“ Gruelling to read ” : Swedish university students ’ perceptions of and attitudes towards academic reading in English

Students around the world are expected to read, comprehend and learn from growing numbers of English texts in higher educational contexts where the official medium of instruction is the local language. Despite this language shift, relatively little attention has been paid to the challenges academic texts in English present for students. The present paper provides insights into first-year university students ’ perceptions of and attitudes towards academic reading in English in Sweden through a sequential explanatory design with questionnaires and follow-up interviews. Sweden is often seen as a model country in terms of second-language proficiency in English, but as this study shows, a majority of first-year university students expressed negative attitudes towards academic reading in English. Student responses suggested more than one third of first-year Swedish university students in social science subjects struggled to comprehend and keep up with their assigned reading, with vocabulary and reading speed cited as their biggest challenges. This paper further shows that a considerable number of students entered higher education unaware that they were going to be required to read academic texts in English, with some questioning this common practice. Finally, implications for teachers are discussed.


Introduction
The use of English has been gaining ground in numerous educational contexts where another language was previously used (Gabriëls & Wilkinson, 2021).In an ever-globalised society, students are expected to read, comprehend, and learn from growing numbers of English texts in higher educational settings where the official medium of instruction is the societal language (i.e., not English).The importance of academic reading is widely acknowledged, but compared to academic writing, there is little previous research on reading for academic purposes in tertiary education (Maguire et al., 2020), particularly on students' experiences of academic reading in a foreign language during their undergraduate studies.
Sweden, a Nordic country in northern Europe that is the site of this study, has "consistently been at the forefront of the process of Englishization" (Gustafsson & Valcke, 2021, p. 215) over the past three decades and generally ranks amongst the top countries in the world on English proficiency tests (Dafouz & Smit, 2020).As a reading language, English has been used frequently in higher education for several decades (Kuteeva, 2014;Malmström & Pecorari, 2022;Pecorari et al., 2011;Salö, 2010).English is an integral part of university education in Sweden and beyond, and as such a certain level of English proficiency is assumed.However, in reality, the shift towards English has resulted in new challenges for students because "assumed proficiency does not always match the actual proficiency of the individuals involved" (Busby, 2020, p. 7).If this is an issue in Sweden, one of the highest ranked countries in the world on E-mail address: linda.eriksson@oru.se.

Journal of English for Academic Purposes
English proficiency tests, it can be assumed that it is also an issue in other parts of the world where students are less likely to have the proficiency required to read English-language publications.
Academic reading research has demonstrated that undergraduate students struggle to meet reading requirements at university (Ismayilli Karakoç et al., 2022).However, outside of the Nordic context, most previous research has been conducted in English-speaking countries.For example, undergraduate students have expressed concerns related to reading, especially in relation to being able to cope with the reading load under time pressure (Weir et al., 2009), reported difficulties reading academic texts (Gravatt et al., 1998) and significantly lower confidence reading journal articles than textbooks (St Clair-Thompson et al., 2018).In one of the few studies focused on academic reading in English at university in a context where the official medium of instruction is another language, engineering students in Thailand were shown to have great difficulty reading textbooks in English, impacting both their learning and their motivation to learn because the texts were "just too difficult" (Ward, 2001, p. 150).
Reading in contexts where English is an L2 is particularly challenging for students in their first year when academic emotions play a major role.Positive academic emotions and optimism have been shown to "correlate with academic success, while negative academic emotions can have the opposite effect" (Perander et al., 2020, p. 2).The participants in this study are first-year students in Swedish-medium instruction programmes.On these programmes, the language of instruction and interaction in course meetings is generally Swedish, but the assigned readings are, to varying extents, in English.The study is part of a larger project which involves data from both upper secondary schools and universities in Sweden (see Eriksson, 2023), and reports on the full extent of Swedish university students' perceptions and attitudes towards academic reading in English.

Aims and research questions
Unlike higher education institutions (HEI) in many other countries, an English proficiency test score is not required for university admission at Nordic HEIs (Busby, 2020), who instead depend on upper secondary schools to provide students with the necessary skills.HEIs take it for granted that Nordic students have the ability to read for academic purposes in English without support (Arnbjörnsdóttir, 2018), an assumption that may be tenuous.The potential impact on student retention and academic success means there is a need for more attention to be paid to students' experiences with reading.The present paper aims to illustrate the issue by discussing university students' perceptions of and attitudes towards academic reading in English through its sequential explanatory design (Creswell et al., 2003) with questionnaires and follow-up interviews.The study addresses the following research questions: 1. What perceptions do first-year university students in Sweden have of their a) preparation for, and b) ability to read for academic purposes in English? 2. Do first-year university students in Sweden report challenges in relation to English reading materials and if so, what challenges do they report?

