This article proposes a robust test for weak instruments in linear instrumental variables (IV) regression, which is valid under heteroscedasticity, autocorrelation, and clustering. The test statistic is a scaled version of the nonrobust first-stage \(F\) statistic. Instruments are considered weak if the two-stage least squares (TSLS) or limited information maximum likelihood (LIML) Nagar bias is large relative to a benchmark. The authors apply their test to estimate the elasticity of intertemporal substitution (EIS), finding that the test cannot reject the null of weak instruments in more countries than the test proposed by Stock and Yogo (2005). The article includes a detailed model and testing procedure, asymptotic distributions, and an empirical application. The test provides a new tool for empirical researchers to assess instrument strength in various data contexts.This article proposes a robust test for weak instruments in linear instrumental variables (IV) regression, which is valid under heteroscedasticity, autocorrelation, and clustering. The test statistic is a scaled version of the nonrobust first-stage \(F\) statistic. Instruments are considered weak if the two-stage least squares (TSLS) or limited information maximum likelihood (LIML) Nagar bias is large relative to a benchmark. The authors apply their test to estimate the elasticity of intertemporal substitution (EIS), finding that the test cannot reject the null of weak instruments in more countries than the test proposed by Stock and Yogo (2005). The article includes a detailed model and testing procedure, asymptotic distributions, and an empirical application. The test provides a new tool for empirical researchers to assess instrument strength in various data contexts.