TOWARDS EQUITY IN ASSESSMENT: ACCOMMODATING A VARIETY OF LEARNING STYLES

LEE SUTHERLAND
UNIVERSITY OF ZULULAND

PREAMBLE

The literature on assessment reports internationally on trends in education which we in South Africa have come to regard as unique to our situation. Gipps (1995) talks of differential performances in the context of both ethnic groups and gender. Gibbs (1992) talks of rising student-staff ratios (p 149). Many authors and researchers talk, of the inadequacies of their own assessment systems. There is in fact little research being done in any country in the world that reports on students who come to a tertiary system from a unitary, homogenous school system. There is currently a school of thinking that goes so far as to say that all classrooms are multilingual in the sense that the linguistic ability of any two students is never the same.

Much of the criticism levelled at assessment systems, is the perceived failure to render an equitable assessment system. It is not surprising that Perkins (1992) says ‘test bashing has become one of the most popular sports in the educational arena’ (p 179). The need for a shift in educational assessment practices in the United States is well documented (see for example Viator, 1994: 39). Gipps (1994: 15) citing Wood (1987) makes the point that assessment has been seen at various times both to promote equal opportunities and to militate against them.

Perhaps our multilingual, multicultural South African situation is not unique after all its search for an equitable assessment system. For this reason much of what I have to say, has already been said in other contexts and in other arenas However, my input will be to draw together the various strands that are developing in assessment, and to weave these into a clothe that is appropriate for the South African context. This is however, not a facile attempt to provide discreet solutions to a very complex problem.

Although gender issues (which deal with biological and psychological factors resulting in differential performance) within the broader equity debate can not be ignored, it is not within the scope of this paper to deal with all equity issues. This paper will attempt to confine itself to issues of environmental disadvantage or ‘differential learning experience’ (Gipps and Murphy, 1994: 264). Three important questions raised by the research of Darling-Hammond and Ancess (1994, cited in Garcia, 1994: 70) seem pertinent to our debates on equity in this regard:

How can ‘historically discouraged learners’ be assisted to achieve while maintaining standards of excellence?

What support structures are needed for these students to achieve?

How can teachers revise their teaching to help these students?

Any debates on assessment need to acknowledge current trends, both in South Africa and internationally. The recent publication of a Draft Proposal for National Guidelines for Assessment is important for informing the debate on local trends.

CURRENT DEVELOPMENTS IN ASSESSMENT IN SOUTH AFRICA

The Draft Proposal for National Guidelines for Assessment focuses on assessment in compulsory schooling, whilst contextualising this within education and training as a whole This document outlines the shift in assessment paradigm that has become essential in the South African context. It also outlines why such a paradigm shift is necessary : ‘A paradigm shift or ‘scientific revolution’ occurs when an old paradigm is unable to deal with an outstanding problem’ (Gipps 1994 quoted in Draft Proposal for National Guidelines for Assessment, 1996: 2). It goes on to say that the old assessment paradigm is inadequate to deal with the challenges presented by new policies aimed at transforming education and training. In changing the paradigm, the old paradigm informs, and becomes part of, that new paradigm.

The following aims characterise the underlying rationale for a paradigm shift : 

The implementation of these guidelines, many of which have a social-cultural perspective, is characterised by the following and may in future promote equity :

WHAT IS EQUITY?

Gipps and Murphy (1994) debate what is meant by equity in education. In this debate, the whole is worth more than the sum of the parts thereof: equity means more than equality of input, equality of outcomes, or the two combined. It is a complex issue that includes factors like equality of access to the curriculum, equal epistemological access, support structures and standards. As Gipps and Murphy point out, it also means more than equality, which refers to the absence of differences (ibid).

