Abstract : Text type is proposed as a psychologically valid construct. Previous research has suggested that text type may play a role in a reader's comprehension and memory for a text. Two experiments were conducted to explore the psychological reality of text types. In the first experiment, students were required to sort twelve texts on the basis of their similarities. The resultant sortings were subjected to a clustering analysis. Despite the fact that other bases for grouping together texts existed--a number of pairs of semantically related texts of different types were included--text type emerged as a powerful determiner of group membership. In the second experiment, students listened to recorded texts and then tried to recall them. As was predicted, text type had a significant effect on recall, with stories being recalled more fully than were instructions or definitions.
A. Munro, Kathy A. Lutz, Lynn Gordon
Journal name not available for this finding