The purpose of this study was to compare the relative effectiveness of five instructional interventions which were designed to correct a size-related science misconception in a group of first graders. The particular misconception chosen for the study is the belief that larger magnets are always of greater strength than smaller magnets. These interventions consisted of: a. a demonstration lesson, b. a hands-on lesson, c. a verbal statements lesson, d. a demonstration-plus-verbal statements lesson, and e. a hands-on-plus-verbal statements lesson. At the beginning of each magnet lesson, students were first exposed to evidence contradicting the size-related magnet misconception. This was followed by the introduction of cognitive conflict via the use of a small weak rectangular magnet and a larger strong rectangular magnet to pick up paper clips. Finally, students interacted with two (apparently) identical rectangular magnets of clearly differing strengths. The second major component of each intervention was the use of iron filings and a magnet to "show" lines of force. The purpose of this last activity was to give the students some information about magnets that would assist them in accommodating the events witnessed in the first part of the intervention. Subjects were tested three days before the treatment, one day after treatment to determine change of knowledge effect, then six weeks after treatment as a check for knowledge retention. Six subjects were randomly chosen from each treatment group to be interviewed using a format based upon Novak's Interview-about-Instances (1984) prior to the instruction, and on two occasions (one day, and six weeks) after the instruction. It was hypothesized that a demonstration treatment would result in the highest frequency of students who received a score of 100% on four misconception-related items on the post-test. It was also hypothesized that the demonstration treatment would result in the greatest retention. Analysis of both test scores and interview data indicates that, although there is strong support for the corrective properties of a demonstration which directly confronts the misconception that a necessary direct relation ship exists between magnetic strength and magnet size, a demonstration alone is not more effective than all of the remaining treatments in achieving change of knowledge or retention. Therefore, there is a lack of support for both research hypotheses. Both treatments containing demonstrations, however, were more effective in achieving correction of the size-related misconception than the treatments consisting of a hands-on treatment alone and verbal statements alone.