In such a debate, the important questions might be: are group differences ‘real’ or are they the result of the measuring system? Does creating equity mean eliminating bias and prejudice? In terms of the challenge for equity in assessment, Gipps says baldly that ‘there is not such thing as a fair test, nor could there be: the situation is too complex and the notion too simplistic’ (1995: 279). Gipps and others (see also Murphy, 1990 and Goldstein, 1993) addressed the fallacy that multiple choice tests and psychometrics and technology will eliminate bias. ‘The traditional psychometric approach to testing operates on the assumption that technical solutions can be found to solve questions of equity with the emphasis on using elaborate techniques to eliminate test biased items’ (1995: 273 - 274, 1994: 149). Perhaps the greatest limitation of this approach is that it does not question what counts as achievement (ibid). In addition, viewing bias as a circumstance where a test is unfair to one particular group or another belies the complexity of the underlying situation. Differential performance on a test is more likely to be the result of environmental factors like differing access to learning (ibid). The notion of a standard test as a way of offering impartial assessment remains a powerful one; however, if equality of opportunity does not precede the test, then the ‘fairness’ of the approach is called into question (Gipps and Murphy 1994: 15). Similarly, decontextualised assessment is not possible or desirable (ibid: 276).

The equity debate often becomes linked to debates about standards- how do we achieve equity in education while improving the quality of education? Fairclough (1992) reminds us that inequalities in education are often disguised as ‘diversify’ Gipps works hard at dispelling the common myth that ‘alternative’ equals ‘non-standard’ equals ‘sub-standard’ (1995: 276). Wiggins (1994) suggests that we need to reconsider our notions of standards: ‘To see test scores as the key indicator of educational well-being is an unthinking response to the problem of standards... standards refer to qualities, not quantities’ (Wiggins, 1994: 5).

The issue of standards is also addressed in the Draft Proposal for National Guidelines for Assessment. It is forthright in stating that ‘issues around standards are very politically loaded. It is important to make distinction between improving [or maintaining] quality and preserving certain existing standards in education. As Adrienne Bird (1994) has pointed out, the phase ‘maintaining standards’ has often been abused in a context where the speaker implies that former white education left nothing to be desired and that other cultures have nothing to offer. For this reason it is important to unpack exactly what is meant by ‘standards’.’ (p 16 - 17).

Coupled to this is the notion of achievement: equity debates need to challenge the notion of achievement and what constitutes achievement. Gipps and Murphy (1994) remind us that we need to recognise that views of what constitutes achievement. How it is defined and revealed will reflect the value judgements of powerful groups in society. ‘We need to encourage the questioning of this world view [that sees white middle-class males as ‘normal’ and all other groups as ‘other’] to challenge traditional attitudes and to value other group perspectives, achievements and concerns’ (p 277). If achievement is defined by groups in power alone, a large number of individuals will appear deficient (ibid, 265). Perhaps one way of challenging the dominant power groups is by an awareness of language and the way in which it positions people. Such study is known as Critical Language Awareness (CLA).

CLA AND ITS IMPLICATIONS FOR EQUITY CHALLENGES

It is not within the scope of this paper to provide a full discussion of the CLA movement (for a full comprehensive reading of CLA refer to Fairclough, 1992); it must suffice to say that CLA is not a branch of language study but rather an orientation towards language: ‘It highlights how language conventions and language practices are invested with power relations and ideological processes which people are often unaware of’ (Fairclough, 1992: 7). CLA has proved itself central to language teaching in a climate of change: ‘... language itself is becoming a target for change, and change in language practices in coming to be seen as significant in the implementation of more general social and cultural change’ (ibid: 5). Its focus in equity debates must lie in its being a prerequisite for democratic citizenship, as emphasised by Fairclough: ‘. . given that power relations work increasingly at an implicit level through language, and given that language practices are increasingly targets for intervention and control, a critical awareness of language is a prerequisite for effective citizenship, and democratic entitlement’ (ibid: 12)

Its role in empowering (or dis-empowering) students is explained by Janks: ‘CLA can make people conscious of how language can be patronising, demeaning, disrespectful, offensive, exclusive or the opposite’ (in Fairclough 1992: 312). While CLA refers directly to language teaching and learning, it has crucial implications for educators It sees the use of language was a component of the meaning-making process. McKenzie is insistent on the intimate relationship between language-teaching and a learner-centred pedagogy (in Fairclough 1992).

However, Janks tells us there are no textbook recipes or short cuts as to students can be empowered through language (in Fairclough 1992). She shows how language can, in fact, contribute to a process of subjection (ibid). Some academics hold the belief that if students wish to enter the academy, they need to accept the norms, practices and standards of the academy without challenging them. On the contrary, CLA shows how educators need to help learners question attitudes and gain access to alternative perspectives (ibid: 313) CLA can operate in an emancipators way, only if it is successful in contesting the practices which disempower people (ibid). Assessment situations are quite obviously an area where there is the potential in this regard.

