Since it was introduced in the 1800s, standardised testing in Australian schools has attracted controversy and divided opinion. In this series, we examine its pros and cons, including appropriate uses for standardised tests and which students are disadvantaged by them.
Educational assessment provides evidence of what students have learned and are able to do. Standardised educational assessments, such as Australia’s NAPLAN, are often used to judge and compare student achievement. They assess all students on the same content under standardised conditions.
The assumption is that this is fair, providing a level playing field for all students. Frequently, the focus of standardised administration becomes maintaining a test’s “integrity”.
What is “fair” for students with disability?
A major challenge in assessment is how to obtain evidence about learning from students with disability. We cannot know how these students are faring with their learning if assessments or tests are structured in ways that create barriers for them. We would not expect a student who is blind, for example, to complete a paper and pencil test.
Fairness can be thought of in two ways. It can mean procedures that treat everyone the same, or it can mean treating individuals according to their needs, to ensure a fair outcome for everyone.
For students with disability, the second way means removing assessment barriers that prevent them from achieving their best results. For a student who is blind, the obvious solution is a braille test, if they are proficient in braille, or a person or technological aid to read the test aloud, and record their response. This is known as an assessment adjustment (in Australia) or accommodation (in the US).
Reasonable adjustment protocol
Australian and international anti-discrimination law requires that reasonable adjustments are provided for students with disability to enable equal access to assessments as other students.
The law also protects the “integrity” of an assessment or certification so that a student with disability is not reported as achieving something that they have not. An underlying assumption of adjustments is that the effect of the disability and assessment barriers are “neutralised”, and the students participate in the assessment on the same basis as students without disability.
Standardised test adjustments often provide common options. For example, these might be the use of a scribe, reading questions (unless a test of reading) or instructions aloud, a support person, assistive technology, braille form, extra time, and rest breaks.
A major concern that policymakers, teachers, and some students have regarding assessment adjustments is that they should not provide unfair advantage for students with disability. Extra time is one of the most contentious areas. Guidelines are usually stringent.
For example, the NAPLAN guide for students who need testing in braille suggests an extra ten minutes per half hour for a writing test, an extra 15 minutes per half hour for a reading test, and an extra 20 minutes per half hour for numeracy, with more time if needed. For other disabilities, the protocol guidelines state:
it is recommended that no more than five minutes of extra time per half hour of test time be granted; however, in some cases, up to an additional 15 minutes per half hour of published test time may be provided.
Three issues with adjustments for students with disability
The first important issue is the goal of test adjustments. Do test adjustments, given their restrictions, only need to enable students to get a pass? Are they able to do their best?
The second, equally important, issue is the lack of empirical research evidence regarding the appropriateness of common test adjustments. For example, one US study identified optimal time extensions for all students with disability, on average, as one and a half to two times the standard test time. For students with visual or hearing impairments, two to three times the standard time is necessary. This creates another disadvantage for students with disability if the test has to be completed in one sitting. Other researchers have shown that simplification of standardised tests can improve results for students with disability, without losing validity or reliability.
The third issue is that disability discrimination can take many forms. It also occurs when no adjustments are available and students with disability are not able to participate in assessments. However, students with disability want to be treated the same as other students, even to the extent that some students do not want adjustments.
For example, in international tests such as PISA or TIMSS, students with intellectual or functional disabilities in mainstream schools who “would be very difficult or resource intensive to test” are excluded. In the last round of TIMSS in 2015, 2.1% of Australian students in mainstream schools were excluded. However, an estimated 18% of Australian students receive some form of adjustment in their education. Potentially, 16% of students who complete these tests may need test adjustments. What adjustments were provided, and how this impacted on their performance, and Australia’s overall performance are not known.
Three ways standardised testing can be improved for students with disability
First, research is urgently needed on the impact of test adjustment restrictions on how well students with disability are able to do.
Second, the extent to which conditions such as time restrictions affect all students, not just students with disability, needs to be reconsidered. As US researchers have noted, time and speed of response are not usually identified as components of what is being tested. The best solution is to eliminate the role of speed in testing. When more time is available, all students do better, but students with disability improve more.
Third, alternative forms of assessment should be provided, as allowed under US legislation, rather than tinkering with standardised tests.
Assessment should provide information on what a student has learned and is able to do. It should not focus on the integrity of a test to the extent that limits equitable participation by students with disability. Often, the outcome of standardised tests is reinforcement for students with disability that there are things they can’t do that students without disability can. Everyone should be given the opportunity to show what they know regardless of disability.