Contextualising the study
The gradual expansion of English in research and publishing has led to more than 90% of indexed social science journals and 95% of natural science journals to be only or partially English-language publications (Dafouz, 2020).In smaller countries, publishing textbooks in local languages is often financially prohibitive (Arnbjörnsdóttir, 2018).As a result, it can be difficult for teachers in higher education to avoid assigning texts written in English when creating reading lists.These reading lists primarily consist of textbooks, chapters in edited volumes, and journal articles, but can also include conference proceedings, reports, manuals, policy documents, and other relevant materials.Research on reading for academic purposes in higher education should thus not be limited to textbooks only as it would not provide the complete context.In this article, I adopt a broad view of reading lists and define academic texts as all texts that fulfil a purpose of education in university.
Prior to entering higher education, the typical Swedish student has attended compulsory school for ten years (year 0-9) and upper secondary school for three years (year 10-12).The syllabus for English in upper secondary school is divided into three courses: English 5, 6 and 7. Learners start studying English formally before the age of nine, with two thirds of Swedish learners achieving a B2 level in

Table 1
Extracts related to academic reading and scientific content from the syllabus for English in upper secondary school (Skolverket, 2011).
Upper secondary school courses in English in Sweden English 5 (Year 10) English 6 (Year 11) English 7 (Year 12) Subject areas related to students' education, and societal and working life; current issues; events and processes; thoughts, opinions, ideas, experiences and feelings; relationships and ethical issues.
Concrete and abstract subject areas related to students' education and societal and working life; current issues; thoughts, opinions, ideas, experiences and feelings; ethical and existential issues.
Theoretical and complex subject areas, also of a more scientific nature, related to students' education, chosen specialisation area, societal issues and working life; thoughts, opinions, ideas, experiences and feelings; cultural expressions in modern times and historically, such as literary periods.Texts of different kinds and for different purposes, such as manuals, popular science texts and reports.
Texts of different kinds and for different purposes, such as formal letters, popular science texts and reviews.
Texts of different kinds and for different purposes, such as agreements, in-depth articles and scientific texts.

L. Eriksson
reading comprehension in the European Survey on Language Competences 2011 (Skolverket, 2012) in their final year of compulsory school (i.e., year 9).The Swedish National Agency for Education (Skolverket) prepares knowledge requirements for schools, which involve explicit goals for compulsory school such as being able to read instructions, descriptions, newspaper articles, and novels in English (Skolverket, 2022a).In order to be eligible for university studies, learners must continue to upper secondary school.New curricula for both compulsory school (Lgr 11) and upper secondary school (Gy 2011) were implemented in 2011.One of the major changes in the new curriculum was the requirement for two years of English in upper secondary school for university eligibility, compared to just one year previously.A third year of English is optional but generates a credit increment that makes it easier for the applicant to be admitted to university.Each upper secondary course has its own set of content which should be covered, as well as its own grading criteria.As seen in Table 1, the syllabus for the two mandatory courses focuses on subject areas such as current and ethical issues, and texts such as manuals, formal letters and popular science texts.
When the vast majority of the participants in this study attended upper secondary school, content such as "theoretical and complex subject areas, also of a more scientific nature" and "in-depth articles and scientific texts" was not introduced until English 7 (Skolverket, 2011), which is the optional course given in the final year of upper secondary school.As a result, a large proportion of first-year university students are likely to have little experience reading academic English, which could affect their studies negatively.