MOTIVE OR METHOD?

Much of the literature on assessment focuses on the how rather than the why. Bintz and Harste (1994) and Kohn (1994) challenge researchers who fall into the former group. Both believe that looking at the methods of assessment will achieve little, while examining the motives and underlying assumptions about assessment will allow assessors to value alternative assumptions about learning. Bintz and Harste say this: ‘developing new forms of assessment without changing old assumptions about learning and curriculum is tantamount to using old eyes to see new possibilities. Alternative assessment requires that we value alternative assumptions’ (1994: 11). Instead of asking what new forms of assessment are worth developing? we should be asking what new assumptions about learning, assessment and curriculum are worth valuing? (ibid) Kohn takes this debate one step further when he says: ‘The easy question is whether a student has learned something; the far more important - and unsettling - question is whether the student has been given something worth learning’ (1994: 40). Implicit in this statement is the need to re-examine the nature of learning.

THE NATURE OF LEARNING

This paper argues that a socio-cultural perspective of learning and assessment is one way in which to address issues of equity. Constructivism is defined by Murphy as ‘a theory of knowledge which involves conceptions of the learner, of knowledge and of the relationship between them’ (1994: 251). Socio-cultural theories go further than this, arguing that individuals are socially, politically, economically, historically and institutionally constructed (Wertsch 1991 cited in Murphy 1994: 254 - 255). Common to both these perspectives is the notion that knowledge is constructed: students are actively engaged in thinking and the ideas and experiences that students bring to situations matter (ibid). Allowing students opportunities to construct meaning is one way of providing equal epistemological access to learning.

‘Valid interpretation of assessment outcomes depends crucially on the assessor understanding the significant characteristics of learners and tasks and how these interact.’ (Murphy, 1994: 250). Further support for the move towards re-examining the nature of learning is the work of Gibbs (1992) who points to research in Europe and Australasia that has identifies students’ approach to study as a key variable in predicting student performance and the quality of learning outcomes: ‘Whether students take a surface or a deep approach (attempting to reproduce subject matter or understand it) has a profound effect on the quality, structure and permanence of students’ learning’ (p 149). One aspect of the research that has particular relevance for the South African context, is the conclusion that Gibbs comes to that the way in which many courses respond to increased student numbers, seems very likely to foster a surface approach in students (1992: 154). In addition, he points out that students in large course often express very conservative views about innovation in teaching and great anxiety if alternatives to lecturing are suggested (ibid, 155).

The issue of construct validity (traditionally defined as the extent to Waco a test measures what it claims to measure) relates to the integration of assessment and teaching and is a fundamental issue with regard to assessment and equity (Gipps and Murphy, 1994: 23) Construct validity relates to appropriateness and ‘fitness for purpose’ (ibid, 260). The converse of this notion would be that unfairness in assessment suggests invalidity (ibid, 25). Viator (1994) relates the issues of assessment, and more specifically construct validity, to issues of accountability: an approach which acknowledges accountability requires that the relationship between instruction and assessment be integrative (p 41). Perkins puts this issue in perspective when he says: ‘There is nothing wrong at all with teaching to a test provided the test tests the outcomes you really want’ (1992: 172).

DIFFERENTIAL PERFORMANCE AND UNDERACHIEVEMENT

A leading question in the debate on equity is given by Gipps (1995: 237): ‘Are the forms, content and mode of assessment appropriate for difference groups and individuals?’ Viewing assessment in a socio-cultural perspective, the assessor has to acknowledge the variability created by students’ different experiences: when students try to make sense of assessment practices they do so using the totality of their understanding of the world. These and other sources of variability, like for example, theories of multiple intelligences (to be discussed later in the paper) have implications for the generalisability and comparability of assessment. Explorations of construct validity must include a focus on the ways in which individuals and groups deal with assessment tasks.

