Almost 217 million secondary school students (60% of the world's adolescents) do not reach minimum levels in reading proficiency at the end of secondary school, according to objective 4.1 of the UN's Sustainable Development Goals. Therefore, the early and efficient identification of this disadvantage and implementation of remedial strategies is critical for economies. In 2018, the Programme for International Student Assessment (PISA) assessed the reading skills of 15-year-old students in 80 countries and economies. This work introduces a methodology that uses PISA's data to build logistic regression models to identify the main factors contributing to students' underperforming reading skills. Results showed that socioeconomic status (SES), metacognition strategies, Information and Communication Technology (ICT) skills, and student-teacher relationships are the most important contributors to low reading abilities.