Student background determinants of reading achievement in Italy. A quantile regression analysis

• Using PISA data we study factors that affect reading achievement of Italian students.
• We model data throughout a quantile regression (QR) setting.
• QR findings are compared with OLS model results.
• Policy implications are discussed to highlight the usefulness of QR approach.

In recent years determinants of students’ achievement has received much attention. Empirical studies have found that students’ characteristics, family background, school attended, and regional residence are major factors affecting student performance. In this paper, we analyze the 2009 OECD-PISA (spell PISA) survey to examine individual background characteristics influencing the reading achievement of Italian 15 years-old students using the quantile regression (QR) approach. The QR approach allows researchers to analyze changes in size and direction of predictor estimates on student performance across the entire distribution of reading achievement scores. Results indicate significant effects of predictors on reading achievement operating differently across quantiles, suggesting different pathways to achievement for low and high performing readers. In particular, some family background predictors (parental education, computer availability at home, and availability of a desk for homework at home), the school program attended and, the region of student residence play important but differing role for low and high performing readers. For example, parental education shows a positive effect on student reading, academic (general) programs perform better than vocational or technical, and Northern regions perform better than Center-Southern ones, with differentiated effects along the distribution of students’ reading scores. These findings should be carefully considered by policymakers when outlining strategies to enhance student performance at all levels along the reading continuum of low and high scores.