Adding Voices to Physical Sciences Preservice Teachers’ Approaches to Learning

Abstract

Students’ approaches to learning have a significant impact on the outcomes of any educational programme. In the light of the documented concerns about the quality of students’ engagement with academic work, this study sets out to determine and explain preservice teachers’ (PSTs) approaches to learning. This was based on the premise that, while science teacher-educators may strive to inculcate a deep approach to learning in their PSTs, various contextual factors may impinge on these efforts. Ninety-four Physical Sciences PSTs’ approaches to learning were explored through a survey questionnaire followed by 32 face-to-face interviews. The approaches to learning theory was used to develop the instruments and interpret the data. Mean agreement scores were used to determine the students’ approaches to learning by profiling them based on a mean agreement scale. The findings indicated that most (68%) of the Physical Sciences PSTs tended to use combinatory pragmatic dissonant and deep-strategic approaches to learning. Several factors influenced the PSTs’ approaches to learning but their socio-economic status and motivation to train as teachers seemed to have the greatest effect. It is imperative for lecturers to explore strategies to promote a deep approach to learning and high self-efficacy in students with low socio-economic status having to balance part-time work and studying. A more extensive study might help to ascertain the correlation between the motivation to train as teachers and students’ approaches to learning.

Keywords:

• Approaches to learning
• deep approach
• strategic approach
• surface approach
• dissonant

Introduction and Background

The number of students enrolling to train as science teachers in South Africa has been increasing rapidly over the years owing to increased government funding. The aim is to alleviate the shortage of science teachers in the country. While such efforts are being made, there have been concerns about the quality of the science teachers produced by universities. The science teachers’ competence in general and subject content mastery in particular have been questioned by various stakeholders (Botha & Reddy, 2011; Ramnarain & Fortus, 2013). University lecturers tasked with science teacher training have raised concerns about the level of engagement with academic work and performance demonstrated by preservice science teachers (PSTs) (Stears & James, 2014), and this puts into question their approaches to learning as these have a bearing on the learning outcomes.

Research on approaches to learning employing quantitative methods has been largely restricted to profiling students’ approaches to learning (Brown, White, Wakeling, & Naicker, 2015; Buckley, Pitt, Norton, & Owens, 2010; Entwistle, McCune, & Tait, 2013). Students’ voices regarding why they use specific approaches to learning have not been adequately exploited. By simply using an established theoretical framework on approaches to learning and statistical data to categorise students, researchers assume that approaches to learning are organic from an academic perspective, thus obviating the existence of other factors. We argue that approaches to learning should be viewed in context and that statistical inferences should be complemented by the participants’ own narratives of how they approach learning. A profile of students’ approaches to learning coupled with an explanation of their choices can help lecturers to devise appropriate teaching strategies and remedy any perceived negative learning traits. This is of particular importance because if PSTs consciously adopt a deep approach to learning, this will have a positive impact on their quality as graduate teachers.

The key research questions were: (a) what are Physical Sciences preservice teachers’ approaches to learning and (b) why do Physical Sciences preservice teachers use specific approaches to learning?

Theoretical Framework

Pioneering work by Marton (1976) led to the development of the approaches to learning theory, identifying two dichotomous approaches to learning, deep and surface. The theory has since been expanded over the years by other scholars to include the strategic (achieving) approach (Biggs, 1987; Entwistle et al., 2013; Entwistle & Ramsden, 1983). Approaches to learning describe students’ engagement with academic work based on their beliefs about learning, their motives and achievement targets (Entwistle & Peterson, 2004). Entwistle and Peterson further opine that variations in study behaviour can also be associated with approaches to learning, self-regulation and academic processing strategies. The deep approach is associated with seeking meaning, whereby students purposefully apply methods to achieve this intention when attempting to understand ideas for themselves. Under the strategic approach, students put effort into organised studying with the intention of achieving good grades in their modules. Students associated with a surface approach are characterised by reproducing module content in order to cope with module requirements. The constructs that characterise each approach to learning are summarised in Table 1.

