The relationship between course ratings, student achievement, and the perceived attitude similarity of students and their instructors were explored with 645 students enrolled in 41 courses taught at English-language institutions in Montreal, Quebec. These institutions provide instruction intermediate between secondary school and the university. Two groups were evaluated. Three hundred and forty-nine students enrolled in 22 classes rated instructor effectiveness on a 28-item rating form and then rated their attitudes plus perceived instructor attitudes on an attitude scale adopted from Byrne. Another 296 students enrolled in 19 classes rated instructor effectiveness and their attitudes plus perceived instructor attitudes on an attitude scale adopted from Hofman and Kremer. Official end-of-term grades were also collected. A modest correlation between students' ratings and perceived attitude similarity was found, as was a modest relationship between similarity and course grades. These relationships were greatly reduced when the substantial influence of teachers on ratings and achievement was removed. The findings fail to support the claim that attitude similarity is a source of bias in ratings with any useful, practical effect. For neither of the attitude scales were student attitudes substantially related to student ratings of achievement independent of teacher effects. However, for the Hofman and Kremer scale, perceived instructor attitudes substantially affected ratings but not achievement. (Author/SW)