This chapter studies quality control methods for real-time kinematic positioning, introducing both robust estimation and the detection, identification, adaptation method for outlier management. Outliers in Global Navigation Satellite System (GNSS) data necessitate specialized processing to mitigate their biasing effects on least-squares estimators. Two principal outlier detection frameworks are outlined, categorized by whether outliers follow a non-stochastic (mean-shift model) or stochastic (variance-inflation model). It also emphasizes the importance of realistic stochastic models in statistical reliability testing, which can minimize false alarms and enhance detection accuracy. Proper modeling of physical correlations, such as those related to satellite elevation and observation time, is shown to significantly improve the reliability of GNSS positioning tests.