The basic assumption of brainstorming is that increased quantity of ideas results in increased generation as well as selection of creative ideas. Although previous research suggests that idea quantity correlates strongly with the number of good ideas generated, quantity has been found to be unrelated to the quality of selected ideas. This article reports the results of a brainstorming experiment aimed at increasing the average creativity of ideas and creative idea selection (rather than idea quantity). Problem scope (narrow vs. broad) and creativity instructions (emphasis on creativity vs. personal relevance) were manipulated. Results show that both narrow (vs. broad) problems and creativity (vs. relevance) instructions led to the generation of ideas that were more creative. However, only under creativity instructions did participants select more creative ideas.