What Is Deep Learning? Part I: Seminal Texts

The association of the word deep with the word learning is pervasive in higher education: it is ubiquitous in the research literature, it crops up in the delivery of teaching courses, in conversations  about learning effectiveness and in institutional consultation and policy documents.  It is a pairing of words that is normally used to denote learning that is somehow ‘good’. Conversely, any learning that is not deep i.e. that might be labelled shallow, superficial or ‘surface’-like in nature must, by implication, be ‘bad’ (or at least inferior).  Given that higher education is when the most advanced, formalised learning qualifications are studied for and awarded, it implicitly follows that higher education students should be engaged with ‘deep learning’.

However, what does deep learning actually mean?  Do those responsible for teaching and supporting student learning in Universities have a conscious and shared understanding of the term; or is our understanding discipline and/or culturally dependent?  Can we define the term at all or is our understanding of it so tangled in a tacit framework constructed of personal experience, beliefs and assumptions that we would flounder if asked to do so?  Even more importantly, is it safe to presume that deep learning is, necessarily, inherently and innately ‘good’ at all?

Reading List

  • J. Biggs (1987) Student Approaches to Learning and Studying. Melbourne: Australian Council for Educational Research
  • N. Entwistle and P. Ramsden (1982) Understanding Student Learning. Beckenham: Croom Helm.
  • F. Marton and R. Säljö (1976) “On Qualitative Differences in Learning: I – Outcome and Process.” British Journal of Educational Psychology. 46. pp4-11.

All texts are available to University of York holders of a Paperpile account.

Meeting Report (3rd  November 2015) 

In our  first meeting we looked at seminal texts from the 1970s – 80s that established a research methodology to investigate student learning and defined “deep” and “surface” approaches to learning, as well as the associated principles of “strategic”/”achieving” approaches to learning (Marton and Säljö. 1976; Biggs, 1987; Entwistle and Ramsden, 1982)

Much of the discussion circulated around the relevance of these terms to the current practice and experience of colleagues, with a general consensus that the principles of deep/surface learning continued to have significant currency in modern Higher Education, but that this currency was often based on unquestioned assumptions about what the terms actually meant. Points were raised, for example, about whether “deep learning” was the same as “critical thinking/learning,” and the discussion also opened out to consider the so-called “Chinese Learner Paradox” (Watkins and Biggs 1996) – whereby international students often demonstrate high cognitive learning (deep learning) despite using strategies that this early research identifies as “surface learning” (e.g. memorization and rote learning).

There was less agreement on the currency of the principles of “strategic/achieving” approaches to  learning outlined in the research. Many in the group felt that there was an implicit pejorative tone underlying the outline of strategic learning in the research, despite its claim that high-achieving students might well also engage in strategic learning. Indeed, it  was felt that developing a strategic approach to learning was actually an important life skill for students. The critical observation was also made concerning the relevance of this research to the current HE context of high student numbers comparative to teaching staff, which means that some of its proposals were often untenable in this environment (such as an emphasis on the importance of sustained one-to-one contact between staff and students, or the general emphasis on individualised approaches to learning and teaching).

The discussion raised further issues about whether it was possible to determine from any given assessment whether a student was engaged in genuine “deep learning” or was merely being successfully strategic in “performing” mastery to satisfy assessment requirements. What does “transformative,” “deep learning” look like, exactly – from a teaching delivery and assessment point of view? Does it describe something innate to individual students or is it, as claimed in the original research by Biggs et al, a variable that can be developed in any individual through reformed approaches to teaching? Can we teach someone to approach learning deeply if they are not already inclined to do so?Does a “deep” approach to learning even yield academic success for students? Given the claim made in the research that our conventional approaches to Higher Education teaching and assessment are at odds with our (assumed) desire to stimulate “deep” learning, what kind of learning design is being proposed through this research to address this issue?  

The lively discussion left us with considerable food for thought, albeit with no clear answers to the questions that the research provoked, which we resolved to carry through to the next meeting. As agreed, we will here consider critical responses to the methodology, conclusions, implications and impact of this research on Higher Education learning and teaching practices in the UK.

Works Cited

  • Biggs (1987) Student Approaches to Learning and Studying. Melbourne: Australian Council for Educational Research
  • Entwistle and P. Ramsden (1982) Understanding Student Learning. Beckenham: Croom Helm.
  • Marton and R. Säljö (1976) “On Qualitative Differences in Learning: I – Outcome and Process.” British Journal of Educational Psychology. 46. pp4-11.
  • D. Watkins D & J. Biggs J (eds) (1996) The Chinese Learner: cultural, psychological, and contextual influences, Hong Kong/Melbourne: Comparative Education Research Centre/Australian Council for Educational Research.

