Paper-based reading differs from digital reading with regards to several aspects. According to Chen (2017) the major difference between these two formats of reading is that the digital one is continuous and has no physical boundaries or visual anchors, for example pages (Chen, 2017). The range of elements that a reader can come across in a printed text are much fewer than the ones in the digital text. While a printed text can be accompanied by pictures, a digital one can contain much more elements besides images like sound effects, music, pop-up windows etc (Kress, 2003). The set of skills that these two types of reading literacy require are also diversified. The reader of a digital text needs to have higher resistance in distractions, because while in a printed text, distractions are external, meaning that they happen aside and away of the printed text, the reader of a digital text needs to avoid internal distractions as well. These are distractions that can pop up the at the same time while reading (Gilster, 1997). The strategies of approaching the two modalities can also differ. Schmar Dobler (2003) explains that digital reading demands the ability of evaluating the amount and variety of information offered in hypertexts such as links leading to additional information, definition or examples before picking up the appropriate. Research has also shown that printed text and digital text have different consequences for our memory, concentration and level of comprehension (Delgado et al., 2018; Baron, 2017; Mangen et al.,2019). According to Mangen et al. (2019) a reader can easier reconstruct temporal and chronological aspects from a printed text than a digital one. Baron et al. (2017) reported that the 91.8% of 429 university students in five countries have replied that they concentrate better when reading printed texts. As far as the comprehension is concerned, Delgado et al. (2018) in their meta-analysis conclude that paper-based reading has a positive effect in deep comprehension.

Furthermore, the skills required for each format of reading as well as the consequences in cognitive levels can differ. Researchers have examined both paper-based and digital literacy and the factors that can influence achievement on these two different formats. Factors like motivation, peer effect, reading strategies/habits, gender, age, length or genre of the text in test, teacher competence and teachers’ judgement of students’ reading achievement have been thoroughly studied across the years and countries and have been found to have an impact on reading achievement (Johansson et al., 2014; Rosén, 2006; Ammermueller & Piscke, 2009; Becker et al., 2010; Yan & Cai, 2021). Digital reading, a rather new and not yet well-defined concept, has taken also a considerable share in research focus the recent years. Factors like computer use, game-playing, engagement in online reading activities, digital reading communication ability, home digital activities/attitudes/practices, social media use and online information seeking have been found to have an impact on digital literacy (Rosén & Gustafsson, 2016; Rasmusson & Åberg-Bengtsson, 2015; Cheung, Mak and Sit, 2013; Chen, Lin & Chen, 2021; Gil-Flores et al., 2012).

Against this background, the main objective of this study is to investigate the similarities and differences of paper-based and digital reading using the international reading assessment, PIRLS and ePIRLS (2016) in Sweden. More specifically, our research questions are:

1. What is the association between paper-based and digital reading literacy skills for Swedish 4^th grade students?
2. How are students’ and classrooms’ digital use related to their achievement on the paper-based and digital reading test?
3. Are there differences in how SES groups perform depending on the format of the test?

The study investigates the difference between paper based reading and digital reading for Swedish grade 4 students, using data from the Progress in International Reading Study (PIRLS) and the corresponding online survey ePIRLS. In 2016, IEA administered two tests to students in some countries to assess both reading for literary experience and reading to acquire and use information through PIRLS, while e PIRLS assesses only the latter part. About 4000 ten-year-old students completed the reading test both on paper and on computer. From the students’ questionnaire we selected variables such as ‘language at home’, home resources, digital habits and exposure to digital means while indicators of SES derive from the parents’ questionnaire. The choice of the specific variables is justified from previous research that has shown the influence of these factors on either the paper-based reading or the digital one. The present study will attempt to investigate the influence of these factors on the difference between these two types of reading. The analysis is conducted both at student level and classroom level via a two-model that is specified to examine the effects of individual factors (e.g., language at home, digital habits, home resources, exposure to digital means, SES) on the difference between paper based and digital reading achievement. By using a multilevel model we account for potential cluster effects and the research questions can be investigated both for individuals and for classrooms. Sampling weights were used to account for the stratification in PIRLS. The data was processed in IBM SPSS Statistics 27, and the two-level modelling was performed using Mplus 8.

