<p>Starches are often added to yogurts as thickeners and stabilizers, increasing the viscosity and firmness of the gel, however, in Brazil, legislation establishes its use in a maximum of 1%. This study aimed to identify the adulteration of commercial yogurts by adding starch using Differential Thermal Analysis (DTA) and Thermogravimetric Analysis (TGA) associated with chemometrics. Yogurt formulations with different starch concentrations were prepared as reference samples (n = 81), and 19 commercial samples were obtained to verify authenticity. The DTA and TGA curves showed that increasing the amount of starch in the yogurt affected the temperatures of the thermal events throughout the storage period. Parameters from these analyses were submitted to Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) and Multiple Linear Regression (MLR). The integrated application of these techniques for the detection of adulteration in commercial yogurts, something that is not yet widely used in routine food quality control, indicated that PCA was important to separate yogurt samples according to starch concentration, classifying the samples into two classes according to the addition of starch (≤ 1% and &gt; 1%). A reliable predictive model from PLS-DA and MLR was established to detect adulterations in commercial samples based on thermal data, with an accuracy of over 83%. From the models created, it was possible to identify that 36.84% of the samples of commercial yogurts had starch levels &gt; 1%. The results indicate that the combination of thermal analysis with chemometrics is an effective and promising tool for the detection of starch in yogurts.</p>