Umeå University's logo

The aims were to investigate (1) the role reading and writing may play in educational attainment, (2) the text quality of students with reading difficulties (RD), and (3) relations between reading, self-efficacy for writing (SEW), and text quality in Swedish (L1) and English (L2). 

As to methods, participants (N =159; ages 17–18) were screened for word recognition and reading comprehension to assess reading ability and identify groups with RD and typical reading (TR). Data consisted of results from writing tasks, information on students’ educational attainment, study programme, study background – one year (SB1year) or two years (SB2years) of Swedish and English study – and online SEW responses.

Regarding what explains the variation in educational attainment for the whole group, logistic regression analyses indicated that especially writing performance but also reading comprehension, word recognition, study programme, and gender to a certain extent affected educational attainment in Swedish, and reading comprehension, writing, and study programme in English. Concerning students’ specific reading profiles and educational attainment, a chi-square test showed that students with RD and SB1year had the lowest attainment scores of all groups in Swedish, English, social studies, and history, whereas no differences were observed between students with RD and TR with SB2years. 

With respect to how reading and SEW related to the text quality of the whole group, multiple regression outcomes demonstrated that word recognition and SEW contributed to Swedish and English text quality, and that reading comprehension contributed to English text quality. When it comes to students with different reading profiles and text quality, an ANOVA revealed that RD took a heavier toll on Swedish text quality in the SB1year-group compared to the group with SB2years. Students with RD in Swedish had greater difficulties in writing a good text in English than peers with TR regardless of study background. In Swedish and English text quality, the group with RD and SB1year received the lowest scores. Overall low Swedish and English argumentative text quality was observed, regardless of reading profile. A Kruskal-Wallis H Test showed that the group with poor reading comprehension had significant challenges with spelling in Swedish, and with spelling, language use, and cohesion in English. Peers with poor word recognition were challenged by spelling in Swedish and English. 

Findings indicate that especially writing, but also reading, may play an important role in educational attainment, and they underscore the importance of including writing and reading instruction in foundation subjects (and syllabuses) in order to achieve better attainment. Results reveal that argumentative writing can be challenging for many upper secondary students. Furthermore, outcomes highlight the need for support in both L1 and L2 writing (including SEW) to SB1year-students with RD. Finally, findings suggest that groups with reading difficulties may need macro level support (e.g., coherence) as well as linguistic support (e.g., spelling, grammar, and cohesion).