Assessment for learning
By Margaret Adolphus
What is wrong with assessment?
When I was at university, some 30 years ago, summative assessment – on which one received one's ultimate degree grading – was by examination. Some 25 years later, when I was myself teaching a university class, I failed to appreciate (and no one bothered to explain) that assessment had changed, and that students' coursework counted for a percentage of the overall mark.
The discovery that I had not assembled the marks for the latter (nor indeed created all of them in the first place) in preparation for the students' end of semester grading was unfortunately made on the day I'd had to absent myself due to a serious family crisis.
Few experiences of assessing students' work are quite that stressful, but most would agree that marking is not the reason they came into teaching. If students got something out of it, then it would all be worthwhile. However, there is serious concern in some quarters as to whether assessment really does develop students' learning. Moreover, a survey of student satisfaction conducted in 2006 by the UK's National Union of Students showed that, while generally students were positive about their experiences, 49 per cent were dissatisfied with assessment and feedback (Shepherd, 2006).
There is general agreement that assessment should be linked with the learning outcomes of the relevant course unit. This process is described in "Learning outcomes and assessment criteria". However, there is also a concern that there is too great a reliance on measurability.
Assessment is a powerful driver of learning, with many students using it as a lens to view their current course. However, it may be that concern for grades could be to the detriment of learning in a wider sense.
In their review of assessment in universities, Elton and Johnson (2002: p. 7) comment on how the former is often held back by traditionalist practices which have not been properly reviewed and reflected on. They contrast this with the more radical views of teaching and learning which shift the onus from the teacher transmitting towards the learner constructing knowledge:
"In demonstrating the overwhelmingly positivist basis of current approaches to assessment, which is in striking contrast not only to philosophical developments of the past half century, but even to current approaches to teaching and learning, we call into question the philosophical basis of virtually all current approaches to assessment" (Elton and Johnson, 2002: p. 94).
It is entirely right that assessment should be related to the course learning objectives: if these are carefully worded to encapsulate that which matters the most, then that which matters the most will be assessed.
However, assessment becomes distorted if it is conceived purely in terms of that which can be measured. According to Elton and Johnson (2002: p. 10), there are certain qualities – "quality of mind", "independent critical thinking", "breadth" – which distinguish a successful graduate.
Similar views are expressed by Price et al. (2008):
"There are important benefits of higher education which are not amenable either to the precise specification of standards or to objective assessment."
Assessment should derive from the practices of academic and professional communities, and the nature of the tasks within which learners engage, as well as just learning objectives and marking criteria.
At the University of Illinois at Urbana-Champaign, all academic units are required to assess student achievement in their undergraduate and graduate programmes. In its assessment, the Department of History brought back recent graduates in order to discover the effects of their training on their current careers. The results of this process were fed back into one of their introductory programmes: "Introduction to historical scholarship".
In other words, assessment has a broader purpose than merely to pass judgement on a student's performance: it should communicate the standard expected so that this can be internalized. Then learning can become not just a race to the finishing post, but a personal commitment to those values of judgement, critical analysis and independence of thought which should characterize academic life.
The search for authentic assessment
Challenges to traditional assessment have been particularly strong in the UK, which has seen accusations of lowering standards and dumbing down of higher education. Recently, a 551-page dossier concerning standards was submitted to a committee of Members of Parliament, which included allegations from Alan Ryan, warden of New College Oxford, that any reasonably well-organized student could get a 2:1 (Innovation, Universities, Science & Skills Committee, 2009).
One of the submissions from this document was from the Assessment Standards Knowledge Exchange (ASKe), a centre of excellence for teaching and learning based at Oxford Brookes. ASKe called for a "root and branch" change in assessment processes and practices, and argued that there were fundamental flaws in the current system, at both macro (degree) and micro levels (the assessment of individual students).
A number of other projects have sprung up to look at assessment. Re-engineering Assessment Practices in Scottish Higher Education (REAP) was put together with the specific objective of redesigning large first-year courses around more formative assessment and feedback: see http://www.reap.ac.uk. It was funded from 2005-2007 and comprised three partners, the universities of Strathclyde, Glasgow and Glasgow Caledonian, which transformed 19 modules across a range of disciplines. Results have shown not only improved learning, but also staff-efficiency gains.