Table 1. Approaches to learning and corresponding constructs

Approach	Constructs
Deep	Meaning seeking (to gain a deep understanding of ideas) Relation of ideas (to prior knowledge and experience) Use of evidence (looking for patterns and underlying principles) Interest in learning the subject content
Strategic	Organised (systematic organisation of studying) Time and effort management Alertness to assessment demands and criteria Achievement of good grades Monitoring of the effectiveness of study methods
Surface	Lack of purpose (seeing little value in the course and assessment tasks) Unrelated memorising (routinely memorising facts and procedures) Syllabus-boundness (a narrow focus on the minimum course requirements) Fear of failure (feeling undue pressure and anxiety about academic work)

Adapted from Entwistle and Peterson (2004, p. 415).

In practice, students may adopt specific approaches to learning as a response to the context in which they learn (Buckley et al., 2010; Entwistle & Peterson, 2004). The teaching–learning context, which refers to how the teaching–learning environment is designed and how module content is selected, organised, presented and assessed, influences approaches to learning (Entwistle & Peterson, 2004). While students may enter university with well-established approaches to learning, these can evolve through the years as influenced by their context (Laurillard, 2005). In their suggestions for dealing with large classes, Hornsby, Osman, and de Matos-Ala (2013) argue that educators should manipulate the learning environment to promote a deep approach to learning. In this regard, Hurtado et al. (2011) found that patterns of science faculty educators’ engagement with students can make a significant difference in promoting a culture of support while still maintaining rigour in the mediation of the curriculum.

Student characteristics also play a crucial role in their approaches to learning. These include previous experiences of teaching and learning, their conceptions the process of knowledge construction and their exposure to instruction on pedagogy (Entwistle & Peterson, 2004). Students attending a course are not always homogeneous. The characteristics of the students will vary from place to place and these will influence students’ approaches to learning in different ways.

Literature Review

Globally, research on approaches to learning has mostly focused on categorising students by approach and determining linkages with academic performance. Generally, a deep approach has been found to be strongly correlated to a better acquisition of cross-curricula generic skills such as problem solving and critical thinking (Goh, 2016) but some scholars have established a weak positive correlation between the deep or strategic approaches to learning and academic performance (Nordin, Wahab, & Dahlan, 2013). Researchers have also deliberately modified the teaching–learning environment with the intention of increasing the use of the deep approach to learning and achieved positive results with preservice teachers (Gordon & Debus, 2002).

Research in African contexts has revealed that the deep approach is the dominant and preferred approach to learning amongst science students (Matoti, 2014; Rollnick, Davidowitz, Keane, Bapoo, & Magadla, 2008). These longitudinal studies have also shown that the deep approach to learning develops with academic experience. This may be because the students would have adapted to the challenges of tertiary education such as work pressure, an overloaded curriculum and assessment procedures. While students tend to self-report the use of a deep rather than a surface approach to learning, in some cases students use a surface approach when they find it convenient (Brown et al., 2015). This might happen when they view a specific course as less significant in the broader scope of their studies. They therefore resort to rote learning just to get done with it. A shift from deep to a surface approach over an academic year was reported by Mogashana, Case, and Marshall (2012). Through interviews, Mogashana et al. (2012) established that data collected using the Approaches to Learning and Study Inventory were not always accurate owing to students’ (in)ability to decipher entire or part of the inventory statements.

Some scholars with an interest in South African preservice science teachers have preferred to focus on aspects that might impact on approaches to learning, such as learning strategies (Dukhan, 2018), knowledge and teaching strategies (Vhurumuku & Chikochi, 2017) and professional learning experiences (Nkambule & Mukeredzi, 2017). While focusing on university first year students’ note-making in Biology, Dukhan (2018) found out that high school experiences strongly influenced the quality of their notes, which in turn impacted on their performance. Although there was no direct focus on approaches to learning, it can be inferred that poor note-taking might be symptomatic of a surface approach to learning, hence the observed poor performance. Vhurumuku and Chikochi (2017) experimented with two distinct teacher approaches with special focus on developing knowledge and teaching strategies with respect to the nature of science. With the experimental group of in-service teachers they used a capsular approach in which important historical, philosophical and sociological aspects of science and scientific literacy were covered before dealing with the teaching strategies relevant to nature of science. The control group was subject to a heurist approach, which focused on the teaching strategies in isolation. They found the capsular approach to be more effective. Making connections between teaching strategies and various elements of science and scientific literacy could be viewed as promoting a deep approach to learning while the heuristic approach inadvertently promoted the surface approach to learning.

