The independence of positive and negative affect has been heralded as a major and counterintuitive finding in the psychology of mood and emotion. Still, other findings support the older view that positive and negative fall at opposite ends of a single bipolar continuum. Independence versus bipolarity can be reconciled by considering (a) the activation dimension of affect, (b) random and systematic measurement error, and (c) how items are selected to achieve an appropriate test of bipolarity. In 3 studies of self-reported current affect, random and systematic error were controlled through multiformat measurement and confirmatory factor analysis. Valence was found to be independent of activation, positive affect the bipolar opposite of negative affect, and deactivation the bipolar opposite of activation. The dimensions underlying D. Watson, L. A. Clark, and A. Tellegen's (1988) Positive and Negative Affect schedule were accounted for by the valence and activation dimensions.