Beyond text comprehension: exploring items’ characteristics and their effect on foreign students’ disadvantage in mathematics

Abstract

National and International standardized assessments have shown sharp differences between native and foreign students’ performance in both text comprehension and mathematics. Based on the data gathered yearly by the INVALSI (Italian National Institute for the Evaluation of Educational Systems) at the national level, we performed a two-step analysis: (1) a regression analysis was first performed to quantify the effect of students’ abilities in reading and text comprehension (and their interaction with students’ citizenship status) on mathematical ability, given a set of covariates; then, (2) a differential item functioning (DIF) analysis was performed to explore the differences between the students depending on their citizenship status. The analysis, performed from a pseudo-longitudinal perspective to account for possible variations in the relationship between citizenship and mathematical performance by age, involved students in grades five (primary school), six, and eight (lower-secondary school). The regression analysis confirmed the relationship between text comprehension and mathematical ability, but the differential item functioning (DIF) analysis individuated additional items, showing statistically significant differential functioning by citizenship status regardless of whether high or low reading demands were placed on students.

KEYWORDS:

• Citizenship
• mathematical differential item functioning
• standardized assessment
• Rasch
• pseudo-longitudinal design

Acknowledgement

We confirm that this work is original and has not been published elsewhere, nor is it currently under consideration for publication elsewhere.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 ‘Ability’ is a technical term used within the framework of Rasch and IRT analysis to refer to a latent trait that an achievement test is intended to measure.

2 In line with the OECD and INVALSI, we categorized the students by citizenship status using the following criteria: Italian (i.e. student born in Italy with Italian parents), first-generation foreign (i.e. student not born in Italy with parents not born in Italy), and second-generation foreign (i.e., student born in Italy with parents not born in Italy).