A research project was conducted to see if students given direct instruction in recognizing and asking literal, interpretive, and critical questions would ask better questions about material they read than students who were not given the same instruction. This premise, based on an extensive search of the literature on questioning and comprehension, was first tested in a pilot study at a university-sponsored elementary school. Subjects were 40 fifth and sixth grade students divided by grade into two control and two experimental groups of 10 each. To evaluate their question generating ability, all were administered a pretest consisting of reading a story and writing down questions they would ask other students about it. These questions were reviewed and categorized in increasing degree of difficulty as having answers that were (1) explicit in the text, (2) implicit in the text, or (3) needed to be supplied from the reader's background. The two experimental groups were then given lessons in how to identify and generate questions using three different teaching strategies. Finally, both experimental and control groups were given a posttest using the pretest story, and asked to write questions about it. Again these were reviewed and categorized. Both experimental groups showed marked increases in the number of questions beyond the literal or textually explicit level, while the control groups showed little or no improvement in their questioning skills. All the teaching strategies seemed successful. Once these results were evaluated, a field study was undertaken for which only preliminary data are available. (RBW)