The ESCAPE Project (Effecting Sustainable Change in Assessment Practice and Experience), is funded from 2009-2010, and is based at the University of Hertfordshire. Drawing on expertise from the University of Hertfordshire's Blended Learning Unit, it looks at ways in which information and communications technology can help assessment.
All these initiatives are looking at ways of supporting learning in a more authentic manner. Authentic assessment should promote active and deep learning. Deep learning seeks to understand and relate to other knowledge and experience, as opposed to shallow learning which concentrates on memorizing and verbatim regurgitation.
The various types of learning have been analysed into taxonomies: academic education seeks to achieve those at higher levels.
The Teaching and Educational Development Institute (TEDI) at the University of Queensland quote Bloom's taxonomies:
- Knowledge (recall and recognition of facts).
- Comprehension (restate, review, identify etc.).
- Application (apply, interpret, solve, illustrate).
- Analysis (analyse, compare, contrast, distinguish).
- Synthesis (compose, design, synthesize).
- Evaluation (assess, judge, predict).
As well as Biggs' taxonomy of learning outcomes, Structure of the Observed Learning Outcome (SOLO):
- Prestructural (the task has not been understood).
- Unistructural (only one aspect of the task tackled).
- Multistructural (several aspects of the task are tackled, but are treated independently).
- Relational (the topic is understood adequately, and the different aspects treated as a coherent whole).
- Extended abstract (not only is the topic and its components understood, but the learner is able to move to a level of abstraction and apply to a new topic).
Source: see Isaacs (2001).
Formative versus summative assessment
As we have seen, one of the main criticisms of assessment is that it is too concerned with grading. Here it is important to distinguish between formative and summative assessment. Summative assessment is graded and often comes at the end of the module; formative assessment is informal and mainly intended to help the student learn.
There is general agreement that the balance between formative and summative assessment in higher education is also biased towards the latter. Rust (2001) points out the value of giving formative feedback early on in the course, so that both teacher and student can pick up on areas which need improvement:
"It is arguable that assessment in British higher education is too often focused on the summative, and the accumulation of marks, coming at the end of courses, while students would benefit from more opportunities to build on their strengths and learn from their mistakes through the feedback from formative assessment activities staged throughout their course or module" (Rust 2001: p. 6).
Maclellan, who in her research on perceptions of assessment (2002) found that there was an inconsistency between teachers' declared support of formative assessment and their practice, makes a similar point, maintaining failure to do this will preclude the opportunity to modify teaching.
Black and William (1998) carried out a meta-analysis of research over a ten-year period, in which they asked three questions about formative assessment:
- does it raise standards,
- is there room for improvement, and
- if so, how?
They concluded that the answer to the first question was yes, and that this was due to:
- enhanced feedback,
- the involvement of learners,
- the ability to adjust teaching,
- improved motivation and self-esteem.
Formative assessment can help both the teacher (who can adjust the teaching to accommodate learner needs) and the learner (by indicating where his or her performance falls short).
Feedback is a major way in which assessment can help students learn. Good feedback should adhere to certain principles:
It must be constructive, pointing out weaknesses in a positive way. It should indicate the successful features of the work and those that are less so; how the work could be improved; how the student might do better in the future.
At many US colleges and universities, students attend first-year writing courses. Here they learn, through detailed feedback from both tutors and fellow students, about the importance of revising drafts – and also about the centrality of feedback to learning (Chickering and Gamson, 1987).
It should be timely. Students in the National Union of Students survey often complained that they received feedback too long after submitting their essay, too late to inform other similar assignments.
No feedback can occur without assessment. But assessment without timely feedback contributes little to learning (Chickering and Gamson, 1987).
3. Given at the right stage
It should come at the right stages in the learning process. Formative assessment and feedback is useful at early points in the course to develop students' abilities. Studies on retention have shown that students are particularly vulnerable in their first term, so early "low stake" feedback is less threatening than getting a grade at the end of the first term, and can be a useful monitor of progress. Isaacs (2001) recommends submitting large pieces of assessable work in stages, for example the plan, the results, the final report with analysis and conclusion.
Students should be encouraged to engage with their feedback. For example, Isaacs (2001) suggests that students are paired up when work is returned in order to discuss each other's work; and also that they should write what they learned on an index card.
