The concept of political support (Norris 1999; 2010) received considerable traction in legitimacy literature and has been widely used to test the levels of state legitimacy in many cross-country contexts (Dalton, 1998; Dahlberg et al. 2015; Both and Seligson, 2009; Linde, 2012). Yet, I argue that outside of the context of liberal democracies this operationalization of state legitimacy suffers from a serious lack of measurement validity because 1) people often conflate their support for institutions with support for current actors in office and 2) more often than previously assumed regard other, non-democracy principles as regime principles. As electoral autocracies are now the most common regime type in the world and together with closed autocracies are a home to 68% of the world population (Luhrman et al., 2018, Alizada, 2021), creating a valid measurement of state legitimacy in electoral autocracies seems to be a valuable task. I provide evidence for the first and second claim through the analysis of in-depth interviews on a sample of citizens from Serbia, a typical electoral autocracy. In addition, I show support for the second claim by running a confirmatory factor analysis using the Latin Barometer survey data. Based on the results I offer some suggestions for an improved measurement strategy of state legitimacy in the context of electoral autocracies.