skip to main content

Research Developments from ACER


Image © Grill

Getting to the essence of assessment

Assessments in education can be made, interpreted and used in different ways, but all serve the same fundamental purpose, as Geoff Masters explains.

A lifetime working in the field has convinced me that assessment in education has become over-conceptualised and over-complicated. Assessment concepts and terminology introduced over the past half century sometimes now function as impediments to clear thinking and good practice; and, worse, the field itself is a mess.

A large part of the problem originates in the belief that there are multiple ‘purposes’ of assessment in education. These different purposes are sometimes described in terms of dichotomies, such as formative or summative, norm-referenced or standards-referenced, school-based or external, and assessment of learning or assessment for learning. Such dichotomies fragment the field and spawn unending concepts and terminology, and unhelpful complication.

In reality, there is only one fundamental purpose of assessment in education: to establish and understand where learners are in an aspect of their learning at the time of assessment. This usually means establishing what they know, understand and can do. When this single purpose is recognised and taken as the starting point for thinking about assessment, it becomes a unifying rather than fragmenting influence in the field.

Understanding assessment in this way helps us to see that assessments can be undertaken at varying degrees of diagnostic detail, using a variety of observation methods, and that the results of an assessment process can be used and interpreted in different ways.

Assessments can reveal varying degrees of diagnostic detail

Because educational assessments are designed to provide information about learners’ knowledge, skills and understandings at differing levels of detail, ‘diagnosis’ is not so much a matter of kind as it is of degree. Assessment instruments differ in their diagnostic power in much the same way that microscopes and telescopes differ in the level of detail that they are able to reveal.

An assessment process might be designed to establish a student’s overall level of proficiency in a school subject, but at a more detailed level it might be designed to establish the student’s levels of proficiency in a number of different areas of learning within the subject or, at a still finer level, their mastery of specific skills and concepts, or errors and misconceptions.

The point is that the fundamental purpose remains the same: to establish and understand where learners are in an aspect of their learning at the time of assessment. The degrees of ‘diagnostic’ detail might vary from a ‘telescopic’ international achievement survey to establish how well students in a country are performing in various areas of the school curriculum to a ‘microscopic’ conversation with a student to establish how they arrived at a particular answer.

Assessments can be based on a variety of observation methods

Assessment tasks provide observations for drawing conclusions (or inferences) about the points that students have reached in their learning. Individual assessment tasks are designed to provide observations that can be used to infer what cannot be observed – a student’s underlying reading ability, for example. Achieving the fundamental purpose of assessment requires that we go beyond recording how students perform on specific tasks to inferring from task-specific observations where students are in their learning. 

Assessment results can be used in different ways

Although there is one fundamental purpose of assessment in education, there are multiple ways of using the results of an assessment process. Sometimes, our intended use determines the degree of diagnostic detail we require.

One important use of assessment information is to inform future decision making. The results of some assessment processes are used by governments and policy makers to inform action to improve student achievement levels. The results of other assessment processes are used by teachers and students to inform and guide next steps in teaching and learning. In both cases, assessment results are used for decision making and improved learning outcomes.

Another important use of assessment information is to determine the progress that students have made over time. The results of some assessment processes are used by governments and policy makers to monitor overall trends in achievement levels and to evaluate progress in achieving more equitable outcomes for disadvantaged students. The results of other assessment processes are used by teachers and students to monitor the progress that individuals make in their learning. In both cases, assessments of learning progress are used to monitor change over time and to evaluate the effectiveness of programs, policies and interventions.

The point is that, when assessments are used to establish and understand where learners are in their learning at the time of assessment, the results can be used either prospectively (or formatively) to plan future action or retrospectively to evaluate past learning progress.

Assessment results can be interpreted in different ways

Once we realise the fundamental purpose of assessment is always the same, it’s easier to see that the results of an assessment process also can be interpreted in different ways. We might, for example, interpret an assessment for a student by reference to the learning domain itself by asking what the student’s result means in terms of their level of knowledge, understanding and skill. This, in turn, requires a deep understanding of the learning domain through which students are progressing. Alternatively, we might interpret a student’s result by reference to the performances of other students of the same age or grade; by reference to year level curriculum expectations or ‘standards’; or by reference to that student’s past performance.

The point is that, although it is common to treat standards-referenced and norm-referenced assessments as different kinds or classes of assessments, standards-referencing and norm-referencing are more appropriately conceptualised as different ways of interpreting the results of an assessment process. Both interpretations can be applied to a wide range of assessment methods, including multiple-choice tests, essays and performance tasks. ■

Further information:
This article draws on Assessment: Getting to the essence by Geoff Masters AO, available at < >


About the author

Professor Geoff Masters AO is the Chief Executive of ACER. 

More [rd] articles by Geoff Masters

View selected works of Geoff Masters

Related articles

Higher Education
University completions and equity | RD

University completions and equity

06 May 2015

University students from disadvantaged groups have a lower completion rate than their more advantaged peers, but most disadvantaged students do complete their degrees, research reveals.

Evaluation, Quality & Standards, Higher Education, Indigenous

Education & Development
Teacher absenteeism in Indonesia | RD

Teacher absenteeism in Indonesia

16 March 2015

A comprehensive new study reveals that teacher absenteeism in Indonesia is declining, and provides evidence for policy makers focused on improving teaching and learning, as Phil McKenzie explains.

Evaluation, Quality & Standards, Survey, School, Education & Development

Higher Education
Unpacking ‘programmatic assessment’ | RD

Unpacking ‘programmatic assessment’

10 October 2019

A programmatic model of assessment can provide a more complete understanding of trainees’ knowledge and skills, but key aspects of the approach are often misunderstood.

Assessment, Higher Education, Vocational & Adult, Featured home, Featured higher education, Featured vocational & adult, Australia, India, United Arab Emirates, United Kingdom