A number of South African reading researchers have argued that students’ poor reading ability at tertiary level is a direct result of poor reading ability at school level (Howie et al. 2017; Pretorius 2007; Pretorius & Lephalala 2011; Taylor & Yu 2009; Van Staden & Bosker 2014; Zimmerman 2014). Studies on the teaching of reading literacy in South African primary schools reveal a number of gaps, such as low exposure to texts and poor reading instruction. For example, Van Staden and Bosker (2014:5, 8) point out that reading skills are introduced to learners too late at school, and little time is spent on reading instruction. She reports that the majority of Grade 4 teachers surveyed across the curriculum spent less than 2 hours on reading. Zimmerman (2014:7), in her observations of the teaching of reading comprehension in Grade 4 in a number of South African schools, found that the teaching of reading literacy is poor during the Foundation Phase at primary school. By the time they reach Grade 4, learners have a reading comprehension backlog accumulated over previous years of schooling, and as a result Grade 4 teachers have to teach at a slower pace and a lower level. Of even more concern is that Zimmerman (2014) found a number of teachers lacking the skills required to teach reading literacy, and particularly comprehension. Consequently students are inadequately prepared to read and comprehend texts independently (Zimmerman 2014:7). Zimmerman and Smit (2014) note similar trends in their studies and express particular concern about teachers’ poor understanding of teaching reading comprehension. In other words, they posit that a number of South African teachers do not succeed in generating the ability to comprehend reading material beyond information retrieval and do not encourage written responses (Zimmerman & Smit 2014:6). They also noted in their observation that successful teachers had declarative knowledge of teaching reading comprehension skills and utilised theoretical models of comprehension and a variety of reading comprehension strategies as needed in response to students’ needs (Zimmerman & Smit 2014:6).

Pretorius and Klapwijk (2016) explored the literacy practices of teachers in rural and disadvantaged areas and found they did not include appropriate reading strategies. This was confirmed by Boakye’s (2017a) study. Her survey showed that the majority of first-year sociology students stated that reading did not receive much attention in their schools, and instruction was for a limited time and given haphazardly. The students further confirmed that deep engagement with texts was not required at school level and they found a huge gap between reading at tertiary level and reading at school level (Boakye 2017a). Unfortunately, very little is done at tertiary level to bridge this reading gap. As Roberts and Roberts (2008) have noted, sociologists expect their students to be able to read, and do not take into consideration inadequacies or challenges students may be facing because of inadequate teaching at school.

At tertiary level, there has been limited research on the state of students’ academic reading ability. However, reading challenges are experienced by students all over the world. Bean (2011) identifies 11 reasons for students’ reading difficulties, such as schools (and the academy) that reward surface reading; lecturers’ teaching of reading material, rather than encouraging students to work independently; students’ difficulty in engaging with unfamiliar ideas; students’ inability to critically digest the argument of a text; students’ inability to understand and engage with the text; students’ resistance to spending time on deep reading; students’ difficulty with syntax and vocabulary; students’ inability to adjust their reading strategy according to the text’s genre; and lecturers’ and authors’ assumptions that students have access to ‘cultural literacy’. These issues result in a lack of ‘deep reading’, which involves conscious engagement with a text for learning and enjoyment purposes (Bean 2011:166). It also means that students do not internalise a text, an activity that Roberts and Roberts (2008) highlight as a mark of efficient reading.

Pretorius’ (2000, 2002) research, which focused on psychology and sociology students at tertiary level, suggested that students’ reading ability was in decline then, inter alia because of historical and socio-economic issues (2000:45). It seems after more than 10 years, it is still in decline (Boakye 2012). She (2000:43, 2002:169) contends that low reading literacy levels are predominant among L2 learners as a result of their poor inferencing and comprehension skills, which are essential for understanding texts and engaging with them. Because of these low reading literacy levels, a number of these students find it difficult to produce accurate or detailed representations of complex psychology or sociology texts (2000:45). Pretorius (2000:46). concludes that a lack of interventions allows these problems to persist, giving rise to a cycle of failure. In response, the current study seeks to provide ways of addressing these challenges by means of integrated strategies and to contribute to the limited body of research on reading at tertiary level.

