This study aims to investigate how Turkish students' involvement in information and communication technologies (ICT) predicts their math and science performance in the 2018 Programme for International Student Assessment (PISA) test. The research also tests demographic variables including socioeconomic status (SES) and gender as covariates. The data were examined through two-step hierarchical regression analyses. Regarding demographics, SES revealed a significant positive contribution to the prediction of math and science performance, whereas gender failed to make a significant contribution. Additionally, after controlling demographics, ICT availability at home significantly and negatively predicted student math and science performance, whilst ICT availability at school was not found to be significantly contributory. Regarding ICT use variables, all constructs significantly predicted student math and science performances, but some negatively contributed to the model, whereas others positively contributed. Amongst ICT attitude variables, all constructs made a significantly positive or negative contribution to predicting both math and science performance, with one exception. Only perceived autonomy in ICT use failed to significantly contribute to the prediction of math performance. Based on the findings, it was concluded that persistent involvement in ICT tends to be related to decreased math and science performance. We propose that policymakers and practitioners drop the myth that ICT use results in better achievement under any circumstances and should therefore refrain from integrating ICT without first undertaking careful planning.