Students’ approaches to learning do not always conform to theory. Gijbels, van de Watering, Dochy, and van de Bossche (2005) asserted that a group of second-year Dutch law students employed dissonant approaches to learning. Dissonant refers to the use of incompatible approaches to learning by one person. This is induced by one or more factors. Contextual factors, such as socio-economic status (SES), largely defined by family income, have a strong correlation with academic achievement (Mirowsky & Ross, 2003; Yoshikawa, Aber, & Beardslee, 2012). Students from poor SES have been found to underperform consistently in comparison with students from high SES (Maringe & Sing, 2014). In South Africa, inequality arising from apartheid is linked to poor SES and low achievement of the largely African student population (Spaull, 2013). While the surface approach is more common, students from medium to high SES use the deep and strategic approaches more frequently than low SES students (Aharony, 2006).

Other factors that influence students’ approaches to learning are self-efficacy, intrinsic (and extrinsic) motivation, goal orientation and outcome expectations (Zimmerman, 2002). Prat-Sala and Redford (2010) found intrinsic and extrinsic motivation to be correlated with approaches to learning. They also established that students with high self-efficacy with respect to reading and writing often used a deep or strategic approach to learning while their low self-efficacy counterparts often resorted to a surface approach. Results from a study by Bolkan, Goodboy, and Griffin (2011) suggest that using interactive teaching styles influences students’ intrinsic motivation through the use of intellectually stimulating behaviours and encourages deep conceptual understanding rather than just doing enough to pass, graduate, get a job and earn an income.

The literature reviewed points to the fact that approaches to learning are shaped by factors within and external to the learning environment. However, educators can deliberately manipulate contextual factors within the learning space to induce a deep approach to learning. Student characteristics may act as a significant determinant of the observed approaches to learning and these may exist without the knowledge of the lecturers. It is therefore incumbent for scholars in education to go beyond determining approaches to learning by probing students to fully understand why specific approaches are prevalent (Mogashana et al., 2012).

Research Design and Methods

An explanatory mixed-methods approach was used to accomplish this study (Creswell & Clark, 2011). Quantitative data were collected using a questionnaire and interviews were used to generate qualitative data. In the first phase, the PSTs completed a questionnaire whose purpose was to determine their approaches to learning. In the second phase, qualitative data were generated to explain the quantitative results. By dialoguing through interviews conducted with the PSTs, it was possible to understand their ‘thought process’ (Atkins & Wallace, 2012) and explain why they preferred specific approaches to learning. The data collection process took place at the end of the second year of study based on the understanding that, after two years of tertiary education, the PSTs would have defined approaches to learning at this level.

The Participants

The participants targeted for this research were a cohort of 105 Physical Sciences PSTs in their second year of a four-year Bachelor of Education degree programme at the School of Education of a particular South African university. Ninety-four of them volunteered to take part in the study. Ninety of the 94 participants were native Africans and four were of Indian descent. Sixty-one were males and 33 were females. Seventy of the participants came from rural areas and the rest were from traditionally black townships. All the participants were funded by a government scholarship programme aimed at solving the shortage of science teachers in the country. The funding included a monthly subsistence allowance. Eighty of the 94 stayed in university halls of residence and the remainder commuted to campus from home. Permission to conduct the research was granted by the registrar and ethical clearance was obtained from the research ethics office of the concerned university. All of the 94 participants completed a survey questionnaire aimed at determining their approaches to learning. An email was sent to all the 94 PSTs inviting them to participant in face-to-face interviews. Thirty-two PSTs responded to the email and consented to be interviewed. The breakdown of the interviewees is described later. The participants’ names were coded to guarantee anonymity, for example PST301 (Cohen, Manion, & Morrison, 2018).