Literature investigating the state of reading literacy in South African schools as well as the limited studies at tertiary level seem to indicate reading strategies as a focal point. According to Hosseini et al. (2012:1357), if the end product of reading is comprehension, then reading strategies can be understood as the ‘quintessence of successful comprehension … at any level of processing’. Appropriate use of reading strategies is important, as it plays a crucial role in assisting students to comprehend the texts they read. Hosseini et al. (2012:1361) advocate for the necessity of applying reading strategies, as it has a positive effect on reading comprehension and critical thinking, especially among second language learners. In other words, utilising reading strategies helps to facilitate deep reading, deep thinking, well-rounded understanding and internalisation of texts (Hosseini et al. 2012:1361–1362).

While there have been several interventions to counteract the South African literacy crisis at school level, there has been limited response to how this crisis affects tertiary education studies and how it should be addressed. Boakye (2017a) suggests a possible solution in terms of response theory. In her presentation of a multifaceted model for designing reading development programmes for L2 learners at tertiary level, she includes affective and social aspects (Boakye 2011). She points out that university students may not necessarily have the required literacy levels, especially if coming from disadvantaged backgrounds. As a result, they often struggle to cope with required reading (Boakye 2011:111). As a possible solution to this dilemma, she argues for literacy interventions that utilise an affective dimension in reading literacy instruction. This approach, she explains, encourages reader engagement through a focus on ‘motivation, attitude, interest, and self-efficacy’ (Boakye 2011:111). In addition, Guthrie and Wigfield (2000) argue for engaged reading as, according to them, it can compensate for the effects of ‘low family income and poor educational background’. In their engagement model, Guthrie and Wigfield (2000) posit that deep reading and better comprehension of texts can be achieved through engagement, which can be obtained through integrated reading instruction. Boakye (2011:117) adapts the model to the South African context and argues for a socio-affective approach to reading intervention. Her intervention with first-year tertiary students in an academic literacy module was shown to improve students’ affective reading levels, particularly the aspect of interest, motivation to read and deeper engagement with texts (Boakye 2017b:14). In addition, the intervention boosted the self-confidence of students in weak reading groups, which increased their motivation to read (Boakye 2017b:15). Thus, both high and low risk students were found to have improved in affective reading level and were motivated to read not only academic texts but also for pleasure. This points to the importance of reading interventions to be sensitive to socio-economic status and the affective dimension in reading literacy instruction.

In relation to reading texts in a specific subject, Roberts and Roberts (2008) investigated deep reading among sociology students and found that the majority of students do not engage in deep reading and neglect to read their course material. They highlight reasons why sociology students neglect to read their course material, and propose a possible solution to students’ apathy towards course literature. They (2008:125) argue that university students fail to read for understanding because high school education encourages regurgitation of information, rather than requiring students to find personal relevance in a text. According to Bloom’s taxonomy, memorising information is the least complex educational learning objective (Anderson et al. 2001). Education that focuses on retention of information results in ‘surface learning based in episodic memory or short-term memorisation for a day or two, rather than deep learning that is transformative of one’s perspective and involves long-term comprehension’ (Roberts & Roberts 2008:127). This superficial reading is further encouraged by the academy that offers rewards for basic memorisation, further fostering a ‘simplistic cost/benefit process’ that most often results in students doing the minimum reading for the maximum output (Roberts & Roberts 2008:129). The authors note that while content quizzes have been suggested as a remedy for this issue, quizzes are time-consuming and ‘reinforce the rational choice approach that is part of the problem’ (Roberts & Roberts 2008:130). These issues are particularly true for the first-year sociology students who are the subjects of this study.

In view of the issues discussed above, Roberts and Roberts advocate for the use of ‘reading responses’. This method requires students to read a text and summarise it or respond to it in one of several ways (Roberts & Roberts 2008:131–132). The various forms that a reading response can take are drawn from Howard Gardner’s forms of multiple intelligences in relation to Bloom’s more complex educational objectives. Students were assessed on their reading responses weekly while their examinations targeted ‘big ideas, analysis, synthesis, [and] evaluation’ (Roberts & Roberts 2008:133). Results from the study showed that the sociology students read more of their coursework and actually read with purpose (Roberts & Roberts 2008:134). While reading responses offer a short-term reward, they introduce students to reading strategies that will hopefully shift the focus away from an emphasis on achieving grades to ‘reading for the sake of learning from it’ (Roberts & Roberts 2008:134). This shift in focus can perhaps be achieved because reading responses aim to incorporate an affective dimension into reading by allowing students to work in modes with which they are comfortable, such as music and visuals. While this study is relevant and achieved success, the model may not fit seamlessly into the South African context, which comprises learners from different educational and social backgrounds who possess varying competencies.