Literature review
Reading for academic purposes is an integral part of higher education and previous research has shown a strong correlation between reading proficiency and success among undergraduate students (e.g., Brost & Bradley, 2006;Fatiloro et al., 2017).Yet in L1 contexts, around half of university students do not complete their required reading (Aagaard et al., 2014) because of lack of preparation and/or motivation, time constraints, and an underestimation of the importance of reading (Maguire et al., 2020).In order to succeed in higher education, reading comprehension alone is not sufficient.Students also need to have the skills required to be able to engage critically with the texts on their reading lists, demonstrate their critical thinking in assignments, and construct arguments based on their readings in essays (Ismayilli Karakoç et al., 2022).
In an L2 setting, students are often expected to read the same texts as L1 users even though reading in an L2 "imposes a number of additional constraints on reading and its development" (Grabe, 1999, p. 11).These constraints include, but are not limited to, the use of counterproductive strategies such as excessive dictionary use, difficulty guessing the grammatical function of unfamiliar words from context, and the need for additional cognitive processes which are required to make up for a lack of automatic recognition of words (Hellekjaer, 2009).Vocabulary knowledge in particular is a critical component for fluency, and has been described as "the greatest single impediment to fluent reading by ESL students" (Grabe, 1988, p. 63).Students who read more have been shown to outperform students who read less on all aspects of academic language, including vocabulary, spelling, reading and writing (Krashen & Brown, 2007).The impact on the development of academic language as well as on content learning is thus greatly affected if students do not engage with their required reading.
In the Nordic context, around one third of university students in Iceland and Norway were found to be struggling with academic English texts (Arnbjörnsdóttir, 2018;Hellekjaer, 2009).In Denmark, university lecturers have expressed that students are "sort of taken aback" (Dimova, 2020, p. 8) when they realize the difference between general and academic English.Nordic scholars have further suggested the English taught in schools "has not kept up with the changing role of English in the Nordic countries" (Arnbjörnsdóttir, 2018, p. 145), thereby impacting students' experiences with academic reading in English in higher education.
In Sweden, a recent report found that nearly half of the required reading on undergraduate courses with Swedish as the official language of instruction is now in English (Malmström & Pecorari, 2022), despite reports that a majority of undergraduate students said they understood less when the textbook was in English, and that more effort was required to read in English than in Swedish (Pecorari et al., 2011).In the same study, more than four in five students expressed a negative attitude towards English textbooks, although only 44% said they would not choose English textbooks if the choice was theirs.In Mežek (2013), reading speed in English was perceived as an issue by students in biology and correlated with vocabulary knowledge.Biology students also had trouble learning English subject-specific terminology.However, if given extra time, both Shaw and McMillion (2008) and Mežek (2013) found that Swedish university students in biology performed at a similar level to that of British L1 students.Finally, Airey (2009) found that some first-year students had considerable difficulties speaking about disciplinary concepts in English; however, his investigation focused on students' experience attending lectures rather than reading, which is the focus of the current study.

Material and methods
The data for this study were collected from three universities in Sweden in 2021 after pilot testing.A total of 23 programmes in behavioural and social sciences at the three universities were identified during a systematic review of reading lists.For inclusion in the study, the selected programmes all fulfilled a set criterion of a minimum of one textbook or three journal articles in English during the first term of the programme but had Swedish listed as the official medium of instruction in the programme syllabus.This consistency measure was important to ensure validity as self-assessment items need to be "based upon task content tied to students' situations as potential users of the language in question" (Bachman, 1990, p. 148).University teachers were contacted with information about the study, with 16 programmes opting to participate.A questionnaire was distributed to students in their first year on the participating programmes during a lecture or a seminar (see section 4.1).After the questionnaires had been collected, twelve students participated in follow-up interviews (section 4.2).Participating students received no compensation for taking part in the study.

