Tell me, I forget. Show me, I remember. Involve me, I understand.
- Confucius
What is Below?
If You Only Read One Paper, Make it This
The Breakthrough of Deep & Surface Learning
What Deep & Surface Learning Actually Is
The Factors Influencing Depth of Learning
The Outcome: What's Best, Deep or Surface?
A Practical Example: Deep Learning in Schools
How the Theory Overlaps with Other Learning Theories
If you only read one thing, make it this: Haggis, T. (2003) Constructing Images of Ourselves? A Critical Investigation into ‘Approaches to Learning’ Research in Higher Education. British Educational Research Journal, 29, 1, 89-104.This article gives a readable summary of the main aspects and ideas of deep and surface learning; as well as a description of the key experimental studies. The paper also gives an account of implications of the research in relation to higher education, as well as providing criticisms and limitations of the deep and surface learning theory.
Background Study: The Breakthrough
Pre- 1976, there was no common theory that could explain the different approaches to learning. A ground-breaking series of experiments by Marton and Säljö (1976) at the University of Gothenburg aimed to detect different levels of processing whilst learning, by interviewing Swedish students who were instructed to read an article. After reading, the instructors posed questions regarding the meaning of the passage and their experience whilst reading including: “Did you find the passage interesting?”, “While reading, was there anything that struck you as particularly important?”,“Could you describe how you went about reading the passage?” and “Did you find the text interesting?”. The students participated in individual sessions, with their conversations transcribed for analysis.
The researchers categorised the answers, and interpreted from them two distinct types of processing which the learners engaged in: deep and surface. Surface learners were defined as those only focused on grasping the main points and memorizing them. Conversely, deep learners explored the meaning behind the topic, and attempted to relate it to other knowledge to aid their understanding. Follow up examinations by Entwistle (1981), Ramsden (1987) and Biggs (1993) elaborated on the theory, and substantiated the research.
What Exactly is Deep and Surface Learning?
From the initial study it was concluded that surface learners tended to concentrate on the writer’s main point, and reproducing the main facts. Associated with anxiety in learning, the surface learners were dispassionate with the material. The students felt pressurised and rushed into retention of information; thus they only memorised and did not understand the meaning behind the material. When it comes to recalling information, surface learning is not as effective because less information is remembered as efforts are placed on memorisation rather than understanding.
On the other hand, deep learners transformed the knowledge they gained by exploring it beyond the main point. Deep learners aimed to understand the meaning behind the text, and interacted with the material by creating relevant arguments and examples related to their daily lives. Rather than memorisation, the learners engaged and thought critically about the information. They showed great interest and were calm in their learning of the information. As deep learners are more engaged, information is more likely to be retained long-term.
Table 1: Examples and Features of both Surface-based and Deep learning approaches.
Deep Learning | Surface Learning |
Relates topic and ideas to past knowledge and experiences | Unreflective approach; facts not elaborated on. |
Thinks critically about newly learned material | No interaction with content or ideas |
Ties in information from other sources | Concentrates only on memorisation |
Creates new arguments and understands logic based on new information | Underlying argument not comprehended |
Recognises a structure in the content | Treats the task as like a monotonous chore |
Motivation from within, wants to learn | External incentive, based on demands of a test |
Aims to understand the meaning behind the material | Aims to recite and regurgitate material inactively |
Conceptions of Learning
Subsequent research by Marton & Säljö produced six conceptions of learning that university students experience during their studying period (Marton & Säljö, 1997). The six conceptions of learning are structured in a developmental hierarchy starting from the lowest: as a quantitative increase in knowledge; as memorisation; as the acquisition of facts for later use; as the abstraction of meaning; as a process designed to understand reality; and lastly for “developing as a person”. The diagram shows how the conceptions are related to the amount of knowledge obtained, as well as their relation with deep and surface approaches to learning.
What Influences Depth of Learning?
There are two types of motivation that relate to both approaches to learning. If a student wants to learn, and there is no fear regarding learning because they are not under threat – then the student exhibits intrinsic motivation which is associated with a deep approach to learning. Intrinsic motivation occurs when the student is interested in learning and mastering a topic, rather than simply remembering in order to achieve high grades.
On the other hand, extrinsic motivation does not come from within the student, rather induced by circumstances pushing them to revise– for instance grades or money. Extrinsic motivation includes those who are uninterested in the topic, but anxious about trying to learn it, so they try and show the desired behaviour to extinguish a threat (E.g. threat of an upcoming test). Those who are fearful and disengaged with the material are shallow learners; whereas those who are relaxed and interested in the material adopt a deep approach (Marton, Hounsell & Entwistle, 1984).
