There has been considerable study of journals in psychology, both through the collection of ratings of excellence by journal readers (Koulack & Keselman, 1975, Mace & Warner, 1973) and through the study of journal citations (Buss & McDermott, 1976; Porter, 1976; Rushton & Roediger, 1978; White & White, 1977). There has been less study of educational journals, one of the most recent efforts being that of Luce and Johnson (1978) who collected ratings of psychological and educational journals from 682 members of the American Educational Research Association. The method of rating journals through the collection of readers' opinions has been criticized on several grounds (Boor, 1973; Buss & McDermott, 1976; Gynther, 1973; Hohn and Fine, 1973; Levin & Kratochwill, 1976), a principal objection being that reader preferences may be unrelated to more objective indicators of journal impact. Therefore, in order to provide additional, nonpreference data on the nature of educational journals, a comparative study of journal citations was made of the American Educational Research Journal (AERJ), the Review ofEducational Research (RER), and the Educational Researcher (ER). To our knowledge, a citation analysis of educational journals has been performed only twice previously, first to study the publication patterns of Canadian faculties of education (Arlin, 1978), and second in the study of a single journal, Adult Education (Boshier & Pickard, 1979). The present citation study was designed to discover (a) the nature of the citation pattern of the three journals (AERJ, RER, ER); that is, which is most cited, what is the average number of citations per article, etc., (b) the sources that most cited the literature appearing in these three journals; this gives an indication of the interjournal communication network within the educational literature, and (c) the citation characteristics of the most cited articles in each of these journals.