Assessment and Reporting in Visual Art

(BACKING OUR CREATIVITY SYMPOSIUM, Melbourne, 13-14 September, 2005)

© Donald Richardson, 2005


The assessment (marking/grading) of students' art has long been a matter of professional concern. How can one put a finite, arithmetical value on something as ingenuous as a child's freely-expressed painting? Is it ethical to do so, even?
With the current national emphasis on reporting assessments, it is important that art teachers and lecturers adopt the right methods in the best interests of their students.

This paper examines the following issues:
* whether it is ethical to judge and mark/assess students' art at all
* the use of ranking, grading, marking etc
* different kinds of report
* the structure of knowledge of the subject

It recommends that products of an art-learning experience be not assessed solely - or even mainly - as finished products, and/or according to the teacher's or lecturer's personal taste, but objectively as indicators of each individual student's learning in and about the subject.

It also examines the relative merits of different systems of reporting student learning, attitude and achievement in visual


'There is no disputing taste', the old saying goes - but, actually, we dispute each others' taste in, and judgments about, art all the time. While this may be no more than an amusing game in the lives of adults most of the time, it is a very serious matter in education because student learning depends upon consistent and positive feedback. If art teachers and lecturers evaluate students' art products purely according to their own personal tastes - and award marks or grades accordingly - their assessments are likely to be inconsistent with those of other teachers or lecturers and the resulting impression of institutional adventitiousness is likely to inhibit students' confidence in the subject. This is a most undesirable outcome at the school level and it may ruin the very lives of tertiary students.

The assessment of student art has always been a matter of great professional concern at every level. Should it be rated in relation to established, professional artists? Should it be graded in relation to class peers? What criteria should be used in assessing the art of elementary and primary children? Is it ethical to assess and mark young children's art at all? What allowance - if any - should be made for 'talent'?

(You probably all know Schulz's cartoon in which Lucy is given a 'C' for her coat-hanger sculpture.)

The general confusion has not been ameliorated by recent discussion about student reporting emanating from the Commonwealth Department of Education and Training. In fact, it has brought about an urgency to come to terms with the problem once and for all.

This paper is based on three premises. The first is that what should be assessed in art classes is not the aesthetic value of the works produced but the amount of student learning the works exhibit. Schools and universities are not art galleries, but learning institutions. Students (the unwilling notwithstanding) are in school to learn - in Art as in Science, Mathematics,Literature etc - not just to have aesthetic experiences (although we hope the experience of art will be a rewarding one). The second is that educational assessment is a diagnostic tool which indicates to both teacher and student the path future learning should take. Thus, feedback to the student (i.e., reporting) should be framed in such a way as not to discourage further engagement in learning. The third is that it is the individual student who learns and whose progress is mapped by assessment.

The teaching-learning principle this paper supports as appropriate to learning in and about art is similar to the concept of 'personal best' performance in athletics. The individual student is assessed in relation to his/her previous 'performance' and not in relation to class averages or other norms. It requires that students are not competitively ranked against each other and that reporting avoids labelling any student 'failure' until the time comes for a final grading - i.e., at the end of a schooling period, be it a term, a year or the final year of schooling. Of course, this depends upon the teacher/lecturer having - and the students understanding - a programme that clearly defines learning increment.

Given the foregoing, there can be no ethical objection to assessing student learning in art at any level; in fact, the student - and all students in the class - will know as well as the teacher/lecturer what learning has occurred.

School is not a horse-race, designed to seek out the one winner and 'the devil take the hindmost'. Contrary to what some contemporary politicians seem to think, schools are concerned to facilitate the learning of all students, not just the 'winners'. 'Nothing succeeds like success', but nothing discourages like failure either.

When teachers/lecturers assess, they are not acting as art critics or competition judges but as educators. This means that what they should be assessing and reporting on is student learning in and about art. This makes assessment more equitable for the students (because untalented students have as much chance of achieving good grades as the talented ones do - more so, if they learn more; we all know students with a talent in a limited field, like cartooning, who rest on this and absolutely resist learning anything else) and also less of a hassle for teachers as well (because they can put aside their personal tastes and prejudices).

The distinction between assessment and reporting
As elementary as it may seem, this is a distinction worth making because these are two very different functions with different purposes, but they are often conflated in practice.

We can enter this discussion by noting that those teachers who claim they 'do not assess' art are not being entirely honest with themselves. It is not possible for any conscious human (or any organism, really) not to be assessing and evaluating his/her environment all the time. If you ask these teachers for an opinion on any of the students that are 'not' being assessed, he/she will usually offer an opinion (i.e., will report an assessment). So, what is actually meant by the claim not to assess students' work is that the teacher is just not reporting the assessment that he/she has made - for whatever reason.

