This study argues for the need to clarify the attitude construct in science education. After positing a conceptual definition for attitudes derived from the social psychology and persuasion literature, this study makes a demarcation between science attitudes and attitudes towards science. Based on this conceptual clarity, the study offers a constructive critique of five commonly used measures of attitudes in science education and offers a theoretically grounded alternative. Data gathered to test the alternative measure's validity and reliability is in the main consistent with strong validity and reliability claims for the new measure. Implications for science education are discussed.