This experiment asked Clinical Psychology and Counseling graduate students to evaluate the mental health of strangers using attitude statements and behavior statements purported to refer to fictitious strangers. Holding behavior statements constant, strangers displaying attitudes dissimilar to the students' were rated as significantly more disturbed than attitudinally similar strangers. Holding similarity constant, strangers described as demonstrating severely disturbed behavior were perceived to be more disturbed. Students reported more confidence in the accuracy of their evaluations of similar strangers. No relationship was found between the ratings given to strangers and the student's major, the degree pursued, or the coursework completed. (Author)