Suggested Reading Questions

The overarching questions for the network meeting might be:

  • How do these elaborations of deep learning and surface learning correlate with their wider usage in discussions on the Higher Educational practices forty years later?
  • A commonly raised point in this research is that contemporary Higher Education learning and teaching design, delivery and assessment discourages deep learning. Do you agree this is (still) the case? If so, how might this issue be addressed? If you disagree that this is the case, why not?
  • What are the implications of this research on learning and teaching in your discipline?

Paper 1: F. Marton and R. Säljö (1976) “On Qualitative Differences in Learning: I – Outcome and Process .” British Journal of Educational Psychology. 46. pp4-11.

Marton and R. Säljö’s are often credited with introducing the concepts of deep/surface learning into Higher Education studies through their analysis of students’ comprehension of passages from academic texts using a phenomenographic methodology. This is a qualitative approach that investigates the different ways people experience things. It proceeds by questioning students on their perceptions of a learning experience and then categorising their responses within an interpretive framework.

Questions to consider when reading this paper might include:

  • What do Marton and Säljö perceive to be the problem with the “traditional method” of describing the outcomes of learning (4)? How do they propose to address this problem with their approach?
  • How do they define deep and surface learning? What model of learning is implicit in this description?
  • Does their definition correlate with your tactic understanding of the terms?

Paper II: N. Entwistle and P. Ramsden (1982) 2Chapter 10: Learning and Teaching in Higher Educaiton” Understanding Student Learning. Beckenham: Croom Helm. pp193-220.

N. Entwistle and P. Ramsden’s study builds on Marton and Säljö’s work using the same methodology to broaden the scope of their enquiry. In addition to the latter’s idea of students adopting a deep or surface approach to learning, they consider ideas of (positive and negative) strategic approaches and describe how learning contexts can steer students towards adapting their approaches to perceived learning expectations (See Fig 10.1 for a tabular summary of these approaches). The additional emphasis on learning contexts draws into the debate issues of broad disciplinary differences between the Arts, Humanities and Social Sciences compared to the Sciences; the impact of lecturers’ attitudes; and (significantly) of assessment practices as motivating factors that can determine which set of approaches students might consciously or unconsciously take.

Questions to consider when reading this paper might include:

  • What do Entwistle and Ramsden mean when they write of “different styles of learning” (197)?
  • How does the learning context affect, either positively or negatively, students’ approaches to learning?
  • What constitutes a “good” approach to learning in their study?
  • What constitutes “good” learning contexts in their study?
  • What recommendations, if any, to the design and delivery of Higher Education programmes are (implicitly or explicitly) suggested in this concluding chapter?
  • What do they mean by the following terms?
    • Strategic orientation versus academic orientation
    • Personal meaning versus evidence in building understanding.
    • Outcome and process o Internal and external student orientations.
    • Syllabus Bound (Sylbs) versus Syllabus Free (Sylfs)
    • Serialist and/versus holist approaches to learning.
    • Learning pathologies (improvidence and globetrotting)

Paper III: J. Biggs (1987) “Chapter 2: The Nature of Student Learning” Student Approaches to Learning and Studying. Melbourne: Australian Council for Educational Research. pp14-27.

The final paper for this session is a synoptic piece from J Biggs, who had been undertaking research along parallel lines to the phomenographers developing his own learning inventory, taxonomy of learning and a set of standard questionnaires to evaluate students’ approaches to study and learning. This chapter summarises his methodology and contextualises it against related theories and approaches. Biggs’ work has been massively influential on Higher Education policy and practice throughout the UK; arguably because his work focused on practical approaches to learning and teaching design and delivery, which was underpinned by the theories of how students learn he outlined in this study.

Questions to consider while reading the chapter might include:

  • What data is generated by Bigg’s Study Behaviour (SBQ) and Learning Process (LPQ) Questionnaires?
  • What model of learning is implicit in this scheme? What impact does this model have on the relationship between teaching and learning? Or between teachers and learners?
  • What constitutes “good” learning and teaching in Biggs’ study?
  • Biggs states the aims of his research include an intention to pursue the “practical implementation of the LBQ and SPQ” (18). What might the implications of this aim be on Higher Education programme delivery and design?

One thought on “What Is Deep Learning? Part I: Seminal Texts

  1. Pingback: Making Life Difficult: Testing to Improve Learning | Scholarship of Teaching and Learning (SoTL) Network

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s