Purpose: This study aims to discern and analyze statistical differences in Haralick textures within lung X-ray images, specifically comparing patients diagnosed with COVID-19 against those with common pneumonia. Methods: This study included three datasets (Covid-19, Pneumonia, Control) each with 194 x-ray imagens. For the extraction of Haralick textures, the MATLAB MathWorks, Natick, MA, USA) was employed. To assess the normality of the data distribution, the Kolmogorov-Smirno test was employed. The adoption of the Kruskal-Wallis test in this analysis stems from the recognition that the data exhibit a non-normal distribution. Results: The Haralick texture Features that demonstrated significant discriminative capability among the three groups were Contrast, Entropy, Sum of Entropy, Difference Entropy, and Informational Measure of Correlation 1 and 2. Conclusion: The findings in this study advocate for a comprehensive and iterative approach to texture analysis methodologies, integrating both informative Features and complementary analytical techniques.