results suggest that while electronic evaluation techniques may be well suited for evaluations involving affective aspects of ETV programs, they may not provide useful evaluation information when objectives emphasize cognitive change, and may actually interfere with some types of learning. Formative evaluation factors Influencing categories of learning are discussed. Resume: lei, nous avons etudle les effets qu'ont 1'utlllsation et la non-utilisation d'un outil d'evaluatlon formative specifique, le systeme devaluation de programme «PEAC», sur les aspects cognitifs, affectifs et evaluatifs de la perception qu'ont les etudiantsde niveau collegial du programmede television educative«ETV» qul porte sur I'apprentlssage cognitlf. Une question specifique en temps reel a et6 utlllsee dans I'etude No.l et une question plus globale dans I'etude No. 2. Dans I'etude No.2, les scores obtenus, en reponse aux questions a reponses breves portant sur 1'evaluatlon du contenu retenu, ont augmentedu pretest au post-test pour les nonutilisateurs mals dlminue pour les utilisateurs. Ce resultat laisse supposer que ACKNOWLEDGEMENT The first author gratefully acknowledges the support of the Social Sciences and Humanities Research Council of Canada and the Fonds pour la Formation des Chercheurs et 1'Aide a la Recherche of Quebec for initial support for this research. The authors would like to thank Robert M. Bernard for his helpful suggestions regarding the research design.