<p style="text-align:justify">This study aimed to compare and examine the effectiveness of interactive STEM learning and paper-and-pencil STEM learning in terms of mathematical literacy skills of elementary school students. This research is of a quasi-experimental type with a non-equivalent pretest-posttest control group design. Sampling was carried out on the elementary school populations in Bengkulu and South Sumatra Provinces in two stages. In the first stage, schools in rural and urban areas were selected, and in the second, classes in each school were randomly selected. The selected sample consisted of fifth-grade students of the Public Elementary School of Terawas, Musi Rawas, with an experimental class A (n = 20) and an experimental class B (n = 19), as well as fifth-grade students of the Public Elementary School of Bengkulu City, with an experimental class A (n = 25) and an experimental class B (n = 22). Data collection was conducted using mathematical literacy skills tests in reference to the PISA and Minimum Competency Assessment (level 1–3). Data analysis was performed using descriptive and inferential statistics; it employed an independent t-test for the comparative testing and an N-gain test for testing the effectiveness of STEM learning. The results showed that there were differences in math literacy skills between interactive STEM and paper-and-pencil STEM for students in urban schools, but not significantly different for students in rural schools. General STEM learning was effective in increasing the literacy of elementary school students, and interactive STEM in particular demonstrated the highest level of effectiveness in the urban school.</p>