Different kinds of report
But, we must recognise that teachers are continually reporting their assessment to students as part of the normal teaching-learning process. When a teacher comments to a student 'that's good' or 'why don't you try more blue in the background?' he/she is reporting to the student the result of an assessment - in a normal, natural way that is designed to help the learning process.

This informal, personal reporting is educationally sound and the most humane kind of report; however, it is not always used to the best advantage. A report,if it is to be of any use to its recipient, should contain appraisals that can be explained and justified and also give an indication of how improvement can be made. If all the students in a class are told 'that's OK' or 'that's interesting' - that is, if they all get the same report, and a report that gives very little assistance on how to improve or to learn more - they will lose all confidence in the teacher's ability to help them learn. Students are aware that they are not identical with each other, that their works are not identical, and that art is about individuality and creativeness.

But what schools normally regard as 'reports' are the formal cards or documents given to students at the end of an assessment period. Whereas it is clear that the informal, personal report is intended for the student and to assist his/her learning, it is not absolutely clear for whom the school report is intended and what purpose it is intended to serve. Is it for the student (and, therefore, not to be divulged to his/her classmates, parents, prospective employers, etc)? Or is it for the parents (and not to be divulged to prospective employers, for example)? Or for the student and the parents (and no others)? Or?

Students, their parents, the teachers of the students, prospective employers, and the school administration itself all need reports on the progress of students, but it is hard to see how one report can fulfill the specific needs of all of them. The kind of report students need is described above. Parents, on the other hand, need to know (or ought to) how their child is 'getting on', how he/she compares with others in the class, whether the home environment can assist learning better, whether remedial work is needed at home, and whether the student is "good", cooperative,etc. Teachers need to know what action they should take to improve their teaching technique. The school administration needs to know whether the student should be promoted or directed into another course etc. Institutions or other schools a student might move to need to know about his/her past performance. And prospective employers need to know what propensities a student may have that could apply to particular jobs.

Norm-referenced and criterion-referenced assessment
Most assessment of students since schools began has been what we now know as norm-referenced assessment. This means that the group being assessed is evaluated, graded or marked against a standard or norm,which is usually the average mark of all students in the group. The norm is the 'pass mark' i.e., 50% or 5/10, or C in a five-letter scale.

In norm-referenced assessment, approximately half the students score above the norm and approximately half score below it (unless 'pass mark' is set at 40%, 4/10 or D, as is sometimes the case). But, wherever 'pass' is set, under norm-referenced assessment a significant number of students must fail. When we consider that about half of the students in any cohort will go right through their schooling never achieving a pass mark in any subject, we must recognise that the use of this method can only be justified educationally at the end of a specified learning period. This is its justification in Year 12 examinations.

And, to relate this more specifically to Art:
* how do you establish a norm for a Year 11 class when six of them have not done art since year 8, three of them are clearly of tertiary-quality and the rest average year 11s who have taken the subject in years 9 and 10?
* how do you assess the one or two exceptionally talented students in relation to the norm? Is it right that they should always get high marks, even if they don't try or if they learn little? And,what does this do to the rest of the students?

But norm-referencing is not the only way. In criterion-referenced assessment the teacher sets a task and students and teachers alike know the criterion by which achievement will be assessed. The advantage of this concept is that the criterion can be set at a level to suit the abilities, age, standard or previous experience of each individual student. Comparison with peers does not come into it at all, whereas it is all-important in norm-referencing. This means that every student, whether 'talented' or not, can achieve an A or B reasonably often and, thus, receive the positive reinforcement necessary for successful and enjoyable learning.

And, if the criteria refer to learning in and about art, it also means that the talented student who coasts along on his/her drawing ability(for example) will get no more As than the average student because the learning of each is emphasised at a level appropriate for each. To give an example: if the objective of a particular lesson is to learn the skills of making linocuts and the 'talented' drawer of horses trots out his/her usual clever drawing in an unprintable form, he/she fails the criterion (and gets a D), whereas an untalented student who uses a less impressive motif, but produces a printable block, gets an A.

Another way of avoiding educationally unsound personal comparisons is to make the criterion a measure of how much better a student is at a particular task now than he/she was at the beginning of the assessment period. This is sometimes called self-referenced assessment. For example, in a term spent on landscape painting, the student's first effort is compared with his/her last and, if the increment of learning shown in the works themselves and revealed through discussion with the student is sufficient, any student may receive an A or B.