5. Doesn't have to be resource intensive
Feedback does not always need to be resource intensive: it can be automated. For example, computer-assisted assessment uses objective tests in which students choose from a series of predetermined questions and receive feedback.
The University of Strathclyde was able to reduce staff workload and increase student involvement by having regular online testing in a number of subjects. For example, in French, online tasks saved 200 hours of staff time while reducing the exam failure rate from 24 per cent to 4.6 per cent compared with the previous year (see http://www.reap.ac.uk/assessment/pilotsSU.html).
Effective feedback should not limit itself to improved performance: it should also help the student become more conscious of his or her learning processes, and more self-critical. This is known as self-regulated learning, and was one of the goals of the REAP project.
Part of the research rationale for REAP was the work of Nicol and McFarlane-Dick (2006), which claimed that students already possess the ability to assess their work and generate feedback. They advocated opportunities for self- and peer-assessment – students assessing their own and one another's work.
Rust (2001) believes that self- and peer-assessment helps students to apply the assessment criteria for themselves, and thereby gain insight into what is required. Students can be given the opportunity to reflect on their work, and analyse their own strengths and weaknesses, perhaps through completing a self-assessment sheet.
Peer-assessment can take the form of students commenting on one another's work before handing it in to the tutor, or commenting on whole class presentations.
At Liverpool John Moores University, a Level 2 module, "Debating gender in the media", requires students to facilitate a seminar. Students organized a workshop based on themes and readings for that week, and 10 per cent of the total mark went on peer-assessment.
Very often, standards and learning goals are "givens" in higher education, developed by faculty alone. However, many believe that self-regulated learning is helped if there can be a dialogue between tutor and taught about standards, which helps the latter internalize what is required. This was one of the tenets of the ASKe group:
Assessment standards are socially constructed so there must be a greater emphasis on assessment and feedback processes that actively engage both staff and students in dialogue about standards. It is when learners share an understanding of academic and professional standards in an atmosphere of mutual trust that learning works best (Price et al., 2008).
The REAP project actively involved students in the business of identifying and formulating criteria as well as in decisions about assessment.
Use assessment methods that are fit for purpose
More emphasis on formative assessment should not detract from the importance of summative assessment, which is still an important marker of progress, and which should also carry feedback.
The main assessment methods are described in "Learning outcomes and assessment criteria". The method used should fit the purpose: there is a tendency to be too reliant on the essay, which calls for the ability to critically evaluate and formulate argument, but which may not be appropriate in circumstances which require personal reflection.
The Assessment Strategies and Standards in Sociology project reported on a major growth in interest in autobiography in sociology and anthropology, as a way of seeing how theory relates to everyday life (Mears and Harrison, 2000). Thus assessments which incorporate autobiography help to generate understanding of sociological theory, and also provide a way in to the academic world for mature students, who have a considerable amount of life experience on which to reflect.
Autobiographical assessment methods include learning diaries or portfolios, as well as essays with an element of personal refection. The former are becoming popular in a range of courses as the value of metacognitive strategies, the ability to reflect on learning, is realized. Courses which require students to build up a portfolio often call for some reflection on the learning process.
- In the theory and method in qualitative research module for the University of East London Level 2 Sociology undergraduate course, students were required to do a "learning response paper" which involved a series of 500-word reports on the weekly sessions. The reports described the session's main content, evaluated lessons learnt from the workshop exercise, and assessed the topic within the context of qualitative research. The exercise contributed 40 per cent to the overall mark.
- At the same university, for the lived histories module for the Level 1 History course, students had to write an essay entitled "Autobiography as tool", 1,000 words "that explore in some way where you are in history" (Mears and Harrison, 2000).
Portfolios are often used in business and management education as students collect practical data and tools for analysis relating to a particular case study. As such, they can be a good way of transferring learning from theory to the actual practicalities of the business world.
At York St John University, students on a business management degree had the opportunity to use their portfolios when researching an entrepreneurial opportunity, as a means of collating and organizing information to simulate a business plan. The portfolios were assessed via a test in order to generate a grade.
The whole topic of technological tools for assessment is too large to be considered here, however they provide plenty of opportunities for innovative forms of assessment. The REAP project involved redesigning courses to involve podcasts, blogs, electronic voting systems, online tests, e-portfolios, discussion boards, simulations, intelligent homework systems and feedback software.
