Both IEA and the Swedish National Agency for Education (Skolverket) stresses the importance of reading literacy as they recognize it as an individual right and a prerequisite to participate in society, to gain knowledge of school subject and being able to argue and participate in decision making situations (Skolverket, 2016; Mullis, Martin & Sainsbury, 2015). The definition of reading literacy has evolved over time from the simple view of reading as a decoding and word recognition to involve further skills as comprehension and meta-comprehension (Gough & Tunmer, 1986; Chall, 1989; Sefton-Green, Marsh, Erstad & Flewitt, 2016) and with the entrance of digital means in everyday and school life has grown further to involve digital and social skills as well (Coiro et al., 2008; Leu et al., 2009). The differences between paper-based and digital reading have been under close scrutiny in recent past indicating that the two modalities differ in several aspects besides the obvious format, such as the skills required, the processes needed and the consequences in comprehension and concentration (Delgado et al., 2018; Baron, 2017). Previous research has explored and confirmed the paper-based reading ability as a predictor of later achievement (Butler et al., 1985; Sparks et al., 2013;). Less is known though about the role digital reading ability as a predictor for later achievement. The present study will shed light to the possible effect of difference between digital reading assessment and paper-based reading assessment as a predictor of later reading achievement and general achievement. So far, most studies have focused on the individual factors that may influence achievement in reading. The present study examines the two modalities in comparison along with individual factors, such as SES and immigrant background.

IEA’s PIRLS and ePIRLS both measures reading literacy in paper-based and digital form respectively. In 2016, Swedish 10-year-olds completed both the paper-based and digital version of the reading assessment PIRLS. The correlation between students’ test scores amounted to 0.79, thus a high, yet not a perfect, relationship. The correlation revealed that there likely are differences between the two formats and that some students perform better in paper-based reading whereas others perform better on the digital test. IEA analyzed the differences that emerged between PIRLS and ePIRLS results for all countries. Swedish 10-year-olds were generally better in digital, but the difference was not statistically significant.

The main purpose of the current study is to investigate the predictive validity of the paper-based and digital reading scores in PIRLS by comparing these scores with the core subject marks and the overall marks in grade 6 for the students that participated in PIRLS 2016. A hypothesis is that the paper-based scores will be more closely related to the subject grade and overall grades, partly because paper-based reading was more prominent in schools at this point in time. Drawing from Clinton’s meta-analysis (2019) suggesting that paper-based reading is associated with deeper comprehension and with metacognitive processes (accuracy in prediction of achievement), we hypothesize that students who are relatively stronger in paper-based reading test to have higher marks in schoolyear 6 than students who scored better in digital reading. Against this background, two specific research questions are posed:

1. What is the relationship between paper-based and digital reading achievement in grade 4 and core subject marks and GPA in grade 6?
2. Is there a stronger relationship between grades in Grade 6 and the reading results in PIRLS for students who are relatively stronger in paper-based reading?

The present study uses Swedish register data together with PIRLS and ePIRLS scores. Sweden participated in both PIRLS and ePIRLS IN 2016 with approximately 4000 students. Through the Swedish unique social security number (personnummer) that was collected from PIRLS students, a number of register variables can be related to their results in PIRLS and ePIRLS. The first subject marks are collected in grade 6, thus for PIRLS students in 2018. Compulsory school in Sweden is 9 years and it is divided into three periods: grades 1-3, 4-6 and 7-9. The evaluating system up to grade 6 is mostly descriptive. By the end of fall semester in grade 6, students are appointed with grades according to their performance in different courses. There are 5 passing grades (A-E), while F indicates that the student has not reached the lowest benchmark for a passing grade. The courses that students attend are 16 (Swedish, mathematics, english, biology, physics, chemistry, technology, geography, history, religion, social studies, visual arts, music, handicrafts, physical education and home economics). Out of them Swedish, mathematics and English are the core subjects. Parental education is used as an indicator of SES and language use at home is used as an indicator of immigrant background. We will conduct hierarchical regression analysis to investigate if and to what extend can the difference between PIRLS and ePIRLS scores predict the overall marks and the core subject marks.