Instruments and Procedures

The questionnaire used to collect quantitative data for the current study was adapted from a 52 item questionnaire, the Approaches to Study Skills Inventory for Students (ASSIST), developed and validated by Entwistle et al. (2013). The original questionnaire items were technically edited in some places in order to use terms that were familiar to the participants. For example, the term course was replaced with module, which is commonly used in the South African context. The participants rated themselves on a scale of 1–5 (1 = disagree; 2 = disagree somewhat; 3 = unsure; 4 = agree somewhat; and 5 = agree) for each of the 52 items in the ASSIST questionnaire. Each construct under the different approaches to learning (Table 1) was linked to a set of items on the questionnaire. For example, the set of questionnaire items linked to the construct seeking meaning were:

4. I usually set out to understand for myself the meaning of what we had to learn.

17. When I read a recommended article or book, I tried to find out for myself exactly what the author meant.

30. When I was reading something, I stopped from time to time to reflect on what I was trying to learn from it.

43. Before tackling a problem or research assignment, I first tried to work out what is inherent in it.

An interview schedule, based on the constructs in Table 1, was designed and validated by the researchers. The interview items were refined during two meetings before asking a colleague outside the research team to review it. This was meant to eliminate leading and closed questions (Atkins & Wallace, 2012). All of the interview participants consented to audio-recording. The interview sessions lasted an average of 25 minutes each. One member of our research team conducted all of the interviews to ensure consistency in asking the questions. The interviewer did not teach the participants, which promoted confidentiality and freedom of expression. The audio-recordings were transcribed and made available to the participants for member checking to enhance the credibility of the data (Cohen et al., 2018).

Data Analysis

The quantitative questionnaire data were processed using the Statistical Package for Social Sciences version 25 (Cohen et al., 2018). The items falling under each construct were collated in order to calculate the composite mean agreement scores and standard deviations (Nordin et al., 2013). The composite mean agreement scores for the constructs were then used to determine the mean agreement scores for each approach to learning. A mean agreement scale [1–1.99 = very low (V); 2.00–2.99 = low (L); 3.00–3.99 = moderate (M); 4.00–5.00 = high (H)] was used to profile each PST across all approaches. Table 2 shows the rubric used to determine each PST’s approach. The deep-strategic and dissonant approaches were retrospectively done after the quantitative analysis. A deep-strategic approach occurs when students seek a deep understanding of concepts but also systematically work on achieving good grades. A dissonant approach emerges when one has similar mean agreement scores for incompatible approaches.

Table 2. Rubric for profiling PSTs’ approaches to learning

Resultant approach	Possible combination of mean agreement scores (deep, strategic, surface)
Deep	HMM, HML, HLL, HMV, HVM, HVV, MLL, MLV, MVV, LVV
Strategic	MHM, MHL, LHM, LHL, VHL, VHV, LHV, LML, LMV, VML, VMV, VLV
Surface	MMH, MLH, LMH, LLH, VLH, LVH, VVH, LLM, LVM, VLM, VVM, VVL
Deep-strategic	HHM, HHL, HHV, MML, MMV, LLV
Dissonant	HHH, HMH, HVH, MMM, MLM, MVM, LLL, VVV

Description of mean agreement scores: high (H); medium (M); low (L); very low (V).

The qualitative data from the interviews were subjected to thematic analysis (Nowell, Norris, White, & Moules, 2017). This involved working through the textual data iteratively to establish a comprehensive set of explanations for the PSTs’ choices of approaches to learning (Creswell, 2014). Each research team member coded the data individually first, before the themes were consolidated. This approach ensured that the data analysis was thorough and no salient themes were left out.