At Strathclyde Business School, the first-year management development programme started using blogging in 2007 as a way of enhancing student reflection on an assessed assignment, "The holiday project". It was found to be a useful way for students to reflect on their learning and on the team process.
A whole institution approach
Assessment is a vastly complex subject, involving the whole institution and not just the individual lecturer. Joughin and Macdonald (2004) recognize this and propose a four-tier model:
- Level 1: the actual experience of assessment at the modular level, and how the teacher interprets the requirements of the module and translates them into learning tasks.
- Level 2: the immediate context of the module (the programme) and the teacher (the department). Influencing factors include whether the department supports good practice, innovation, and staff development.
- Level 3: the institution. What are its policies, how does it allocate resources, what is its teaching and learning strategy? Is good teaching rewarded? What about the attitude to ICT?
- Level 4: the overall context within which the institution operates, including government bodies and funding.
Major changes, such as the introduction of new technologies, inevitably need to be managed at an institutional level, as they will require considerable investment not only of financial resources but also of staff training. The REAP website contains a number of useful papers on how to manage institutional change.
This article has looked at ways in which the assessment process can better support learning, by introducing more formative assessment, providing feedback, helping students regulate their own learning, and using varied and imaginative methods. Much of the impetus behind the drive for better assessment has come from the UK's concern about standards, hence many of the examples are British. However, the lessons are universal: by paying attention to the process as well as the end result, assessment can really be for, as much as of, learning.
Black, P. and Wiliam, D. (1998), "Inside the black box: raising standards through classroom assessment", Phi Delta Kappan, Vol. 80, No. 2, pp. 139-148, available at: http://www.pdkintl.org/kappan/kbla9810.htm [accessed April 1 2009].
Chickering, A.W. and Gamson, Z.F. (1987), "Seven principles for good practice in undergraduate education", AAHE Bulletin, Vol. 39 No. 7, pp. 3-7, March, available at: http://www.aahea.org/bulletins/articles/sevenprinciples1987.htm [accessed April 1 2009].
Elton, L. and Johnson, B. (2002), Assessment in Universities: A Critical Review of Research, Learning and Teaching Support Network Generic Centre, available at: http://eprints.soton.ac.uk/59244/ [accessed April 1 2009].
Innovation, Universities, Science & Skills Committee (2009), Students and Universities, House of Commons Publications, London, available at http://www.publications.parliament.uk/pa/cm200809/cmselect/cmdius/memo/… [accessed April 1 2009].
Isaacs, G. (2001), Assessment for Learning, Teaching and Educational Development Unit, The University of Queensland, available at http://www.tedi.uq.edu.au/downloads/Assessment_for_Learning.pdf [accessed March 26 2009].
Joughin, G. and Macdonald, R. (2004), A Model of Assessment in Higher Education Institutions, The Higher Education Academy, York, available at http://www.heacademy.ac.uk/assets/York/documents/resources/resourcedata… [accessed March 26 2009].
Maclellan, E. (2002), Perceptions of Assessment: An Audit of Practise, The Higher Education Academy, York, available at http://www.heacademy.ac.uk/assets/York/documents/resources/resourcedata… [accessed March 31 2009].
Mears, R. and Harrison, E. (Eds) (2000), Assessment Strategies and Standards in Sociology: A Resource Handbook, available at http://www.c-sap.bham.ac.uk/subject_areas/sociology/fdtl_assessment.htm [accessed March 26 2009].
Nicol, D.J. and Macfarlane-Dick, D. (2006), "Formative assessment and self-regulated learning: a model and seven principles of good feedback practice", Studies in Higher Education, Vol. 31 No. 2 pp. 199-218.
Price, M., O'Donovan, B., Rust, C. and Carroll, J. (2008), "Assessment standards: a manifesto for change", Brookes eJournal for Learning and Teaching, Vol. 2 No. 3.
Rust, C. (2001), Basic Assessment Issues and Terminology, The Higher Education Academy, York, available at http://www.heacademy.ac.uk/assets/York/documents/resources/resourcedata… [accessed March 26 2009].
Shepherd, J. (2006), "Courses deliver, but feedback falls short", Times Higher Education, August 26.