The purpose of this study was to compare the effects of different types of questioning interventions on students' reading comprehension. Fourth-grade students (n = 246) were identified as struggling, average, or good readers and assigned randomly within school to one of three questioning interventions: two inferential conditions (Causal or General) or one literal condition ("Who, What, Where, When" or W-questioning). Teachers delivered the interventions for 20-30 min, 2-4 times per week, for 8-10 weeks. All readers made reliable pre- to posttest comprehension gains as measured by story recall (ps less than 0.001 to 0.04). Differential effects for intervention were found between two subgroups of struggling comprehenders-elaborators and paraphrasers. Elaborators benefited more than paraphrasers from Causal questioning (d = 0.86) whereas paraphrasers benefited more than elaborators from General questioning (d = 1.46). These findings suggest that identifying subgroups is important in developing and evaluating the effectiveness of reading comprehension interventions. (Contains 1 figure and 4 tables.)