Results

PSTs’ Approaches to Learning

The PSTs could not be conclusively linked to a definite approach to learning (Table 1) as a collective owing to similar group mean agreement scores. The group means for the three approaches to learning were found to be moderately high. However, the means for the deep and strategic approaches to learning were the same (M = 3.70) with comparable standard deviations (SD =  0.47 and SD =  0.46 respectively), implying that even the spread of the mean agreement scores was similar. The surface approach had a slightly lower mean agreement score and higher standard deviation (M = 3.33, SD =  0.49) compared with the deep and strategic approaches. Owing to the blurred boundaries between the approaches, using the highest mean agreement score to classify the PSTs was problematic. We therefore profiled the PSTs based on a mean agreement scale (Table 2). Only 30 of the 94 PSTs used definite (deep, strategic or surface) approaches to learning. Each of the remaining 64 used either a combination of compatible (deep and strategic) or non-compatible (dissonant) approaches. Table 3 shows the frequencies of the approaches to learning together with the profiles of selected PSTs as illustrations.

Table 3. Frequency of approaches and examples

Approach to learning	Frequency (%)	Exemplar PST	Mean agreement scores			Profile of exemplar PST
	(n = 94)		Deep	Strategic	Surface
Deep	13 (14%)	PST301	4.13	3.90	2.63	HML
Strategic	12 (13%)	PST302	3.81	4.15	3.31	MHM
Surface	5 (5%)	PST331	2.88	2.05	3.69	LLM
Deep- strategic	20 (21%)	PST381	4.25	4.30	2.94	HHL
Dissonant	44 (47%)	PST384	4.06	3.95	4.13	HMH

Mean agreement scale: 1–1.99 = very low; 2.00–2.99 = low; 3.00–3.99 = moderate; 4.00–5.00 = high.

Reasons for Using a Particular Approach to Learning

The 32 PSTs that were interviewed were distributed as follows: deep, 6; strategic, 5; surface, 4; deep-strategic, 7; and dissonant, 10. As alluded to earlier, the interview data allowed us to give meaning to the quantitative data.

The dissonant approach

Forty-seven per cent of the PSTs were found to be dissonant in their approach to learning. Incompatible combinations of approaches such as deep-surface and surface-strategic were reported as being linked to the perceived level of difficulty of specific topics, the type and availability of resources and the nature of assessments.

One group of PSTs using a dissonant approach tended to abandon deep in favour of surface learning when they found the topics difficult. They apparently resorted to rote learning as a survival strategy. This group of PSTs consistently reported that the availability of lecture presentations, past examination papers, module outlines and planners on the e-learning platform assisted them to be strategic (focus on what they thought was of interest to the lecturer) and surface oriented (memorising content on lecture PowerPoint presentations) as they found convenient. They expended effort on doing what they deemed necessary to pass the course. A typical rationale was voiced by PST334:

Some topics like Hess’ Law and buffer systems were new to me and very difficult so, I just had to memorise the slides used by the lecturer. Equilibria and Gas Laws are easy so, I used a different approach, I deepened my understanding from high school. The past papers also helped, you know how the lecturer asks questions and I prepared accordingly. I used the lecturer presentations on the e-learning platform. They summarised everything we needed to know (for tests and examinations). I also used past examination papers and marking guidelines. It worked for me in high school. Sometimes I use my cell phone to record the lectures to make sure I get it as it is from the lecturer. If I use this strategy I know I will get at least a 50%. (PST334)

A second group of PSTs using the dissonant approach to learning linked this to their weak socio-economic background. They did part-time jobs to earn income for subsistence, leaving them with limited time to study, hence they replaced deep by surface learning, again relying on resources provided by the lecturer such as notes, the course outline and the module planner, as illustrated by PST384:

I had to balance my studies with part-time work so, sometimes I didn’t have enough time to study. Some topics were difficult, so I forgot about the deep understanding the lecturer preached about and just memorized the notes from the lecture presentations. The course outline and module planner provided by the lecturer helped because you know what to focus on. My father chose the teaching profession for me because he thought there are better prospects of getting a job. I therefore have to pass by all means in order to graduate and change my family’s life. (PST384)