In criterion-referenced assessment, the student still has grades or marks but it will not be valid to use them to make comparisons of different students, and this may make criterion-referenced assessment of little use to school administrations, except when they need normative data to decide which group a student should be promoted into etc. So, norm-referenced data must be provided for this purpose. But the two should never be confused and students and their parents must know the difference and it must be public knowledge whether an assessment has been norm- or criterion-referenced. It should, for instance, be indicated on all reports. If it is not,everyone assumes the grades have been norm-referenced, because that is how it has been ever since.

It is also essential that students know which system is being used because, otherwise, they might receive encouraging results all the term and a low assessment at the end because the former was criterion-referenced and the latter norm-referenced. This sort of thing often happens in art assessment because a teacher has pushed up the marks during the term to encourage the student only to find that norm-referencing must occur finally. There will be no problem with this if the students, the teacher and the parents know what kind of assessment is occurring at each stage.

Quartiles and A-E Scale
Recent calls from the federal Minister for Education and Training for the national adoption of reporting student assessments in both quartiles and A-B-C-D-E grading requires comment. Quartiles (as the name suggests) is a four-stage ranking of students in one or other of the 0-10%, 11%-50%, 51%-75% and 76%-100% ranges of marks. We will not discuss here how students find themselves in any of these ranges; however, it is imperative that we recognise that, using quartiles (as with any use of percentages), it is inevitably that approaching 50% of students are likely to be 'failures' - and the educational implications of this have already been discussed under norm-referenced assessment. In A-E rankings, on the other hand, a large percentage of students in a cohort (perhaps the largest) will be ranked C, with a smaller percentage falling into the A, B, D and E categories. Only those in the latter two could regard themselves in any sense as 'failures', so this system has greater educational relevance than the quartile system.

But, insisting on both being used simultaneously (on the grounds that this would be 'plain English' reporting) is not only illogical and confusing but educationally unjustifiable because the A-E scale is a five-part division (into quintiles) whereas quartiles is a four-part division. This means (as the diagram shows) that about half of students ranked C will join the Ds and Es as 'fails' because their marks are below 50%, which is educationally unjustifiable.
grades: / A / B / C / D / E /
quartiles: / 1st / 2nd / 3rd / 4th /

Assessment of learning and assessment of attitude
Art teachers have often tried to ameliorate the unhappy effects of norm-referenced assessment regimes by including an assessment of student attitude in a report. Sometimes this is done by adding an unidentified loading for 'trying' to the assessment. This is not a sound practice because, unless the two components of the assessment are identified,the report will give a distorted picture of both the amount of learning and the amount of effort put into the activity.

If reports are to be given at all they should at least contain meaningful information to learners (and their parents) on how they can improve their learning.

But an even less justifiable practice is to identify a separate assessment for attitude or behaviour which is then combined statistically with the assessment of learning. As the following table shows, three quite different students can get the same grade from this process. The final (average) C, which would be all that would appear on each of the three students' reports, gives no useful information to anyone.

Assessment of Learning Attitude, Behaviour, etc Total(Average)
Student 1 A E C
Student 2 E A C
Student 3 C C C

This is not to say that an assessment or comment on student attitude should not be given - it should, because positive reinforcement can be provided in this way. But it should not be conflated with the assessment of learning.

Types of Learning
Art is an omnibus subject, involving several different concepts and activities and, therefore, different types of learning. The 'structure of knowledge' of the subject may be categorised as
* Learning as an artist or designer. This is the creative aspect of the subject, usually involving making works - functional works (works of design) or non-functional, purely aesthetic, works (works of art)
* Learning as a craftsperson. This involves the skills necessary to practice as either artist or designer. It is a useful aspect of the subject at the school level for students who, for whatever reason, are not 'creative' because they benefit from practising the skills for their own sakes.
* Learning as a consumer or critic. This is the study of works produced by others - usually famous works - and need not involve any practice at all. 'Art Appreciation' classes come into this category, all those who 'appreciate' art being consumers of art. Criticism here does not mean professional art journalism but the ability to talk and write coherently about works. This is an important aspect of student learning in art because few will become practitioners, but all are potentially part of the public for art and design.

It is logical to assess each of these aspects separately, but - for practical purposes - it may be appropriate to combine them into the one grade or mark.

The specification of criteria in advance and sound teaching-learning practice go hand-in-hand. Specifying criteria for learning must bean integral part of programming, if teaching is aimed at anything more than 'an experience'. How do you assess a 'paint anything you like' lesson? If students are set this task, how can any get less than A grades if they do just that? In such lessons, in practice, teachers usually assess the work on some private criteria of their own - often,even, the criterion is that of adult art, or some other measure of quality of which the student is not appraised.

Teaching and learning should be as goal-directed in art as in any other subject, and assessment-of learning according to known criteria is not inimical to the subject if those criteria genuinely reflect its nature.