Swedish students had an average of 559 points on ePIRLS and 555 points on PIRLS. We calculated correlations which indicated a strong -yet not perfect- relationship between paper-based reading and digital reading, about .80. This suggests that most of those who perform well in one format also perform well in the other. The average score in Swedish language was 14 (0-20), 13.4 (0-20) in maths, 15 (0-20) in English and 205 (0-320) in GPA. In the first step, we calculated the difference between the two test scores (PVepirls –PVpirls). The correlation between the difference of the two tests and Swedish language grade was not significant, while the correlation between the difference and math grades and English language grades was moderate (.52 and .68 respectively) and weak to the GPA (.35). Preliminary results showed that there is a relationship between grades in mathematics and English language and the difference between PIRLS and ePIRLS scores, indicating that students that scored better in the digital assessment achieved higher marks in mathematics and English language. However, the effect on the Swedish or the overall marks was not statistically significant. Having in mind the considerable predominance of English language online (Meurant, 2009), the present results comes in agreement with previous research indicating the relationship between digital literacy and English language (Alakrash et al., 2021). The present study also confirms previous research that found a pattern of digital literacy related with achievement in mathematics (Hu et al., 2018; Skryabin et al., 2015). Nevertheless, the relationship between digital literacy and Swedish language that was found not significant needs further exploration. A possible explanation can be the fact that Swedish language teachers do not focus on digital literacy enough, as well as the fact that the most common language online is by far English.

Alakrash, H., Razak, N. A., & Krish, P. (2021). Social network sites in learning english; an invstigation on attitudes, digital literacy and usage. Linguistica Antverpiensia, 2021(1), 26–43. Baron, N. S. (2017). Reading in a digital age. Phi Delta Kappan, 99(2), 15–20. Butler, S. R., Marsh, H. W., Sheppard, M. J., & Sheppard, J. L. (n.d.). Seven-Year Longitudinal Study of the Early Prediction of Reading Achievement. Chall, J. S. (1989). “Learning to Read: The Great Debate” 20 Years Later: A Response to ‘Debunking the Great Phonics Myth.’ The Phi Delta Kappan, 70(7), 521–538. http://www.jstor.org/stable/20403953 Clinton, V. (2019). Reading from paper compared to screens: A systematic review and meta‐analysis. Journal of Research in Reading, 42(2), 288–325. https://doi.org/10.1111/1467-9817.12269 Coiro, J. (Ed.). (2008). Handbook of research on new literacies. Lawrence Erlbaum Associates/Taylor & Francis Group. Delgado, P., Vargas, C., Ackerman, R., & Salmerón, L. (2018). Don’t throw away your printed books: A meta-analysis on the effects of reading media on reading comprehension. Educational Research Review, 25, 23–38. https://doi.org/10.1016/j.edurev.2018.09.003 Gough, P. B., & Tunmer, W. E. (1986). Decoding, Reading, and Reading Disability. Remedial and Special Education, 7(1), 6–10. https://doi.org/10.1177/074193258600700104 Hu, X., Gong, Y., Lai, C., & Leung, F. K. S. (2018). The relationship between ICT and student literacy in mathematics, reading, and science across 44 countries: A multilevel analysis. Computers and Education, 125, 1–13. Leu, D. J., O’Byrne, W. I., Zawilinski, L., McVerry, J. G., & Everett-Cacopardo, H. (2009). Comments on Greenhow, Robelia, and Hughes: Expanding the New Literacies Conversation. Educational Researcher, 38(4), 264–269. https://doi.org/10.3102/0013189X09336676 Meurant, R. C. (2009). The Significance of Second Language Digital Literacy Why English-Language Digital Literacy Skills Should be Fostered in Korea. 2009 Fourth International Conference on Computer Sciences and Convergence Information Technology, 369–374. https://doi.org/10.1109/ICCIT.2009.192 Mullis, I. V. S., Martin, M. O., & Sainsbury, M. (2016). PIRLS 2016 Reading Framework. Sefton-Green, J., Marsh, J., Erstad, O., & Flewitt, R. (n.d.). Establishing a Research Agenda for the Digital Literacy Practices of Young Children. 37. Skolverket (2016). Att läsa och förstå. Stockholm: Skolverket. Skryabin, M., Zhang, J., Liu, L., & Zhang, D. (2015). How the ICT development level and usage influence student achievement in reading, mathematics, and science. Computers and Education, 85, 49–58. Sparks, R. L., Patton, J., & Murdoch, A. (2014). Early reading success and its relationship to reading achievement and reading volume: Replication of ‘10 years later’. Reading and Writing, 27(1), 189–211. https://doi.org/10.1007/s11145-013-9439-2