The deep-strategic approach

The deep-strategic approach, used by 21% of the sample, was apparently extrinsically motivated. These PSTs were motivated by awards offered to high performers every year, the need to pass in order to retain funding or secure funding for postgraduate studies. A deep understanding of the concepts was considered ideal but the need to achieve good grades was paramount as expressed by PST320:

When you are on a government bursary, they expect you to perform. I want to get a good understanding of the concepts but I also want the awards so I practise a lot using the questions in the prescribed textbook and past examination papers after studying any topic. I also want to come back for Honours and good marks will help me get a place and funding. I therefore have to know how the lecturer examines by going through past papers. (PST320)

The deep approach

The 14% of PSTs who used a deep approach were apparently intrinsically motivated. They wanted to become lecturers in future and knew that they needed a good mastery of the subject content. Alternatively they wanted to be competent Physical Sciences teachers upon graduation. A teaching approach with a bias towards practical work helped to develop a deep approach to learning. Contextualised assessment tasks (tests and examinations) also helped to foster a deep approach. They used learning tools such as concept maps and YouTube videos with simulations. PST313 said:

I want to be a lecturer one day so, I work hard to get a deep understanding of the concepts in each topic. To achieve this I must get good marks and qualify for Honours. The practical work we did this semester was about the content we did in lectures. It helped me to have a very good understanding of Hess’ law, rates of reactions and other topics. In tests and examinations, the lecturer asks questions based on contexts. It’s difficult to pass if you don’t have a good mastery of the concepts, and I have to be able to apply my knowledge to pass. So, I use concepts maps to summarise each topic. I also watch a lot of YouTube videos which have simulations and illustrations to improve my understanding. (PST313)

Strategic approach

The strategic approach was used by 13% of the participants. They were driven by the need to retain funding given that they were from poor socio-economic backgrounds as exemplified by PST302’s words:

Retaining funding for my studies is very important for me. I have to keep getting good grades so that I don’t have to worry about tuition. My parents do not have proper jobs so they won’t be able to pay for my education if I lose my funding. I make sure I know the type of questions asked by the lecturer in tests and examinations. I therefore keep a portfolio of the course outline, past papers and marking guides for my revision and exam preparation. (PST302)

The surface approach

The 5% of PSTs who resorted to a surface approach found the content entirely new and difficult. A reported novelty in the content studied and deficiencies in English (the language of teaching and learning) apparently promoted a surface approach with a view to achieving the minimum pass mark. PST331 described how rote memorisation worked for him:

The course content was very difficult. Some of the topics were completely new to me. We never did Hess’ Law, buffer systems and order of reaction in high school. My English is not good so, I struggled to read the textbooks on my own. I simply memorized the lecture slides. It helped me to get at least a pass mark but I don’t understand most of the content. It worked for me in matric and I managed to pass Physical Sciences. (PST331)

Other PSTs using a surface learning approach found the content very difficult and the use of technology in learning challenging. They found it challenging to access course material on the e-learning platform using computers. PST380 expressed difficulties in adapting to a ‘paperless’ context where hardcopies of lecture notes were not provided.

We are experiencing new things. Like learning using the computer is very difficult. It’s my first time to use a computer for everything related to my studies. So, I think we have more challenges in our studies. I usually don’t find the relevant books in the library. Books would be better for me because I can read them to and from work after lectures. I can’t carry a computer with me to work because of the robbers on the streets. (PST380)

Discussion

The current study established that the approaches to learning identified from the quantitative data resulted in five coherent groups of PSTs with similar reasons for using the respective strategies. Sixty-eight per cent of the PSTs used dissonant or deep-strategic approaches to learning and the remainder used definite approaches to learning owing to the influence of SES, motivational factors and teaching–learning experiences. The emergence of compound approaches to learning from students’ narratives in the current study gives credence to the argument that inventories on their own may not give a holistic understanding of the phenomenon (Mogashana et al., 2012).