The three research questions posed initially are partially answered by Scarr (1994, in Gipps and Murphy 1994: 16) ) who concludes that in order to give disadvantaged students equal opportunity, educators have to give them more and instruction. The Draft Proposal for National Guidelines on Assessment points out that an increasing number of educationist (see for example, Popham, 1987, Crooks, 1988 and Madaus, 1988) have shown the great positive potential of testing as a way of intervening to raise the quality of education (p 8). Murphy (1995: 268) says that ‘assessment used formatively and equitably can reveal and begin to break this cycle of underachievement for students’.

The Draft Proposal for National Guidelines for Assessment clearly recommends that all assessment systems should be characterised by ‘The opportunity to demonstrate competence in a variety of ways across a variety of contexts’ (p 44). Multi-modal assessment, or multidimensionality, which accommodates a greater number of learning styles is but one way of addressing the need for equity.

MULTIMODAL ASSESSMENT

The rationale for suggesting multi-modal assessment practices is far reaching and perhaps the strongest is the notion of multiple intelligences, although not new, most recently acclaimed by Gardner (1983, 1993). Sternberg (1994) uses his own theory as well as that of Gardner to advance a very strong case for multi-modal assessment. Having been struck by the fact that not many of the techniques of instruction and evaluation teachers work for everyone, he realises that a single approach to instruction and evaluation will tend to favour certain patterns of abilities over others. He therefore concludes that if we wish to reach the most students possible, we must diversify instruction and assessment (p 47). Proving that he does not exist entirely in Utopia he realises that one cannot and should not always teach to students’ strengths ‘Students must learn to remediate and compensate for weaknesses, too’ (Sternberg, 1994: 51) Gipps and Murphy (1994) also talk about the indicators of successful learning and quote Linn (1992: 44) as saying that ‘multiple indicators are essential so that those who are disadvantaged on one assessment have an opportunity to offer alternative evidence of their expertise’ (ibid, 275).

The move towards performance assessment might be one way of tackling the issue of differential performance and underachievement through multi-modal assessment, in that it emphasises best rather than typical performance, based on assessment tasks which emulate the kind of process-based higher-order tasks thought to represent good practice (Shepard, 1991 cited in Gipps and Murphy, 1994: 262).

PERFORMANCE ASSESSMENT: POSSIBILITIES AND PROBLEMS

Authentic assessment, the paradigm with which performance assessment most strongly aligns itself, supports fairness and equity, in that it identifies strengths, minimises comparisons and allows room for student learning styles, aptitudes and interests (Wiggins 1989b cited in Gipps 1994,156). Performance assessment, more specifically, attempts to capture the process that students utilise to accomplish tasks, rather than just assessing the students’ end products (Garcia, 1994: 64) and it attempts to represent or simulate performance in real-world settings (ibid, 68). Performance assessment calls for the use of a range of tasks covering different modes of presentation and response.

Gipps asks: ‘Is performance assessment a good form of assessment?’ (1995: 276) While there is much to be said in favour of performance assessment, the debate on its use must consider the various criticisms that have been levelled at the form of assessment. What started off as a form of assessment used almost exclusively for formative functions of assessment, there have recently been attempts to broaden this application to include summative assessment functions. Some critics feel that it is not sufficiently external, rigorous or accountable to be used for ‘high-stakes’ evaluation. Because of the importance of the role of communication in performance assessment, it might pose a threat for minority language groups (Gipps, 1995: 278 check reference). While tasks that fit into performance-based models attempt to represent genuine learning activities, Murphy (1995) points out the lack of comparability across students’ marks and limited generalisibility of performance across tasks (p 250) However, she also makes the point that ‘if we take seriously the evidence of the ‘situated’ nature of students’ learning, we need to reconsider approaches to comparability and generalisability in assessment practice (ibid, 267).

Other possibility limitations of performance assessment models are that tasks that are assessed require qualitative judgements (Gipps, 1995: 275). In addition, Gipps points out that performance assessment has no underlying score distribution as is possible with more traditional forms of assessment (ibid: 276). However, not withstanding these criticisms, she points out that performance assessment can be beneficial in changing curriculum focus and broaden teaching (ibid, p 276).

A DIVERSITY MODEL OF EDUCATION: SOME RECOMMENDATIONS

‘A diversity model of education starts by understanding, acknowledging, and legitimating the system of’ values, beliefs, and meanings different students hold about the social world This system reflects what students value and why they value it, and is central to creating curriculum in the classroom that is culturally relevant and personally meaningful’ (Bintz and Harste, 1994: 10). The socio-cultural perspective is strongly evident in this view.