Student questionnaire
The questionnaire was designed to include demographic, closed-ended and open-ended items, and consisted of 54 questions.Example questions have been included in Appendix 1.To ensure a reasonably high response rate, the questionnaire was designed and constructed in Survey&Report (Version 4.3; Artlogik 2019) so that it could be filled out online during or after a lecture or seminar.Multiple-choice questions using Likert scales were used in addition to a small number of open-ended questions which allowed participants to provide more detail.Questions which invoke the episodic memory of specific skills is one of the best bases for selfassessment (Ross, 1998), and in this study, questions were formulated so that the participants would be required to remember past personal experiences such as times and associated emotions, which are properties of the episodic memory (Clayton et al., 2007).Furthermore, the questionnaire was constructed in Swedish to not exclude students who had difficulties reading or responding in English.
The questionnaire was pilot tested on more than 50 students to ensure that all questions were easy to understand, and that the questionnaire could be filled out in about 10 min.Questionnaire items included in this paper focused on two themes: perceptions of their ability to read for academic purposes in English and reading speed.The questionnaire was introduced and distributed to students in person by the author to try to reduce self-selection bias, but due to the Covid-19 pandemic, lectures and seminars were held in Zoom, which may have affected the response frequency as detailed in Table 2.
The data from the demographic items show that most of the participants were female (n = 157, 77%). 1 The youngest participant was 18 years old and the oldest 46, with a median age of 21 (mean = 23.2,SD = 4.9), which is similar to the median age for first-year students in all subjects which was 21.4 in 2021/22 (Söderman, 2022).Most students were first-year students (n = 165, 81%) who had Swedish as an L1 (n = 193, 94%) and had attended their entire schooling in Sweden (n = 198, 96%).The closed-ended questionnaire items are presented using descriptive statistics, while the open-ended questionnaire items were analysed in NVivo using a conventional approach to content analysis with inductive category development (Hsieh & Shannon, 2005).For example, the first question in the questionnaire asked students about their attitudes towards assigned texts in English.In total, 318 items from 198 open-ended responses were coded (see section 5.1 for examples) and sorted into three categories during analysis: positive, negative and neutral attitudes.

Follow-up interviews
Semi-structured interviews were conducted with twelve students who had responded to the questionnaire and opted to fill in their e-mail when asked if they would be willing to participate in an interview on the same topic.The first three questions included in the interview protocol, which were used as a starting point for the interviews, were the following: 1. How did it feel when you received the first text in English at university?Do you remember what it was like? 2. Did you know that you would be required to read texts in English when you started university?3.In the questionnaire you said that you felt prepared/did not feel prepared to read assigned literature in English when you started university.Can you tell me a bit more about that?
The interview participants represented all three universities and reflected the demographic data of the student questionnaire in regard to gender, age and time spent in higher education.Participants were asked to give oral consent to the interviews being recorded.The interviews resulted in 216 min of data which were later transcribed by the author.The interview transcripts were then analysed using qualitative content analysis.

Results
The questionnaire was designed to provide direct measures of student attitudes and self-reported behaviour regarding reading for academic purposes in English.The sequential explanatory design (Creswell et al., 2003) with follow-up interviews complements the questionnaire results and provides opportunities for exploring the quantitative results in more detail (Ivankova et al., 2006).This section gives the responses for four areas: attitudes towards assigned texts in English, perceived preparedness for reading for academic purposes in English, challenges involved in reading for academic purposes, and reading speed.

Attitudes towards assigned reading in English
The first research question asked what perceptions first-year university students in Sweden have of their preparation for and ability to read for academic purposes in English.In the questionnaire, participants were asked how they felt when they saw a text in English on their reading list.Thirty-nine items (12%) were categorised as positive attitudes, and included codes such as fun, international and preparatory for the future.Sixty-two items (20%) were categorised as neutral attitudes, of which 43 items were coded as nothing/ anything special.This comment was typical of the code: Nothing special, I understood quite quickly that a majority of what we're going to read is in English.I don't have any major issues with my English, so it doesn't cause any negative premonitions.Of course it takes longer time to read for example a scientific article in English compared to Swedish, so you still have a feeling of that the page numbers do not correspond to the normal reading time a Swedish text would result in.
Whilst s ome respondents simply wrote that they did not feel anything special when they saw English texts on their course reading lists, many started by stating they did not feel anything special but later conceded that academic texts in English were more difficult or took longer to read compared to Swedish.The remaining 217 items (68%) were categorised as negative attitudes.These included codes such as annoying, difficult, stressed, panic, and time-consuming.Responses included the following: I feel very unsure of how it will go, frustrated that I will not learn as much as I would have if it had been in Swedish.

P5
I think: damn it, horrible, feels very hard and difficult to get through and understand, very time-consuming, especially scientific articles.P6 At first I was indifferent, but gradually it has turned into a bit of anxiety as you have to focus more on the language than the content.
As seen in the response from P3, even students who perceived themselves to be good at English expressed negative attitudes to academic texts in English.This also includes students who had the highest possible grades in the optional English 7 course in upper secondary school, 51% (n = 19) of whom responded that seeing a text in English on their reading list made them feel it would be more demanding and time-consuming, and in the case of some, such as P2 and P6, angst-ridden.
The gradual increase in anxiety mentioned by P6 and use of various avoidance strategies were mentioned by a small number of students in the data (n = 6, 3%).In one of the interviews, one student elaborated on their answer and said that despite being almost a year into the programme, they avoided reading English texts to the greatest extent possible.As an example, the student discussed a final essay they were working on which required searching a database for previous research.The student said even though they knew there is more and perhaps better information in the English research available, they only searched for journal articles written in Swedish.In contrast, a slightly larger number of students mentioned in their responses that reading assigned texts in English had gotten easier with time (n = 14, 7%), which perhaps suggests some of these students had been able to overcome the additional constraints imposed by reading in an L2 and develop their academic language.

