This study evaluates and adapts the Extramural English (EE) Scale by Sundqvist and Uztosun

(2024) and investigates its relationship with English reading comprehension (ERC) among

Swedish Scholastic Aptitude Test test-takers (n = 3,787). As English is increasingly integrated

in Swedish society from an early age through informal contexts, assessing young peoples’

engagement in EE and understanding its implications for academic achievement is important.

The originally proposed eight-factor EE Scale was evaluated through confirmatory factor

analysis (CFA). Although the proposed eight-factor model showed a good fit to the data, some

items and scales had nonoptimal psychometric characteristics. Therefore, alternative

conceptualizations were explored using exploratory factor analysis (EFA) followed by CFA.

Results supported a refined five-factor model. Structural equation modeling (SEM) showed

that EE Media and EE Internalized positively predicted ERC scores, whereas EE Gaming and

EE Social Interaction were not significantly related to ERC. The EE Digital Creativity factor

emerged as a negative predictor. Overall, the adapted scale explained 19 % of the variance in

ERC scores. This study contributes to the development of a psychometrically robust measure

of EE, facilitating more consistent assessment, replication and comparison of EE research

findings across populations and contexts.

This study uses the EE Scale and SweSAT data from the spring administration of 2025 (n = 3,787). A non-probability purposive sampling method was used, intentionally targeting first-time test takers to answer the questionnaire. EE data were collected through Sundqvist and Uztosun’s (2024) constructed items. However, item response options were altered into a time-based response scale represented by a 5-point Likert scale: 5 = Daily, 4 = Once or a few times per week, 3 = Once or a few times per month, 2 = Once or a few times per year, and 1 = Less often than once per year or never. EE Scale consisted in the following factors: EE Digital Creativity included three items about creating materials to share with others or for oneself in a digital context, whether that would be a video, podcast or music. EE Gaming factor included seven items about playing digital games alone, with others, and watching others play, and focusing on a specific language skill. EE Internalized factor was about three items gathering data about engagement in activities one does alone, such as daydreaming and the presence of inner speech. EE Music included three items about listening to music, reading music lyrics, and singing. EE Niche Activities factor consisted of three items collecting the amount of engagement in activities such as writing fanfiction, playing board games, and learning English. EE Reading and Listening factor included three items measuring receptive language skills, which refer to understand and process language received. EE Social Interaction, in which six items aim to gather data about speaking and writing to others in real life and digital form. EE Viewing factor includes four items collecting information about watching content and following English-speaking content creators. Moreover, ERC SweSAT subtest data were also used, which includes five sentence-completion items and 15 items related to interpreting short text and long texts, and test-takers are expected to identify relevant information rather than isolated details. The complexity relies on the text rather than the questions, which are straightforward and constructed with simple language. Moreover, the material contains an extended range of topics, such as social science, arts, technology, and natural science (Stage & Ögren, 2004). All the analyses were conducted using R, version 4.5.1 (R Core Team, 2025), to compute Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), modification indices (MI), and Structure Equation Modelling (SEM).

This research contributes in addressing the persistent lack of a common measure for EE, a gap that has not allowed to efficiently compare results across different linguistic and educational contexts, particularly within high-stakes assessment such as the SweSAT. Findings indicate that while the original EE Scale offers a viable baseline to address the existing issue, an adapted 5-factor model showed an adequate fit for the SweSAT data, and acceptable factor loadings among the items. Additionally, the adapted EE Scale factors predicted ERC scores, explaining 19% of the variance in ERC score, although the impact of EE was not equal across dimensions: viewing, reading, listening, music and internalized related activities were stronger predictors. The shift from the original 8-factor to a 5-factor structure was a result of adapting the scale to SweSAT data. These dimensions suggest that certain EE activities do not have clear boundaries, that is, SweSAT test-takers perceive these as a collective mode of engaging in EE and applying English skills in a similar way. Interestingly, the EE Internalized factor remained in the adapted EE Scale and positively predicted ERC scores. Regarding EE Digital Creativity, all three items loaded in the adapted EE Scale. However, the activities that represent this factor were the least common among test-takers, and negatively predicted ERC scores. This finding requires further investigation on clarifying the application of English in these EE activities. Furthermore, the EE Social Interaction factor displayed lack of significance in predicting ERC scores. Moreover, the EE Gaming factor remained unchanged from the original EE Scale. However, this factor did not significantly predict ERC scores. Additionally, the EE Niche Activities did not load into the model due to low factor loadings, low communalities and high complexity.

Benson, P. (2011). Teaching and researching autonomy in language learning (2nd ed.). Pearson Education. European Commission (2024). Europeans and their languages. Brussels: European Commission. Löwenadler, J. (2023). Trends in Swedish and English Reading Comprehension Ability among Swedish Adolescents: A Study of SweSAT Data 2012–2018. Scandinavian Journal of Educational Research, 67(4), 591–606. https://doi.org/10.1080/00313831.2022.2042841 Malmström, H., Pecorari, D. (2022). Language choice and internationalisation: The roles of Swedish and English in research and higher education. Institute for Language and Folklore. Neagu, T., Eklöf, H., Laukaityte, I., & Wedman, J. (2025). Reading, watching and gaming: exploring the relationships between extramural English activities and academic L2 English reading comprehension in a Swedish university admissions test context. Education Inquiry, 1-16. https://doi.org/10.1080/20004508.2025.2556572 Poole, F. J., & Clarke-Midura, J. (2020). A systematic review of digital games in second language learning studies. International Journal of Game-Based Learning (IJGBL), 10(3), 1-15. http://dx.doi.org/10.4018/IJGBL.2020070101 R Core Team (2025). R: A language and environment for statistical computing. R Foundation for Statistical Computing[Computer software]. Vienna, Austria. https://www.R-project.org/ Stage, C., & Ögren, G. (2004). The Swedish scholastic assessment test (SweSAT). Development, results and experiences. Sundqvist, P. (2009). Extramural English matters: Out-of-school English and its impact on Swedish ninth graders' oral proficiency and vocabulary [Doctoral dissertation, Karlstad University]. http://kau.diva-portal.org/smash/record.jsf?pid=diva2:275141 Sundqvist, P., & Uztosun, M. S. (2024). Extramural English in Scandinavia and Asia: Scale development, learner engagement, and perceived speaking ability. Tesol Quarterly, 58(4), 1638-1665. https://doi.org/10.1002/tesq.3296Digital Object Identifier (DOI) Sundqvist, P., & Wikström, P. (2015). Out-of-school digital gameplay and in-school L2 English vocabulary outcomes. System, 51, 65-76. https://doi.org/10.1016/j.system.2015.04.001 Sylvén, L. K. (2022). Very young Swedish children’s exposure to English outside of school. In C. Bardel, C. Hedman, K. Rejman, & E. Zetterholm (Eds.), Exploring Language Education: Global and Local Perspectives. Stockholm: Stockholm University Press. https://doi.org/10.16993/bbz.e Warnby, M. (2022). Receptive academic vocabulary knowledge and extramural English involvement–is there a correlation? ITL—International Journal of Applied Linguistics, 173(1), 120–152. https://doi.org/10.1075/itl.21021.war