To examine the relationship between participation in a research-based, academic first-year seminar (FYS) and the first-year achievement of STEM students, university data sets were collected from four cohorts (N = 2,543) of incoming STEM students of whom 581 were FYS participants over the four years. Chi-square tests of homogeneity were used to compare persistence, and Welch's t-test was used to assess differences in first-term GPA between FYS and non-FYS cohorts by gender, admission status, and first-generation status. Analyses revealed significant differences in one-year persistence (some as much as 13% higher) between FYS and non-FYS participants among all groups excluding females. Additionally, first-term GPA was significantly higher for all FYS STEM students (some as much as 0.4 points higher on a 4.0 scale). Due to these results, participation in a three-credit, research-based academic FYS in the first semester is recommended for STEM majors.