Abstract
Subjects were exposed to one of three survey versions differing only in question order. Those who first faced difficult political knowledge questions reported significantly lower levels of both political interest and news attention than those who did not first face the knowledge test. However, when the knowledge questions and the interest and attention questions were separated by a “buffer” item that could serve as an excuse for poor knowledge, self-assessed interest and attention were less depressed. Characteristics of survey questions that may make them particularly susceptible to these types of question-order context effects are discussed and strategies for dealing with such effects are noted.
Get full access to this article
View all access options for this article.