Abstract
BACKGROUND: This study aimed to estimate the frequency and characterize the nature of methodological errors related to the selection, application, and interpretation of ANOVA and related tests in in vitro biomaterials studies published in the Journal of Clinical and Experimental Dentistry (JCED) over the last ten years. MATERIALS AND METHODS: A methodological review was conducted of original in vitro biomaterials articles published in JCED between 2016 and 2025. Studies reporting the use of ANOVA or derived procedures were included. Data were extracted on publication characteristics, statistical tests applied, verification of statistical assumptions, post hoc procedures, and statistical software. ANOVA use was classified as inappropriate when inconsistencies between reported and applied methods were identified, when assumption verification was inadequate or not specified, when post hoc procedures were inappropriate, or when statistical methods were incompletely reported. RESULTS: A total of 345 in vitro biomaterials studies were included. Although ANOVA was widely used, complete verification of statistical assumptions based on model residuals was uncommon. The most frequent methodological errors associated with misused ANOVA were failure to specify the post hoc test used (50.0%) and inappropriate application of post hoc procedures (31.6%). Additional issues included incompatible post hoc tests following global analyses and discrepancies between methods and results. CONCLUSIONS: Despite its central role in in vitro dental research, ANOVA is frequently misused or inadequately reported in biomaterials studies. Strengthening statistical training and enforcing clearer editorial standards may substantially improve the validity, transparency, and reproducibility of experimental dental research.