Motivation put to the Test
Fransson (1977) studied this concept by examining a set of students who were 1st years in an Education department, thus interested in reading a text about “The Examination System in the Education Department”. Another group of students were analysed who were not in the education department of their institution, therefore presumably uninterested in the text. The groups were divided into subgroups. One group were told they had to complete a video-recorded oral report regarding the material – thus creating a source of fear and an extrinsic motivation to learn. The other group were told that they would not have to present information. Findings showed that intrinsic motivation was the best: absence of anxiety and absence of threat both independently and together were linked with a deep approach (Fransson, 1977). Extrinsic motivation was correspondingly associated with a surface approach.
Is Motivation always correlated with learning approach?
Recently, Sengodan & Iksan (2012) studied 78 students whilst learning maths, and found that certain components of intrinsic motivation such as effort and self-efficacy were significantly linked with learning styles, yet worry was not (Sengodan & Iksan, 2012). This suggests that perhaps students employ different kinds of intrinsic and extrinsic motivations which may affect learning to different degrees; or that motivation effects on learning may differ depending on the type of subject being learned (e.g. Mathematics and Reading are two very different activities). Also intrinsic motivation can differ depending on individual student characteristics (Ryan & Deci, 2000).
Learner Intentions
The main difference between deep and surface learning approaches lies with how learners go about learning and why they do so – this is partially shown through Marton and Säljö’s Six Conceptions of Learning described above. Hence, naturally, learner intentions are a massive determinant of whether learning is deep or surface (Marton & Säljö, 1976). Intentions of Deep Learning are generally to understand, to engage with and operate in a subject while evaluating it. This might be reminiscent to the type of learning that takes place during an assignment that requires the review of a research area. Whereas, the Intentions of Surface Learning are merely preoccupied with memorisation, without interest or engagement in the subject matter and with the only motivation being to acquire a specific grade or qualification (Lublin, 2003). For example, the type of learning that might be adopted during a hectic exam period.
Demands of Assessment
As explained above, if put under the pressures of situations requiring a vast amount of information to be learned at one time, for example – exams, it is likely that learners will adopt surface strategies with merely the goal of passing such an exam in mind. However, research suggests that the nature of assessment in itself can influence the type of learning that takes place – for example if an assessment system simply rewards surface or rote learning (Entwistle, 2000). The view conveyed is that surface learning will simply be more adaptive to a student if the assessment system does not emphasise and reward personal understanding but rather tolerates the simple regurgitation of memorised facts (Entwistle, 2000). To enhance deep learning, Biggs (1999) argues for constructive alignment where learning objectives emphasise the importance of personal understanding and assessment that mirrors these aims.
Teacher Influence
Part of the constructive alignment that Biggs (1999) calls for is the designing of a learning environment that will support and complement deep learning objectives. This therefore lies in the hands of teachers and teaching approaches have been found to influence the learning methods adopted by learners as well as their outcomes (Entwistle, 2000). Marton and Booth (1997) argue that it is the responsibility of a teacher to produce “meetings of awareness” between their own knowledge and the knowledge of their students to ensure that what they are teaching can be manipulated in such a way that will ensure that students understand it. Hence, producing deeper learning through the mapping of new knowledge on to pre-existing.
Conceptions of Teaching
Trigwell and Taylor (1994), additionally argue that similarly to student conceptions of learning as described by Marton and Säljö (1997). On analysis of interviews conducted by various researchers, such as Trigwell & Taylor (1994) and Kember (1998), where teachers were asked to elaborate on what they meant by teaching and learning as well as their views on assessment three categories of teacher conceptions were produced (which actually closely parallel those of Marton & Säljö). Two main categories were obtained and are described below.
Teacher Centred/Content Orientated was the first group produced. From this group, teacher intentions were covering the course syllabus and ensuring that students receive the appropriate information. Teachers saw assessment as chance for students to demonstrate the detailed factual knowledge that they had acquired and saw the outcomes of these assessments as entirely the learner’s responsibility, as produced from their personal motivation and ability. (Surface Learning)
Student Centred/Learner Orientated, the second group determined where teachers aimed to guide students toward personal understanding and more sophisticated notions of what they were being taught. These teachers relied on varied assessment systems, incorporating different methods and realised their own responsibility of ensuring the development of a deep understanding in students. (Deep Learning)
Student Activity Focused was the third group produced that can be viewed somewhere in the middle of the two above. Teachers in this group set assignments that ensured active learning (meaningful learning that relates to personal experience) and the development of efficient study skills. However, similarly to the Teacher Centred/Content Orientated teachers, they maintained to see learning merely from one perspective – i.e. – as the responsibility of the learner only.
