Numerous studies focus on the tightness–looseness (TL) dimension of national culture, reporting important implications for business and management. However, the main TL measure has generated a number of theoretical and empirical concerns. It likely reflects unfounded national auto-stereotypes, the conceptual boundary between TL and individualism–collectivism (IC) is blurred, respondents in different societies have different TL-related concepts, and TL yields mostly low and often insignificant correlations with its presumed correlates. Using the latest TL data from Eriksson, Gelfand, and associates, we find that some TL items are uncorrelated with aggregates of corresponding self-reports. TL is not predicted by its main presumed antecedents, and has marginal or no effects on its presumed main consequences, including all metanorms (reported frequencies of reactions to inappropriate behavior). Our results challenge the validity of measures of culture based on respondents’ generalized impressions of their compatriots. We suggest a revision of TL theory in which much of the TL concept can be subsumed under IC. We highlight the fact that a society can be tight in one domain and loose in another, which depends largely on its IC position.