The purpose of this study was to determine whether the direct instruction and guided practice of question-generating as a thinking skill was an influence on students' higher-level thinking skills. Additionally, this study investigated the levels of questions generated by the students throughout the five week study. Thirty sixth grade students in a Southwestern urban public elementary school were instructed daily in the use of Bloom's taxonomy as a guide in designing and composing questions. Higher levels of the taxonomy were emphasized for higher-level question generating (analysis, synthesis, and evaluation). A pre- and posttest measuring cognitive abilities was given to determine the level of the students' higher-level thinking skills. Scores were analyzed to determine the influence of the treatment. A significance difference was found between means by use of a t-test for correlated samples. Student questions generated throughout the study were coded according to Bloom's taxonomy levels by independent coders with a reliability of .93. Qualitative matrices were developed to display the levels, numbers, and percentages of the questions generated. A significant increase of higher-level questions were generated between week one (3.8%) and week five (80.5%). A shift occurred in the fourth week, with a higher percentage (61.7) of higher-level questions generated than lower-level (38.3). The results indicate support for the proposition that the direct instruction and guided practice in question generating as a thinking skill influenced students' higher-level thinking skills. The analysis of the question levels suggest support for recommendation that autonomy follows with mastery of instruction and guided practice in the thinking skill before using that skill in a new content area. Guided practice in this study was in the familiar content area of reading.