It is clear to see how the conceptions held by a teacher might affect the teaching methods adopted, and in turn the learning style engaged in by their students. Therefore, in line with the constructive alignment called for by Biggs, the conceptions that teachers portray to learners should be, alongside learning objectives, those that emphasise the importance of deep learning.
Issue up for Debate? However, a study by Ramsden, Beswick and Bowden (1986) suggests that when teachers attempt to convey that what they are looking for in terms of learning is that of deep, personal understanding – learners who would generally engage in surface learning merely adopt different surface strategies that will produce outcome mimicking that of deep learning.
The Outcome of Deep & Surface Learning – What’s Best?
It appears that there is a relationship between approach to learning and the learning outcome – those who adopt deep learning achieve higher knowledge and higher grades than shallow learners. The table below illustrates results from one of the original Maton and Säljö (1976) tests in which 30 students were instructed to read a newspaper, then questioned on the content. It shows how well students learned in relation to the level of processing.
Table 2: Relationship between Level of Processing and Level of Outcome
Level of Processing | ||||
Level of Outcome (Where A =high, D = Low) | Surface | Unclear | Deep Level | Totals |
A | - | - | 5 | 5 |
B | 1 | 6 | 4 | 11 |
C | 8 | - | - | 8 |
D | 5 | 1 | - | 6 |
Total | 14 | 7 | 9 | 30 |
Similar results have been produced by numerous researchers in higher education contexts, by Newble & Entwistle (1986) for example. Chin and Brown (2000) have also demonstrated the benefits of deep learning strategies such as generative thinking; analysing ideas and relating them to personal experience, compared to surface approaches in school level chemistry. Further, Trigwell and Prosser (1991) provided parallel evidence of the superior effects of deep learning over a whole course, rather than, as the above studies incorporate, merely on one academic task.
Is Surface Learning Ever Superior?
Despite, robust evidence on the superiority of deep as opposed to surface learning approaches, it has been suggested that there are specific instances in education where surface learning is necessary/efficient. Artherton (2011) highlights higher education subjects where the mere memorisation of basic facts or laws is essential before the whole subject can be mastered – for example, law or medicine. However, the impossible role of surface learning later in the study of these subjects is clear. Similarly, in a school context there are instances where rote learning is not only necessary but is effective and information retained for long periods of time – the learning of timestables; memorising the periodic table or learning the basics of a foreign language (Lublin, 2009). In fact, it might be argued that the acquisition of a second language is based largely on surface learning. Clearly there is a place for surface learning in education and in every day contexts, memorising phone numbers for example but by large, deep learning is superior.
Strategic Learning: A Happy Medium?
Deep, Surface, Strategic! Which to Choose?
A Practical Example of Deep Learning: Active Learning in Schools
Active learning is dominant in schools throughout many countries and is a nice way to view attempts at deep learning. The introduction of the Curriculum for Excellence in Scotland in 2010 pushed active learning to the forefront.
Contrary to what you might think, active learning is not about movement or exercise but is learning that emphasises the importance of meaning to pupils; it seeks to engage them, challenge their thinking and build new knowledge by linking it to what is already known as well to real life contexts – a lot like the features of deep learning approaches!
In Active Learning environments pupils learn by doing, through experience and reflection and teachers challenge, guide and build on previous experience. For example, when teaching a class about money – instead of having them read out of a textbook the teacher will provide currency to be used in practical exercises. Understanding is achieved through direct experience as opposed to passive learning – listening to a teacher/lecturer speak.
And Does it Work?
Yes, is the simple answer. Alike studies directly investigating deep learning, research on active learning suggests that it generally enhances and produces superior learning. In fact, statistically analysing 253 active learning studies, Hattie claims that active learners perform approximately a grade and a half better than controls. The Learning Cone developed by Edgar Dale – below – nicely summarises the features and outcomes of active/deep learning as well as the importance of these approaches for retention.
So you have looked at the evidence of the benefits of adopting a deep learning approach and you now want to become a deep learner – the good news is you can! Researchers argue that an individual is not simply a deep or a surface learner, that these traits are not stable (Biggs, 1999) and can be used by any one person at different times. So based on the literature, how can you become a deep learner?
- Organise and structure your content into a coherent whole (Ramsden, 1988).
- Have determination to achieve, understand and engage in academic work.
- Attempt to develop some intrinsic curiosity for the subject; think about how it might of personal interest to you or how the theory might apply to every day experience.
- Put thought into your assessment answers and bring ideas from different areas together.
