Abstract
Normality testing is an essential part of statistical analysis to determine whether observed data are normally distributed or not. This study aims to document and analyze various types of errors frequently made by non-mathematics students when conducting normality tests. Furthermore, this descriptive qualitative research involved 32 second-year non-mathematics students as participants, who were given tasks requiring them to perform normality tests on datasets. The research findings indicate that the most common type of error made by non-mathematics students in conducting normality tests is encoding error, accounting for 40.7% of errors. This occurs when students mistakenly compare decimal values between Lcalc and Ltable. Consequently, the conclusions drawn from these normality tests may be inaccurate. Additionally, other types of errors identified include reading error, comprehension error, transformation error, and process skills error.