Unraveling Chemometric Analysis Performance: A SEM-PLS Investigation of Simple Technology Implementation, Statistical Understanding, and Outlier Detection

Abstract

This study aims to examine the effect of Simple Technology Implementation on spectral preprocessing accuracy using the Savitzky–Golay method, with data literacy as a moderating variable. A quantitative approach was employed using the Structural Equation Modeling–Partial Least Squares (SEM-PLS) method. Data were collected from 113 respondents using a five-point Likert scale questionnaire. The results indicate that Simple Technology Implementation has a positive effect on improving descriptive statistical understanding, although its impact on outlier detection effectiveness remains limited. The measurement model reveals that several constructs do not fully meet validity and reliability criteria, particularly the data literacy variable. These findings suggest that while user-friendly technology can enhance analytical efficiency, it is not sufficient to support more complex analytical tasks. Therefore, improving analytical performance requires both accessible technology and stronger user competencies in data literacy and statistical understanding