- Attempt to develop a little background knowledge of what you are attempting to learn (Biggs, 1999; Entwistle, 1988; Ramsden, 1992).
- Link new knowledge, concepts and principles on to that background knowledge! Make connections; relate ideas and look for patterns (Pask, 1976; 1988).
- Critically analyse new ideas (Marton & Säljö, 1976).
- Monitor the development of your own understanding; reflect and evaluate your work (Entwistle, McCure & Walker, 2000).
Challenge ideas and yourself!
Memorisation can be Deep Learning?
There has been debate as to whether “deep” learning is actually better for understanding. According to the Deep and Surface Learning model, routine memorisation is undesirable and generally leads to a lack of understanding of the material. However, in studies examining high-achieving Chinese students; it seemed that memorisation led to a good level of understanding of material, normally associated with deep learning (Marton et al, 1996). This was termed “The Chinese Paradox” as the findings contrasted the model, because memorisation is normally linked with shallow learning. However, similar findings of good understanding via memorisation have also been found in Nepalese students (Dhalin & Regmi, 1997), which questions the underlying principles of the deep and surface model. It has been suggested that there may be different definitions of memorisation which could explain the inconsistency (Marton et al, 1996). The model may be culture-specific, and may not represent all students in higher education.
Can We Properly Measure Student Learning?
As student learning, understanding and study habits are vague concepts to measure, the literature has been criticised with regards to methodology. For instance, some believe that interview questions are set up in such a way that responses provoke socially desired answers or agreement to questions; which may not reflect the truth (Haggis, 2003). Furthermore, uncertainties have been raised regarding the psychometric validity of learning surveys including Bigg’s Learning/Study Process Questionnaire, and the Entwistle’s Approaches to Studying Inventory (Richardson, 2000). Arguably the self-report instruments may not measure students’ true conceptions, and that their responses reflect the learner's impressions rather than the learner's behaviour.
What is “Meaning” to the Individual?
Perhaps a difficulty with the concept of a “deep” learning approach is the focus on meaning, in that those who abstract more meaning learn more and thus have adopted a deep approach. However, meaning is very non-specific, and could have a number of interpretations. For example, meaning can be interpreted as finding the right links between knowledge within the subject area (McLean, 2001). This is specifically defined by the discipline of the subject, and the lecturer who is teaching that subject, and may not apply to students working within different subjects. On the other hand, meaning could refer to an individual’s personal implications, in that the personal meaning for the student may be linked with other aspects of life unconnected to studying. For instance studying may only be a fraction of “meaningful” activities undertaken by a student.
One interesting way to look at deep and surface learning theory is the degree it overlaps or corresponds to other models of learning. Draper (2005) suggests that, although addressing separate ideas about learning, Laurillard’s (1993) Conversational Framework of Learning and Perry’s (1968; 1970) proposed stages of learning might overlap with deep and shallow learning theory in practice.
This is most easy to see when considering, as Draper does, the aspect of Perry’s model concerned with analysing theories and concepts that might not be entirely true or false; learning to evaluate evidence – similar to the ‘critical thinking’ that has been associated with deep learning approaches. And this goes further, when considering theory that might not be true, Perry is about evaluating the evidence, as stated but additionally about considering the relationships/patterns that said theory has with other theories/ideas as well as with personal experience – again, reminiscent of the deep learning strategies that have been suggested throughout the literature.
As Draper states, overlap of Perry’s model with Laurillard’s – preoccupied with how meaning looks in the world; personal and public experience, is limited to the component of personal experience of Perry’s, but this does not really pose a problem for the overlap with deep and shallow learning theory. It is clear to see how deep learning strategies relate to the emphasis of personal experience conveyed in Laurillard’s model, especially when looking more closely at the 12 activities of learning proposed by Laurillard. Activities can be grouped so that 1 to 4 correspond to conceptual descriptions of information; 5 to 8 to the personal meaning and action taken on that information by the student and 9 to 12 to reflection and modification of the links made between information and personal experience. As well as personal experience, the interaction that goes on between the teacher and student is a dominant aspect of Laurillard’s model – feedback and reflection are vital – this is extremely alike certain deep learning strategies.
Moving away from models, what other learning strategies might apply to deep and surface learning? Discussing the idea of teacher-student interaction above provokes the importance of discussion to become emphasised. Hence it is probable that teaching methods using such activities as Collaborative Learning will influence deep learning – where each member of a group becomes a ‘master’ in one area of a topic and then presents this information to the remainder of the group. This idea of presenting and teaching others can be seen as the optimum of Active Learning as described in Edgar Dales’ Cone of Learning in the Practical Section above.
