Despite evidence that the maze is a reliable measure of reading comprehension, existing research suggests potential problems with the manner in which maze probes are developed. Specifically, research suggests students may be able to respond accurately to a large portion of target words without having to comprehend what they are reading. The purpose of the current study was to examine differences in student accuracy as a function of (a) whether sentences were placed in the context of a passage and (b) target word type (function vs. content). Following the administration of an intact maze probe, 225 third-, fourth-, and fifth-grade students were administered an intact maze probe and a probe with sentences drawn randomly from three different maze probes (scrambled probes). Although statistical analyses suggested context improved student performance, results bring into question the degree to which the maze procedures measure comprehension beyond the sentence level.
School Psychology Review, v. 41, issue 3, p. 262-271
Available at: http://works.bepress.com/stacy-ann-january/31/