Perceived preparedness for reading assigned texts in English
The first research question also asked what perceptions first-year university students in Sweden have of their preparation for reading for academic purposes in English.In response to a question concerning their perceived preparedness in the questionnaire, more than two in three students responded that they felt prepared (n = 139, 68%), as seen in Fig. 1.
However, almost one in three students responded they 'did not feel sufficiently prepared' (n = 57, 28%) or 'did not feel prepared at all' (n = 9, 4%).Students were further asked how easy or difficult they thought it was to read the texts on their university reading lists on a five-point Likert scale, to which a slightly larger number of students (35%, n = 72) said they thought reading assigned texts in English was 'quite difficult' or 'very difficult' compared to 6% (n = 12) in Swedish, as seen in Fig. 2.
In contrast, 36% (n = 74) of respondents said it was 'quite easy' or 'very easy' to read assigned texts in English, compared to 84% (n = 172) in Swedish, suggesting that a considerable number of students had a harder time reading in English than the official language of instruction.
In response to an open-ended question about preparedness, 15% (n = 20) explained that they had said they felt prepared because they had been told to expect considerable reading in English, primarily from friends and family.This was confirmed by students in interviews, as exemplified by P7: P7 Yes exactly but it was because I knew beforehand that I was going to be reading English texts, English literature but it was probably the prior knowledge that motivated my answer.
Similarly, 16% (n = 10) who said they did not feel prepared noted in their comments that this was because they were not aware they were going to be required to read in English.Several students wrote that their programme was "supposed to be in Swedish", with some questioning the inclusion of texts in English.
Other explanations provided by the students as to why they did not feel prepared can be categorised in two groups: 1) insufficient knowledge of English in general, and 2) insufficient knowledge of academic English.Students in the first group explained they were not good at English in general, that they rarely read in English or that they had not used English since they finished school several years ago.Students in the second group described how they were unfamiliar with academic texts in English and blamed "bad teachers" in upper secondary school.Students also explain that there was a large gap between the level of English taught in upper secondary school, which was described as "everyday English", and the level required at university.

Challenges involved in reading for academic purposes in English
The second research question asked if first-year university students in Sweden report challenges in relation to English reading materials and if so, what challenges they report.Asked how often they had a hard time understanding the content when they read the English texts on their reading lists, 23% of the participants responded 'always' or 'often' (n = 46), and an additional 46% 'sometimes' (n = 95) (Fig. 3).
Students were also asked how often they had issues with new vocabulary items when they were reading the English texts on their reading lists, to which 42% responded 'always' or 'often' (n = 85), and an additional 35% 'sometimes' (n = 72), with only 1% responding 'never'.Asked to explain in their own words what they thought was difficult about reading assigned texts in English, nearly two in three (n = 121, 62%) suggested vocabulary is the biggest issue.These comments were typical of the responses:

P8
You get bored because you have to look up so many words, the reading has no flow and that makes you tired.I also get frustrated when I feel "bad" at English.

P9
One is forced to look up words ("normal" words, not technical terms) which makes it so that I have worse flow when reading.P10 There are many words I don't recognize.It takes an extra-long time to read.Sometimes I have to use Google.
Respondents noted issues with vocabulary obstructed reading fluency, making the reading more time-consuming (n = 46, 24%).

L. Eriksson
Respondents also wrote that the English texts on their reading lists required more concentration and as a result, made them tired.A smaller number of students (n = 9, 5%) mentioned difficulties with sentence structure, as exemplified by this response:

P11
The sentence structure is very complicated and, in my opinion, very unnecessarily constructed.I understand that it's academic texts but writing with a simple language should not be considered to be wrong.
In interviews, students explained that there was a scale of difficulty amongst different text types and that it was primarily journal articles in English that were challenging, as seen in P12, whereas English textbooks provided a good introduction to the subject, as seen in P13:

P12
I had not read any scientific articles before and I was not prepared for those.The textbooks I have, they have been a bit more, maybe how one would expect, but it's the articles that are difficult.