Biggs, J. (1999). What the student does: teaching for enhanced learning. Higher Education Research & Development, 18(1), 57-75.
Chin, C., & Brown, D. E. (2000). Learning in science: A comparison of deep and surface approaches. Journal of Research in Science Teaching, 37(2), 109-138.
Dhalin, B. & Regmi, M. P. (1997) Conceptions of learning among Nepalese students, Higher Education, 33, 471–493.
Entwistle, N. (1981). Styles of Learning and Teaching; an integrated outline of educational psychology for students, teachers and lecturers. Chichester: John Wiley.
Entwistle, N. (1988). Motivational factors in students' approaches to learning.
Entwistle, N. (2000, November). Promoting deep learning through teaching and assessment: conceptual frameworks and educational contexts. In TLRP Conference, Leicester [citado abril 30, 2008]. Disponible en: http://www. etl. tla. ed. ac. uk/docs/entwistle2000. pdf.
Fransson, A. (1977). On Qualitative Differences in Learning: IV – Effects of Intrinsic Motivation and Extrinsic Test Anxiety on Process and Outcome. British Journal of Educational Psychology, 47, 244-257.
Garcia, T., & Pintrich, P. R. (1994). Regulating motivation and cognition in the classroom: The role of self-schemas and self-regulatory strategies.
Haggis, T. (2003) Constructing Images of Ourselves? A Critical Investigation into ‘Approaches to Learning’ Research in Higher Education. British Educational Research Journal, 29, 1, 89-104.
Kember, D., & Leung, D. Y. (1998). The dimensionality of approaches to learning: An investigation with confirmatory factor analysis on the structure of the SPQ and LPQ. British Journal of Educational Psychology, 68(3), 395-407.
Lublin, J. (2003). Deep, Surface & Strategic Approaches to Learning. Centre for Teaching and Learning. Good Practice in Teaching and Learning.
Marton, F., & Booth, S. (1997). Learning and awareness. Lawrence Erlbaum.
Marton, E., Dall’alba, G. & Kun, Tse Lai (1996) Memorising and understanding: the key to the paradox? In: Watkins, D. & Biggs, J. (Eds) The Chinese Learner (Hong Kong, Comparative Education Research Centre/Australian Council for Educational Research
Marton, Hounsell and Entwistle (1984) The Experience of Learning: Implications for Teaching and Studying in Higher Education. Accessed Online: http://www.tla.ed.ac.uk/resources/EoL.html
Marton, F. & Säljö, R. (1976). On Qualitative Differences on Learning: I – Outcome and Process. British Journal of Educational Psychology, 46, 4-11.
Marton, F. & Säljö, R. ([1984] 1997) Approaches to learning, in: Marton, F., Hounsell, D. & Entwistle, N. D. (Eds) The Experience of Learning (Edinburgh, Scottish Academic Press).
McLean, M. (2001) Can we relate conceptions of learning to student academic achievement? Teaching in Higher Education, 6, 399–413.
Newble, D. I., & Entwistle, N. J. (2009). Learning styles and approaches: implications for medical education. Medical education, 20(3), 162-175.
Pask, G. (1976). Styles and strategies of learning. British journal of educational psychology, 46(2), 128-148.
Pask, G. (1988). Learning strategies, teaching strategies, and conceptual or learning style. Learning strategies and learning styles, 1(988), 83-99.
Prosser, M., Trigwell, K., & Taylor, P. (1994). A phenomenographic study of academics' conceptions of science learning and teaching. Learning and instruction, 4(3), 217-231.
Ramsden, P. (1987) Improving teaching and learning in higher education: the case for a relational perspective. Studies in Higher Education, 12, 274–286.
Ramsden, P. (1988). Improving learning: New perspectives. Nichols Pub Co.
Ramsden, P., Beswick, D., & Bowden, J. (1986). Effects of learning skills interventions on first year university students’ learning. Human Learning, 5(3), 151-164.
Richardson, J. (2000) Researching Student Learning (Buckingham, Open University Press).
Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary educational psychology, 25 (1), 54-67.
Sengodan, V. & Iksan, Z. H. (2012). Students’ Learning Styles and Intrinsic Motivation in Mathematics. Asian Social Science, 8, 17-23.
Trigwell, K., & Prosser, M. (1991). Improving the quality of student learning: the influence of learning context and student approaches to learning on learning outcomes. Higher education, 22(3), 251-266.
Vermunt, J. D. (1998). The regulation of constructive learning processes. British journal of educational psychology, 68(2), 149-171.