In the first step, the correlation between paper-based reading and digital reading were calculated at student and classroom level respectively. We find high correlations, about .78 at student level and .97 at classroom level. This suggests that most of those who perform well in one format also perform well in the other. However, while the correlation was very high there were also some discrepancies between the scores within the same classroom, indicating that students did better on one of the formats. To further study the difference of the two test scores, the difference between them was calculated (PVpirls – PVepirls) and modelled in a two-level regression. In a second step we related digital use and student background to the difference score that were our dependent variable. Interestingly, no significant relations were found for either digital use or social background at the within level. On the aggregated level, more pronounced differences were shown, however. The most influential variables were exposure to digital means and home educational resources. In classrooms with higher levels of digital use, the performance on the digital format was higher than for paper-based. The difference was estimated to a beta-value of about 0.4. A similar effect was noted for students’ home educational resources, thus also indicating that classrooms with higher SES performed relatively better on the digital format. The analysis was repeated for various difference scores, i.e., test-scores that represented different reading processes (interpreting etc.), but the results remained much the same (R2Between= 0.4). The results will be discussed in relation to existing large school differences in educational achievement and the implications that digitalization may have for these.

Ammermueller, A., & Pischke, J. (2009). Peer effects in European primary schools. Journal of Labor Economics, 27(3), 315-348. Baron, N., Calixte, R., & Havewala, M. (2017). The persistence of print among university students: An exploratory study. Telematics and Informatics, 34(5), 590-604. Becker, M., McElvany, N., & Kortenbruck, M. (2010). Intrinsic and extrinsic reading motivation as predictors of reading literacy: A longitudinal study. Journal of Educational Psychology, 102(4), 773-785. Chen, S. (2017). Modeling the influences of upper-elementary school students' digital reading literacy, socioeconomic factors, and self-regulated learning strategies. Research in Science & Technological Education, 35(3), 330-348. Cheung, K., Mak, S., & Sit, P. (2013). Online Reading Activities and ICT Use as Mediating Variables in Explaining the Gender Difference in Digital Reading Literacy: Comparing Hong Kong and Korea. The Asia-Pacific Education Researcher, 22(4), 709-720. 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. Gil-Flores, J., Torres-Gordillo, J., & Perera-Rodríguez, V. (2012). The role of online reader experience in explaining students’ performance in digital reading. Computers and Education, 59(2), 653-660. Gilster, Paul. (1997). Digital literacy. New York : Wiley Computer Pub. Johansson, S., Strietholt, R., Rosén, M., & Myrberg, E. (2014). Valid inferences of teachers' judgements of pupils' reading literacy: Does formal teacher competence matter? School Effectiveness and School Improvement, 25(3), 394-407. Mangen, A., Olivier, G., & Velay, J. (2019). Comparing comprehension of a long text read in print book and on kindle: Where in the text and when in the story? Frontiers in Psychology, 10, 38. Rasmusson, M., & Åberg-Bengtsson, L. (2015). Does Performance in Digital Reading Relate to Computer Game Playing? A Study of Factor Structure and Gender Patterns in 15-Year-Olds' Reading Literacy Performance. Scandinavian Journal of Educational Research, 59(6), 691-709. Rosén, M. (2006). Analysing Trends in Levels of Reading Literacy between 1970 and 2001 in Sweden, The 2nd IEA International Research Conference, The Brookings Institution Washington DC, United States, 9-11 November, 2006. Rosén, M., & Gustafsson, J. (2016). Is computer availability at home causally related to reading achievement in grade 4? A longitudinal difference in differences approach to IEA data from 1991 to 2006. Large-Scale Assessments In Education, 4(5), 1-19. Elizabeth Schmar-Dobler. (2003). Reading on the Internet: The Link between Literacy and Technology. Journal of Adolescent & Adult Literacy, 47(1), 80-85.