P13
There have been markers in the text, it has been more colourful, all of those things that makes it easier to study and so I have really preferred the English books.
English textbooks provide examples without using too many technical terms.Students further emphasised that English textbooks were more pedagogical than Swedish textbooks, making them easier to read.

Reading speed
An additional challenge reported by students was reading speed.One question asked students in which language they perceive to read assigned texts faster.As seen in Fig. 5, 89% (n = 183) reported they read faster in Swedish than in English.
As previously mentioned in section 4.3, students reported that reading in English was time-consuming (n = 46, 24%) when they were asked what they thought was difficult about reading assigned texts in English.Student responses ranged widely from "slightly longer" to "about 3 times as long".Asked how often they had struggled to read their assigned reading on time during the previous term, which for most participants was their first term at university, 11% (n = 23) said they 'always' struggled in English compared to just 1% (n = 3) in Swedish, as seen in Fig. 6.
The results shown in Fig. 6 suggest students' perceived reading rate in English was not only an issue in the sense that it took them longer to read their assigned reading, but that it took so much longer that almost one third frequently struggled to keep up.This is also reflected in participant responses, with one student remarking the following:

P15
I think it will take longer to get through than if it would have been in Swedish.I prioritize reading the assigned literature that is in Swedish first, which usually results in the English being read very sloppily or not at all.This is especially concerning in light of the observation that 9% (n = 18) of respondents said they gave up and stopped reading if they did not understand a sentence or a paragraph in their assigned reading.

Discussion
The negative attitudes expressed, the reported issues comprehending assigned texts in English, the considerable amount of time required for reading them, and the avoidance strategies at least some of the participating students used undoubtedly had an impact on their quality of work for the course, such as group work, assignments, and exams, but most importantly on the quality of their learning (section 6.1 and 6.2).Considering that Sweden usually ranks amongst the highest countries in English proficiency in the world (Dafouz & Smit, 2020), it is not unreasonable to assume that the issues raised in this paper are impacting other countries where EAL learners have to read English academic texts in higher education, and implications of the results of this study, with a focus on teachers, conclude the paper (section 6.3).

Student attitudes and reading ability
A partial aim of the present study was to investigate what perceptions students have of their preparation for and their ability to read academic texts at university.The ability to read and understand academic texts is one of the most important skills that EFL university Fig. 5. Student responses regarding which language they read assigned texts faster in.students need to acquire (Nergis, 2013).Reading comprehension is essential to academic learning, but also to career success and lifelong learning in all subject areas (Dreyer & Nel, 2003), but the findings of this study suggest more than one third of first-year Swedish university students in social science struggled to comprehend and keep up with their assigned reading, which corresponds with Arnbjörnsdóttir's (2018) and Hellekjaer's (2009) findings from two of the other Nordic countries.
Language learning is not an objective for the participating programmes in this study, but the overwhelmingly negative attitudes expressed by students in relation to reading for academic purposes in English may impact their academic success.This notion is in line with previous research which has suggested negative academic emotions correlate with poor results (Perander et al., 2020).For example, stress has disruptive effects on brain function and structure and is associated with alterations in cognitive processes such as learning and memory (Pakulak et al., 2018), which is likely to impact content learning and knowledge retention.Optimal language learning conditions suggest assigned reading should be slightly more difficult than the students' current level of language proficiency (i + 1, Krashen, 1982).However, in this case, it appears much of the assigned reading on social science programmes at Swedish universities presents students with input that is not fully comprehensible (i + 2 or more), effectively hindering their language learning and disrupting their learning and motivation (Ward, 2001), resulting in the negative attitudes expressed in this paper.
In contrast with Pecorari et al. (2011), who found that negative comments outnumbered positive comments four to one, the negative comments expressed by the participants in this study (68%) were considerably lower.Since this study has only surveyed first-year students in social science programmes, it is possible student attitudes have not improved in general; however, it is also possible the implementation of two years of mandatory English in upper secondary school after Pecorari et al. (2011) collected their data has had a positive impact on student preparedness for further education, somewhat improving their attitudes towards assigned texts in English.Further research on a larger population which includes students in other subject areas and in all three years of undergraduate studies is necessary in order to investigate students' attitudes further.Ideally such a study would be longitudinal in nature and investigate students' perceptions and attitudes at multiple occasions during their undergraduate studies to be able to identify ways in which students learn to adapt to academic texts in English.Such a study would be an important contribution as there are some indications that students adapt to studying in English after about one year of university studies (e.g., Klaassen, 2001).In the present study, a small number of students mentioned assigned reading in English had gotten easier with time.Airey (2009) similarly found that Swedish university students adapted to being taught in English, for example by beginning to read sections of the assigned literature before lectures.However, Airey also found that students adapted in ways that did not necessarily aid their learning, with some using lectures for mechanical note-taking while others ceased to take notes entirely in class.Investigating how these adaptations affect learning is crucial for better understanding the implications of teaching and learning in a foreign language.

