This paper challenges the assumption that meaning is in some way magically encoded within texts and argues that questions, when they are used simply as tools to lead readers to a single meaning, are inadequate as a means of reading comprehension assessment. The paper then goes on to describe one of a series of micro case studies that explored alternative strategies for assessment and what happens when assessment is used as an integral part of teaching and learning. Students in two fifth-grade classes in the same school in Australia were introduced to a picture book. The first two pages were read out loud, and then students read silently the next six pages of the text with two interruptions at separate points in the story. In one class, students were given a set of written questions to complete at each interruption point. The other class was asked to complete a sketch of what they thought would happen next (the "Sketch to Stretch" strategy). Results indicated that: (1) use of Sketch to Stretch provided a different type of data on which to make predictions about student understanding of text; (2) alternative non-question techniques can simultaneously fulfill the dual tasks of assessment and teaching; (3) alternative strategies, like drawing, may have the potential to help readers construct qualitatively different mental texts; (4) experimental strategies like Sketch to Stretch provided greater scope for the sharing of responses between readers; and (5) students experienced greater involvement, interest, and concentration in the experimental lessons than in traditional lessons. Findings of this and the other micro case studies give legitimacy to a variety of "non-question" comprehension strategies as vehicles for assessment and learning. (Four tables of data and four figures of student sketches and of other alternative assessment strategies are included; 15 references are attached.) (RS)