A study compared people's reactions to different types of assertions concerning social issues. For purposes of the study, assertion was defined as having three components: a reference to the issue itself, an attribute having evaluative implications and a verb phrase linking the issue with the attribute. Eight types of messages were composed concerning four social issues in order to assess the generalizability of any effects. The four issues were capital punishment, handgun control, nuclear weapons freeze, and establishing a universal national service program. To elicit salient attributes of the issues, 22 college students were asked to list both the positive and negative consequences of both having and not having each policy. The five most frequently mentioned positive and negative attributes were arranged into 32 different message combinations. Next, 174 college students received the information in booklets that were presented as edited essays by other students. The subjects were guided through the four sections of the booklet, each containing a page-long summary of the topic and followed by one of the eight essays. Participants were asked to respond to two questions on their attitudes towards the issue, and the seven category scale ratings were combined into a single index. Results indicated that although essay ratings tended to vary as a function of message type, the pattern of effects was not consistent across issues. Furthermore, the relative "goodness" of messages did not correspond with relative effectiveness. (CRH)