Challenges involved in reading academic English
The second research question asked if first-year university students in Sweden report challenges in relation to English reading materials and if so, what challenges they report.The perceived differences in reading speed between Swedish and English was one of the most prevalent themes in the data, and in line with Hellekjaer (2005), a majority of the respondents said that they read slower in English than in Swedish.This is not surprising but may, in combination with their struggle to read their assigned texts in English on time, suggest that at least some of these students did not have issues with reading proficiency in general, but language-specific issues that affected their reading in English.Slow reading speed can be an indication of processing difficulties, with reading comprehension hindered by the limitations of our working memory (Grabe, 1999).Previous research involving Swedish biology students found that the students performed at a similar level to that of British L1 students if they were given extra time (Mežek, 2013;Shaw & McMillion, 2008).Further studies involving students in social science and other areas are necessary; however, by considering the findings of this study and the additional constraints on reading imposed by reading in an L2, university teachers should, at minimum, have strong arguments for adjusting the reading requirements.
Another prevalent theme in the data was vocabulary, which almost two in three participants reported as a challenge.Participants frequently mentioned needing to look up or translate vocabulary items, indicating the use of counterproductive strategies associated with L2 reading, impacting not only their reading but also their reading development (Grabe, 1999).Vocabulary knowledge is a critical component for fluency (Grabe, 1988) and by providing vocabulary support for students, content learning would likely improve.Participants further brought up that there was a perceived scale of difficulty amongst different text types.For at least some of the participating students, it was primarily journal articles in English that provided a challenge, while English textbooks were seen as good introductions to the subject.This is line with St Clair-Thompson et al. (2018) who found university students were significantly less confident reading journal articles than textbooks, but further research is needed to investigate the impact of genre on students' Academic reading is a "hidden curriculum" without explicit academic reading instruction (Isakson & Isakson, 2017, p. 156).The findings of the present study suggest that the transition from upper secondary school to university is complicated by the perceived gap between the level of English taught in upper secondary school, which was described by some of the participants as "everyday English", and the level required at university.Although the reported level of English taught can be placed in contrast with the current Swedish curriculum for the final mandatory English course which includes content such as complex and formal texts, novels and extracts from older pieces of literature, the curriculum for English in upper secondary school "lacks explicit guidance about timings and priorities when it comes to the use of class time to cover each of the elements" (Siegel, 2019, p. 267).Ambiguous steering documents allow teachers to "put the syllabus into practice according to their own interpretations" (Siegel, 2019, p. 275), which may explain students' experiences of everyday English in upper secondary school.
However, further research is needed to investigate why students report that teachers focus on everyday English.For example, teachers may believe everyday English is more useful for the students or that they are not prepared to learn academic English.Another possibility is that teacher training programmes have not provided teachers with sufficient training to teach academic English.Regardless, the failure to introduce scientific texts, in-depth articles and theoretical subject areas in the obligatory English courses required for further education is a sign that the curriculum has not kept up with the changing role of English in Sweden and elsewhere (see e.g., Arnbjörnsdóttir, 2018).A recent addition to the optional English course in Sweden, "texts used in higher education" (Skolverket, 2022b), is a step in the right direction, but as the findings in the present study suggest, it is likely not enough to prepare students to read academic texts in English at university.

