In Australia, university admissions have been in the news recently. The system used to admit most school-leavers, the Australian Tertiary Admission Rank (ATAR), is attracting a long line of critics.

converts the results of different state-based school systems into a common rank, downwards from 99.95 for students in the top 0.05?per cent of their age cohort. Universities use ATAR to ration places on high-demand courses; spots at prestigious universities or on courses leading to high-paying professions often require ATARs well into the 90s.

For universities, ATAR is a cheap and efficient selection system that plausibly links entry criteria to academic outcomes. First-year fail rates and attrition go up as ATAR declines. ATAR with grade-point average at university, and . An ATAR of 60 or above roughly equates to a student¡¯s chance of completion; a student with an ATAR between 60 and 69 has a nine-year completion rate of 65?per cent, while one with an ATAR between 70 and 79 has a 75?per cent completion rate, and so on. Most recent school-leaver?undergraduates have .

ATAR¡¯s critics do?not necessarily dispute its connection to academic performance. But they believe that it and . School students . Students from high socio-economic backgrounds dominate its upper ranks. Universities from applicants with disadvantaged backgrounds, but this changes the social profile of the most selective courses only modestly.

ÖйúAƬ

Despite these concerns, ATAR should not be dropped. No university will voluntarily give up its right to select students, so if ATAR were abolished, universities would find another way to measure school results or would create their own tests. Such action would also affect school behaviour and add to student stress.

ATAR is less useful for choosing between applicants with ATARs under 60. The prospects of an applicant with an ATAR of 55 are?not reliably different from one with an ATAR of 45. Both have about a six-in-10 chance of completion. One reason is that their ; ranking systems such as ATAR exaggerate small differences at the lower levels.

ÖйúAƬ

But for less-selective institutions, the main issue is filling their places. In practice, nearly 90?per cent of school-leavers who apply for university . The offer rate for applicants with ATARs in the 50s, and is nearly half for people with ATARs below 50. More lower-ATAR applicants could enrol if they chose a less-selective university.

ATAR supports self-selection more than university selection. Lower-ATAR school students are much for university, and much an offer. These sensible decisions would be harder to make if school results were less clear.

This self-selection continues after accepting an offer. All Australian universities provide a . In most, students do?not pay if they leave within the first four weeks. Those with ATARs below 60 are six times as likely to do so as students with ATARs above 90.

Of those who remain enrolled, those in the lower ATAR group are four times more likely not to continue to the second semester. One reason is that lower-ATAR students are much more likely to fail subjects. Although universities do?not usually discontinue students for unsatisfactory progress at this point, subject fails provide strong academic feedback that informs student decisions.

ÖйúAƬ

Perhaps other selection methods could help to identify which lower-ATAR students have the most academic potential. But there is little evidence that the alternatives do better than just admitting lower-ATAR students and seeing how they go.

These experimental lower-ATAR admissions do not abolish academic selection. Students still need the ability to pass subjects to progress ¨C it is just that the judgement is made post- rather than pre-admission.

ATAR cannot do all the work in matching school-leavers with university places. There may be no reliable substitute for the real-world experience of attending classes, completing assignments and sitting exams. But by reusing school results, ATAR is usually a cost-effective tool for both students and universities.

Andrew Norton is ÖйúAƬ programme director at the Grattan Institute. His latest report is .

ÖйúAƬ