Scores track a student's circumstances as much as their ability

The argument

The central argument for abolishing standardised testing is that the scores do not measure what they are used to measure. Proponents argue that performance on a timed written test depends on a long list of things besides the understanding being assessed: familiarity with the format, having been taught the specific material the test samples, access to coaching and practice papers, a quiet place to study, a stable home, health, sleep, and fluency in the language of the paper. Each of these is distributed by family circumstance, so the resulting ranking reproduces the distribution of advantage and presents it as a measure of merit. Proponents argue that this is what makes the tests harmful rather than merely imperfect. An inaccurate instrument used for description would be a minor problem; these scores are used to allocate places, funding, tracking decisions and judgements about teachers and schools. A student whose score reflects their circumstances is then assigned an educational path on that basis, which tends to confirm the original disadvantage, and a school serving such students is judged to be failing for conditions it did not create. The measurement does not merely record inequality but acts on it. Proponents add that the appearance of objectivity is the most damaging feature. Because the test is the same for everyone and marked without regard to who wrote it, its results carry an authority that subjective judgements do not, and they are therefore harder to challenge. A biased measure which looks impartial is worse than an openly contestable one, proponents argue, because it converts unequal preparation into a number that appears to be about the student. On this view the instrument cannot be fixed by better questions, since the inequality enters through everything surrounding the test rather than through its content.

Premises

Counter-arguments

Defenders argue that the alternatives are more biased, not less. Teacher assessment, interviews, portfolios and references all depend on judgements that vary with a student's manner, confidence, background and the resources of their school, and the historical reason for standardised tests was precisely that such judgements excluded outsiders. They argue that a test which correlates with circumstance is partly measuring real differences in preparation that exist whether or not they are recorded, and that abolishing the measurement does not remove the gap but conceals it. Defenders add that without a common measure there is no way to detect that a school is failing its students, so the group most harmed by abolition would be those whose disadvantage currently shows up in the data.

Rejecting the premises

[Rejecting P1] Every assessment method is affected by circumstance, so this cannot distinguish standardised tests from the alternatives. [Rejecting P2] A correlation with background partly records real differences in preparation that exist whether or not they are measured. [Rejecting P3] Consequential use is an argument about how scores are used rather than a reason to stop collecting them. [Rejecting C] Removing the common measure conceals the gap rather than closing it, and removes the evidence that disadvantage is being produced.