This study compares the Beta-value and M-value methods for quantifying methylation levels using microarray analysis. Both methods are used to measure methylation levels, but their relationships and strengths have not been thoroughly studied. The research demonstrates that the relationship between Beta-value and M-value is a Logit transformation. The Beta-value method has severe heteroscedasticity for highly methylated or unmethylated CpG sites, while the M-value method is more statistically valid for differential methylation analysis. The study evaluates the performance of both methods in identifying differentially methylated CpG sites using a methylation titration experiment. Results show that the M-value method performs better in terms of Detection Rate (DR) and True Positive Rate (TPR) for both highly methylated and unmethylated CpG sites. Imposing a minimum threshold of difference improves the performance of the M-value method but not the Beta-value method. The study provides guidance for selecting the threshold of methylation differences. The Beta-value has a more intuitive biological interpretation, but the M-value is more statistically valid for differential analysis. Therefore, the study recommends using the M-value method for differential methylation analysis and including the Beta-value statistics when reporting results to investigators. The M-value method is more appropriate for homoscedastic assumptions in most statistical models used for microarray analysis. The study also provides guidance for selecting the difference thresholds for the M-value method, suggesting a range between 0.4 and 1.4 (or 1.32 to 2.64 in non-log scale). The Beta-value method is less suitable due to its severe heteroscedasticity in the low and high methylation ranges. The study concludes that the M-value method is more statistically valid for differential methylation analysis and is recommended for use in such studies.This study compares the Beta-value and M-value methods for quantifying methylation levels using microarray analysis. Both methods are used to measure methylation levels, but their relationships and strengths have not been thoroughly studied. The research demonstrates that the relationship between Beta-value and M-value is a Logit transformation. The Beta-value method has severe heteroscedasticity for highly methylated or unmethylated CpG sites, while the M-value method is more statistically valid for differential methylation analysis. The study evaluates the performance of both methods in identifying differentially methylated CpG sites using a methylation titration experiment. Results show that the M-value method performs better in terms of Detection Rate (DR) and True Positive Rate (TPR) for both highly methylated and unmethylated CpG sites. Imposing a minimum threshold of difference improves the performance of the M-value method but not the Beta-value method. The study provides guidance for selecting the threshold of methylation differences. The Beta-value has a more intuitive biological interpretation, but the M-value is more statistically valid for differential analysis. Therefore, the study recommends using the M-value method for differential methylation analysis and including the Beta-value statistics when reporting results to investigators. The M-value method is more appropriate for homoscedastic assumptions in most statistical models used for microarray analysis. The study also provides guidance for selecting the difference thresholds for the M-value method, suggesting a range between 0.4 and 1.4 (or 1.32 to 2.64 in non-log scale). The Beta-value method is less suitable due to its severe heteroscedasticity in the low and high methylation ranges. The study concludes that the M-value method is more statistically valid for differential methylation analysis and is recommended for use in such studies.