Another intervention of similar approach was conducted by Parrott and Cherry (2011) using structured reading groups to facilitate deep learning (2011). Parrott and Cherry (2011:354) acknowledge the difficulty in getting sociology students to engage in deep reading of their coursework texts. However, they propose a slightly different approach, which assigns students to ‘reading groups’, each of which has a different, rotating role: discussion leader, passage master, devil’s advocate, creative connector and reporter (Parrott & Cherry 2011:354). The groups meet throughout the semester, and before each meeting students must complete a number of readings and then prepare for the role they have been assigned (Parrott & Cherry 2011:354). The strategy has two underlying ideas: firstly, to encourage students to commit to deep reading, and secondly, to use the information they have engaged with for a particular purpose, such as discussing ‘hot topics’ (Parrott & Cherry 2011:354–355). In studying the efficacy of these reading groups, Parrott and Cherry (2011:360) found the method not only popular among students, but it also increased the likelihood of students reading course material, helped them to understand the texts and helped them to make connections between the information and everyday life (Parrott & Cherry 2011:361). In other words, the reading groups helped students achieve the course outcomes of ‘developing a deep understanding of course material, using course concepts to better understand social processes, and being able to use class information to discuss current social issues’ (Parrott & Cherry 2011:354–355). The reading groups also fostered a sense of accountability that motivated students to do their readings, as not doing so would mean letting the group down (Parrott & Cherry 2011:361). The authors also argue that this method improves upon the method previously proposed by Roberts and Roberts (2008) in that the work is done not individually but also in groups, which teaches teamwork skills (Parrott & Cherry 2011:355). A similar method was employed in this study.

These teaching techniques highlight the affective aspects of reading as a means of improving students’ reading comprehension. According to Grabe and Stoller (2011), the affective dimension is pertinent to reading comprehension as it contributes to students’ willingness to read frequently and to become involved in their reading. The affective factors focused on in this study are motivation, self-efficacy, attitude and interest.

Motivation is defined as ‘the individual’s personal goals, values, and beliefs with regard to the topics, processes, and outcomes of reading’ (Guthrie & Wigfield 2000:405). In order to increase students’ motivation, they were required to read and share their understanding, taking on different roles. Bandura (1986:391) describes self-efficacy as ‘people’s judgements of their capabilities to organise and execute’ various actions in order to achieve certain goals. Thus, students’ self-efficacy can be understood as how they view themselves as readers: whether they believe they are competent and successful readers or not. Students’ self-efficacy levels were targeted through scaffolding of the texts to be read, as well as discussions that took place in small groups among peers where students felt comfortable to share their views.

Attitude can be perceived as ‘a sense of “liking” or a continuum with positive and negative extremes’ (McKenna 2001:147). McKenna (2001:147) suggests there are two parts to reading attitude, namely attitude towards reading for pleasure and attitude towards reading of academic texts. The intervention sought to instil a positive attitude towards reading of academic texts. While interest relates to two forms of curiosity, namely personal and situational, personal interest is perceived as internal and is the ‘enduring attraction to a topic even before a particular text is read’ (Boakye 2017b; Hidi & Anderson 1992:216; Schiefele, Krapp & Winteler 1992:152). Situational interest, on the other hand, is ‘short-lived emotional states’ drawn from a particular context (Hidi & Anderson 1992:216). Interest therefore differs from attitude and motivation, as it does not rely on particular goals or past experiences to be present in an individual (Hidi & Anderson 1992:216). In order to develop students’ interest, texts were subjected to lively discussion and connections were drawn between texts and students’ personal and everyday life issues.

Boakye (2017b:4) posits that by giving attention to these affective factors students develop the desire to read. They are able to form personal connections with the texts and enjoy reading. The cognitive outcome of the development of affective reading is the application of appropriate reading strategies, which leads to efficient reading and deeper comprehension (Boakye 2017b:4; Grabe & Stoller 2011; Guthrie 2008). This is particularly important for non-native English speakers and students from low socio-economic backgrounds who, because of limited exposure to reading materials, have low reading literacy levels and consequently become easily frustrated while reading, which causes them to develop negative attitudes towards reading (Pretorius 2000).

In view of the issues discussed and reading challenges of students at tertiary level, as well as the limited research on reading at tertiary level, an intervention was undertaken with sociology students to improve their reading of sociology texts, using the cognitive strategies in Parrott and Cherry (2011), integrated with affective strategies.

The aim of the study was to assess the effect of innovatively combining cognitive and affective strategies through tutorials with a view to improving students’ reading of sociology texts.