Below are some ways in which a diversity model of education and assessment might be achieved:

‘...assessment results should be one part of a system of multiple indicators of the quality of education’ (Gipps 1994: 149)

Recognise the tentative nature of judgements made about students’ achievements (Murphy p 267).

Use assessment formatively to break the cycle of underachievement (Murphy 1995: 268).

Value alternative assumptions about learning (Bintz and Harste, 1994: 11 )

Test development enterprise must include individuals from diverse backgrounds (Viator, 1994: 41)

Avoid a deficient model that sees middle-class males as ‘normal’ (Gipps and Murphy, 1994: 277).

Offer a large range of different ways of answering question which do not privilege one form of answer over another (Gipps, 1995: 279)

Expand the range of indicators of success (ibid: 280)

Review the construct validity of assessments to make sure that they reflect that which we want to measure and that which we taught (ibid 280).

Clearer articulation of test or examination developers construct on which the assessment is based (ibid: 279)

Constructs and assessment criteria made available (ibid: 279) to all stakeholders

Encourage a wide range of modes and task styles.

CONCLUSIONS

Assessment development is a fruitless endeavour unless coupled with broader curriculum reform. Perkins (1992: 173) says that ‘installing a test that demands complex cognition without revamping instruction is a recipe for disaster.’

Teachers need to develop a reflective practice in order to meet the challenges for equity. As Garcia points it: ‘authentic assessment practices will not benefit students from diverse backgrounds unless teachers and educational personnel recognise how their own cultural values and upbringing influence their judgement and evaluation of student behaviour and performance’ (1994: 68).

While Gipps following remark might appear somewhat pessimistic, it has a distinct note of realism: ‘It is unrealistic to imagine a time when differential values and power positions will be removed, they will no doubt change but difference in values and barriers to equality will continue’ (1994: 151). However, Scarr (1994, cited in Gipps and Murphy 1994: 15) points out that abandoning tests altogether however, cannot make society more pluralistic or ensure equal rights, social or economic benefits The challenge is to provide an equitable approach where the concerns, contexts and approaches of one group do not dominate (Gipps, 1994: 156).

 REFERENCES

BINTZ, P AND HARSTE, J C (1994):
Where are we going with Alternative Assessment? And is it really worth our time? Contemporary Education. Vol. 66 No 1.

FAIRCLOUGH, N (1992):
Critical language Awareness. Longman. London.

GARCIA, G (1994):
Equity challenges in authentically assessing students from diverse backgrounds. The Educational Forum. Vol. 59

GIBBS, G (1992):
Improving the quality of student learning through course design in Barnett, R (Ed) Learning to Effect. Oxford University Press. Buckingham.

GIPPS, C (1994):
Beyond testing: towards a theory of educational assessment. Falmer. London.

GIPPS, C AND MURPHY, P (1994):
A fair test? Assessment, achievement and equity. Oxford University Press. Buckingham.

GIPPS, C (1995):
What do we mean by equity in relation to assessment? Assessment in Education. Vol. 2 No 3.

JANKS, H AND IVANIC, R (1992):
Critical Language Awareness and emancipators discourse. in Fairclough, N (1992): Critical Language Awareness. Longman. Essex.

KOHN, A (1994):
Grading: the issue is not how but why. Educational Leadership. October 1994.

McKENZIE, M (1992): ‘What I’ve known but never been told’: euphemisms, school discourse and empowerment in Fairclough, N (1992): Critical Language Awareness. Longman. London.

MURPHY, P (1995):
Sources of inequity: understanding students’ responses to assessment. Assessment in Education. Vol. 2 No 3.

PERKINS, D (1992):
Smart schools: from training memories to educating minds. Macmillan. Ontario.

STERNBERG, R (1994):
Diversifying instruction and assessment. The Educational Forum. Vol. 59.

VIATOR, K (1994):
Involving community in promoting equity in school accountability assessment systems. The Educational Forum. Vol. 59.

WIGGINS, G (1994):
Reconsidering standards and assessment. The Educational Forum. Vol. 59.

WB01339_.gif (1535 bytes)   Back to Top