Implications for teachers and conclusions
The results outlined in this paper indicate the need for university teachers to introduce English texts more gradually, allowing students to first become acquainted with subject content and terminology in their L1 before 1) introducing textbooks in English, and 2) journal articles and other texts which have been written for an expert audience, in line with student perceptions that journal articles are more challenging than textbooks (see also St Clair-Thompson et al., 2018).A gradual introduction of English may be hindered in some fields by a shortage of published works in the local language, particularly journal articles.However, considerable differences between the number of assigned texts in English have been identified at Swedish universities by Malmström and Pecorari (2022), with several universities assigning less than 25% of texts in English while others assign up to 65% in English.These differences indicate that at least some universities could assign considerably larger amounts of texts in the local language than they do currently.Irrespective of the amount of literature available in the local language, university teachers may want to consider adjusting the number of pages assigned in an L2 compared with an L1.For example, formulas used by Åbo Akademi and others in Finland that were first introduced by Karjalainen et al. (2006) involve assigning 10-12 pages per hour in the students' first language or for texts that are relatively easy to read but only 4-6 pages per hour in a foreign language or for texts that are difficult to read (Åbo Akademi, 2022, p. 20).The transition from upper secondary school to university would likely also be made easier for students if difficulties with vocabulary in assigned readings in English were anticipated, as previously suggested by Shaw and McMillion (2011).
For English teachers in upper secondary school and EAP teachers, devoting time to improving students' reading fluency, for example through extensive reading and repeated exposure to vocabulary items from the Academic Word List (Coxhead, 2000) or subject-specific vocabulary lists (e.g., Coxhead & Hirsch, 2007;Wang et al., 2008;Ward, 2009), would possibly ease the transition from English classes to content learning at university.The impact of extensive reading in particular on reading achievement cannot be overlooked (Kirchhoff, 2013) and has been found to correlate with academic English reading proficiency (Hellekjaer, 2005).Participating students would likely also benefit from exposure to authentic textbooks and journal articles used in higher education, with a particular focus on strategies which help students read and understand journal articles, for example how journal articles are commonly structured and what readers can expect to find in each section, how to differentiate between different types of journal articles, and what questions readers can ask themselves about the content while reading.
The lack of awareness some of the participating students expressed in relation to assigned reading in Swedish medium instruction programmes being in English is a further cause for concern when considering the increasing amount of required reading in English.The lack of information provided not only by universities when describing their courses and programmes, but also by upper secondary school teachers and guidance counsellors, have resulted in large numbers of Swedish students who were unable to make informed choices prior to entering higher education.By providing all upper secondary school students with accurate information about how much English is used at university and for what purposes would allow them to make different decisions regarding, for example, whether to attend optional English classes in school or to engage in extensive reading in English in and out of school.
To conclude, the updated knowledge the present study has provided about Swedish students' perceptions of and attitudes towards academic reading in English has the potential to inform educational policy and practises around the world.Sweden is often seen as a model country in terms of second-language proficiency in English, and if students perceive academic reading to be an issue in Sweden, it is likely an issue elsewhere as well.L2 reading research is largely dominated by experimental research.Although these studies provide valuable contributions, policy makers and practitioners also need to consider non-experimental factors which may contribute to students' reading involvement.Whether or not students are able to read at a certain speed or answer questions accurately about a text in an experiment is irrelevant if the same student perceives that they cannot understand or keep up with their reading at a satisfactory level in their daily life.The findings of this study suggest a considerable number of students have a negative attitude to assigned reading in English, perceive difficulties comprehending and keeping up with their assigned reading, and were unaware they would be required to read in English as new students.Teaching and learning in a foreign language, particularly in cases such as these where students have not actively chosen to study in an L2, is likely to have a negative impact on the quality of learning in the present for some students, and career success and lifelong learning in the future, reinforcing the need for further investigations and policy changes to reduce educational inequalities.
P2I feel a bit of panic and anxiety, as I'm well aware I have a very hard time with English.I have to put in an extreme amount of effort to understand an English text, but despite this I often times cannot keep up at all and understand what it really says.P3 I usually feel tired, it's gruelling to read in English despite my knowledge in it being very good.It works, but I prefer Swedish.P4

Fig. 1 .
Fig. 1.Student responses to whether they felt prepared to read assigned texts in English when they started university.

Fig. 2 .Fig. 3 .
Fig. 2. Student responses regarding how easy they think it is to read assigned texts at university a) in English and b) in Swedish.

Fig. 6 .
Fig. 6.Student responses regarding how often they had a hard time reading assigned texts on time during the previous term.

Table 2
Overview of participating university programmes.