The following research questions were posed for the study:

The methodology comprised both quantitative and qualitative analysis. Although data were collected through verbal means (open-ended questionnaire), the data were analysed using both qualitative and quantitative methods. Verbal data were analysed using content analysis to provide percentages, and specific examples were used to support the resulting percentages. The methodology therefore comprised both quantitative and qualitative analysis. According to Hsieh and Shannon (2005:1278), in conventional content analysis, verbal data can be coded into categories and then described using statistical analysis. They state that this approach is referred to by Morgan (1993) ‘as quantitative analysis of qualitative data’ (Hsieh & Shannon 2005:1278). They point out that such a design would aim to describe a phenomenon, in which researchers would avoid using preconceived categories but rather allow categories to flow from the data.

The participants were first-year sociology students at the University of Pretoria. The students were registered for various programmes in different faculties. Only a limited number of students proceed to major in sociology; the majority take sociology as an ancillary subject at first-year level. Although they had lectures twice a week in a large group, they also had to attend a 1 h tutorial consisting of about 20–25 students once a week to discuss texts and topics taught during lectures. The tutorials were conducted by postgraduate students who had taken sociology as a major subject or obtained a high mark for their first-year sociology exam. A workshop was conducted to train tutors in the integrated instructional approach of the intervention. Although 425 students were registered for the first-year sociology module, only 254 completed the questionnaire. Other students may not have attended tutorials during the week of data collection or may have opted not to complete the questionnaire because it was voluntary.

The intervention was organised around prescribed texts. The first semester sociology course comprised readings on Emile Durkheim, Max Weber and Karl Marx. In previous years, tutors merely attempted to explain the texts to the students during tutorials. During the intervention, excerpts were taken from sections of these readings. Students were divided into groups, and the group leader had to assign a role to each member. The different roles were first explained by the tutor to the group as a whole. Depending on his or her assigned role, a student had to summarise the text in his or her own words, find meanings of unfamiliar words and concepts, or connect the text to everyday life issues. During tutorials, students engaged with the texts, with each member contributing according to his or her assigned role in a non-threatening environment. Each group then presented to the whole class, followed by a class discussion of the texts, with the tutor providing positive feedback to guide the students. The intervention lasted for 12 weeks and came to an end when the tutorial sessions for the semester ended. At the end of the intervention, students’ opinions and experiences of the intervention were elicited by means of a questionnaire. The questionnaire consisted of a Likert scale section that focused on specific issues and an open-ended section that aimed to determine students’ opinions on the intervention. The open-ended section, which is reported on in this article, required students to comment on and explain how the tutorials and the role-play reading activities influenced their: (1) reading of sociology texts; (2) understanding of concepts; (3) writing of sociology assignments; and (4) writing the sociology examination. They were also asked to provide any additional comments they wished to make regarding the intervention.

During the first week of the second semester, the questionnaire was administered during tutorials to elicit students’ opinions and experiences of the techniques used to facilitate comprehension of the texts during the first semester. Ethical requirements were complied with. In addition to obtaining ethical approval from the Faculty of Humanities, students were assured of anonymity. They were alerted to the fact that their participation was voluntary, and they had to sign an informed consent form to indicate their willingness to participate.

Students’ verbal responses to the open-ended questions were analysed using emerging themes. The categories or themes therefore emerged from the data, as explained by Hsieh and Shannon (2005). The predominant themes were first identified and coded. They were then counted, calculated and presented as percentages to show their predominance.

Some questions were not answered by some of the students, so although there were 254 students the total number of answers for each question was different. For each question: (1) reading of sociology texts; (2) understanding of concepts; (3) writing of sociology assignments; and (4) writing the sociology examination – predominant themes were identified and counted. The total for each of the five emerging themes in each question was calculated as a percentage. In other words, the number of times a specific theme occurred was divided by the number of students who had answered that question and multiplied by one hundred. The major themes in the responses were identified as follows: (a) the techniques helped students to understand what they read, even difficult concepts; (b) the group discussions in particular were helpful; (c) the mind-map and role-play activities aided understanding; (d) students initially struggled with the assigned readings but the tutorial activities helped them overcome the challenges; (e) the intervention was not helpful; (f) students were still having difficulty with reading the texts. Question 5 required students to provide any other information they wished to add.

In presenting the findings, examples for each theme or category are provided from the data (Hsieh & Shannon 2005:1279). Thus, in addition to the percentages, verbal examples are provided to shed more light on the predominant emergent themes.