SES

The SES tended to influence the PSTs to adopt dissonant approaches similar to what Gijbels et al. (2005) found for students in Belgium. Doing part-time work meant reduced study time and the need to retain funding for studies apparently led the PSTs to consciously and conveniently using varying combinations of approaches to enable them to complete their degree programme, get a job and assist their families financially. Thus, their learning approaches can be described as pragmatic-dissonant. The deep-strategic approach arose out of a need to retain or secure funding for studies. This approach can therefore be described as pragmatic deep-strategic. The pragmatic approaches arose from the lived experiences of the PSTs. The PSTs dealt with their learning sensibly and realistically based on practical rather than theoretical considerations.

Motivational factors and self-efficacy

The PSTs who used a deep approach apparently had high self-efficacy and were intrinsically motivated (Prat-Sala & Redford, 2010) by ambitions to pursue postgraduate studies and become lecturers or competent teachers. Those who fell under the surface approach seemed to have low self-efficacy owing to, for instance, deficiencies in the language of teaching and learning or technological skills. PSTs did not always practice the deep approach to learning even though they knew that this was the best approach. It is regrettable that surface learning practices developed in lower levels of schooling remained entrenched in students at tertiary levels of education even in the wake of different teaching–learning experiences (Duff & McKinstry, 2007). The categorisation and distribution of students by approach differ from the findings by Rollnick et al. (2008) perhaps owing to different degree programmes, our inclusion of the strategic approach and ‘adding voices’ to explain the use of each approach.

Teaching–learning experiences

Practical work linked to the theory covered in lectures, contextualisation of concepts and student-centred tutorials reportedly fostered a deep approach. Challenges encountered within the teaching–learning environment coupled with individual student characteristics (Entwistle & Peterson, 2004) seemed to influence the emergence of the surface-strategic approach to learning. The nature and availability of course learning material, the novelty of the course content, previous teaching and learning experiences, technological deficiencies and poor English proficiency apparently constituted a collection of factors which influenced the PSTs to adopt a surface approach. While prior studies have suggested that the flexibility and variety offered by technology enhances the learning experiences and chances for students with poor socioeconomic backgrounds to complete degree programmes (Devlin & McKay, 2016), our findings suggest that this could be a hindrance in adopting a deep learning approach.

Conclusion

The findings of the current study show that contextual factors can lead to the emergence of pragmatic approaches to learning. The SES apparently resulted in pragmatic-dissonant and pragmatic deep-strategic approaches while motivational factors and teaching–learning experiences invariably led to definite approaches. It can be argued that these contextual factors are not peculiar to the participants of this study. Students from economically disadvantaged backgrounds all over the world often receive scholarships to study far away from home, including in foreign countries. The need to work outside lecture times to generate extra income for subsistence often leads to compromises that shape students’ approaches to learning. Perhaps a much broader study might be useful in understanding the approaches to learning for such students. Such an understanding would be useful to lecturers who work with these students since approaches to learning often have an impact on academic performance. The challenge for lecturers lies in designing strategies, which promote a deep approach to learning and good academic performances. Capable students from low SES can be identified and appointed as tutors to avoid them working outside campus, which would cut out travelling time and increase time available for studying. Furthermore, tutoring others potentially promotes deep learning. More research is required regarding the PSTs’ motivation to train as teachers as this is one of the key factors driving their approaches to learning. Promoting self-efficacy by working towards eliminating perceived challenges linked to technology usage and competency in the language of teaching and learning may reduce surface learning. We recommend a comprehensive computer literacy course for first year students so that they acquire the necessary knowledge and skills to access and use resources on e-learning platforms. Upon graduation, technology-competent teachers will be able to use technology in their classrooms and to better-equip high school learners with life-long learning skills that are required at tertiary level and in the workplace. Although the surface approach had a low frequency, there is need to understand why English proficiency remains elusive two years into university education given that the language is taught from primary school.

Acknowledgements

This research project was funded by the University Teaching and Learning Office (UTLO), University of KwaZulu-Natal.

ORCID

Tamirirofa Chirikure http://orcid.org/0000-0002-9808-7671

Lebala Kolobe http://orcid.org/0000-0002-7791-4850

Sebenzile Ngema http://orcid.org/0000-0001-5609-2993