How Does Engagement in Reading Predict Digital Reading Proficiency Among 15-Year-Olds?

• In Event: Perspectives on PISA (Programme for International Student Assessment) Reading

Abstract

Objectives and research questions
The OECD report PISA 2009 Results: Learning to Learn (OECD, 2010) investigated in depth how engagement in reading relates to print reading. This paper will focus on the relationships between reading practices, attitudes and digital reading proficiency.

The following research questions will be addressed:
- do students who enjoy reading and read a diversity of print material read better online?
- do students who read more frequently on line read better on line? Which kind of online reading activities are more positively related to digital proficiency?
Methodology
The PISA 2009 database and data from the 19 countries that participated in the digital reading option were used.
Digital reading scores (plausible values) were treated as the dependent variable in the analysis. Four composite indices were used as independent variables: an index of enjoyment of reading (11 items), an index of diversity and frequency of print material read (five items), and two indices of online reading practices (six items).
Factorial analyses were performed in order to build the online composite indices. Regression analyses were then performed to estimate the percentage of variance in digital reading performance explained by the four indices; the four indices were then incorporated in a broader regression analysis aimed at identifying the best predictors of digital reading.
The following indicators were used for reporting the results: mean index for all students by country, by gender and magnitude of the gender difference; mean digital reading score by national quartile of the indices; change in the digital reading score per unit of the indices; and percentage of explained variance.
Results
Digital reading performance is positively related to enjoyment of reading (14% of variance of the digital reading performance explained) and to diversity of print reading (3% of variance explained).
The amount of variance of the digital reading score explained by online practices is higher (6%) than it is for print reading (3%). Factorial analyses performed on the online reading items show that two separate kinds of online reading activities can be identified: searching information online activities (for example, reading news, and using a dictionary); and social activities (reading emails and chat). While searching information online is positively related to digital proficiency (on average, 8% of student variance explained), social activities on-line are weakly related to digital reading proficiency (only 1% of student variance explained). Besides, relationships with digital reading are curvilinear: students who read very little and very often on line perform worse on the digital reading tasks, while students moderately engaged in online reading perform the best.
The index of online social activities and the index of online searching-information activities are two of the variables that explain the difference in performance between print and digital reading. When students tend to engage in more social activities and searching-information activities on line, they tend to perform better in digital reading than students who do not, even when all of these students are similarly proficient in print reading.

Authors

• Dominique Marie Lafontaine, University of Liege

• Ariane Baye, Université de Liège