Australian universities employed 24,000 fewer people in 2020 than in 2019, yet they increased their full-time equivalent workforce by about 4,500. This is one of many improbable conclusions that can be drawn from a?hotchpotch of incomplete, inconsistent and sometimes error-ridden statistics that make it difficult to draw firm conclusions about staffing levels at any institution.

While the lack of statistics on Australia¡¯s casual university workforce is widely recognised, a?Times Higher Education?analysis has found that figures on tenured and fixed-term staff are also ad?hoc and unreliable. Headcounts at individual universities can differ as much as threefold, depending on which data source observers employ,?while some universities claim to have more actual full-time members of staff than the number of full-time equivalent staff they say they have.

The data provide conflicting accounts of Covid¡¯s impacts on staffing last year.?More than 70?per cent of universities increased employee numbers, according to one source; a different source says they?cut?personnel.

Inconsistencies occur within as well as across datasets, with implausibly large changes from one year to the next ¨C for example, casual employee numbers at one institution rising sixfold. Apparent copy-and-paste errors are commonplace, with exact figures repeated down to the decimal point.

ÁñÁ«ÊÓƵ

Many discrepancies can be explained by variations in reporting requirements, with agencies imposing different staffing definitions and census dates. University headcounts mushroom if volunteers and subsidiary organisations¡¯ employees are included, while double counting of staff with multiple roles or contracts can also blow out the numbers.

Casual employment adds more uncertainty. Casual staff are excluded from Department of Education, Skills and Employment (DESE) headcounts?but are generally included in those reported by the Australian Charities and Not-for-profits Commission (ACNC) and by the Workplace Gender Equality Agency (WGEA).

ÁñÁ«ÊÓƵ

Casual counts also vary depending on whether they involve one-day snapshots or more sustained periods, and whether they are recorded mid-semester ¨C for example, the DESE census date of 31?March ¨C or in holiday periods?such as late December, when ACNC figures are tallied.

Some data sources impose no consistency around census dates, with the WGEA allowing universities to choose their own. And some universities completely omit employee numbers from their annual reports, while others offer rudimentary figures with little explanation of when they were recorded or who they cover. Only Victorian universities¡¯ annual reports offer consistently detailed data.

The different reporting rules and conventions do not explain data that ¡°just fails the pub test¡±, according to the National Tertiary Education Union¡¯s Victorian assistant secretary, Sarah Roberts. ¡°Blind Freddie could see there was a problem. Where¡¯s the fact-checking?¡±

Ms Roberts said it was ¡°astonishing¡± how many universities did not appear to know how many people they employed, sometimes hiring people to generate retrospective employment records. PhD students working as tutors ¡°don¡¯t get paid for six weeks because nobody¡¯s found their paperwork¡±, she said. ¡°These are multimillion-dollar organisations. What¡¯s going?on?¡±

University administrations¡¯ difficulties in quantifying what they routinely describe as their most valuable resource belies the authenticity of what Ms Roberts called the ¡°hand-wringing¡± over redundancies. The ¡°rubbery figures¡± also thwart efforts to research the sector ¨C ironic given that universities themselves employ researchers who rely on precise data, and whose journal articles can be withdrawn over inadvertent errors.

THE?asked nine universities to explain apparent discrepancies in their reported staffing levels. Most blamed differences in reporting criteria, adding that streamlined data requirements would reduce their administrative burdens while boosting consistency.

ÁñÁ«ÊÓƵ

But the?THE?analysis suggests inconsistent interpretations of the reporting requirements. Ten universities reported headcounts to the WGEA that exceeded the figures reported to the ACNC by at least 1,000 people ¨C and in one case, almost 6,000 ¨C while five universities exhibited the opposite pattern, with ACNC headcounts between 1,000 and 6,700 higher.

Of the nine universities approached by?THE, only two confirmed the accuracy of all their figures. One acknowledged reporting errors, while two others stood by the figures in their annual reports but not other data collections. Two said they would review or amend their ACNC?data.

ÁñÁ«ÊÓƵ

Two indicated that they had not provided the figures attributed to them by the ACNC. Another nine universities have missed deadlines to report to the ACNC, according to its website.

Macquarie University accounting professor James Guthrie said staffing data from the various sources were impossible to reconcile. He blamed administrators for downplaying job losses during the pandemic.

Others blamed a culture of complacency nurtured by agencies that impose onerous and duplicative reporting duties on universities and then make no effort to check the results ¨C a suggestion that both the ACNC and the DESE denied.

Australian National University policy expert Andrew Norton argued against ¡°routine audits without cause, given their cost to both sides¡±. But he supported ¡°reality checking¡± of staffing data that were?inherently unreliable because of seasonal employment fluctuations and confounding factors?such as job-sharing.

Professor Norton said agencies had limited interest in checking the accuracy of data that did not affect them financially. ¡°None of the staff datasets were designed with research in?mind,¡± he added.

University of Sydney sociologist Salvatore Babones said the lack of data on Australia¡¯s taxpayer-supported institutions was ¡°shocking¡±. ¡°Universities around the world ¨C certainly the English-speaking world ¨C provide detailed data on staff and students as a matter of course,¡± Dr Babones said. ¡°It¡¯s a travesty that Australian institutions are so secretive.¡±

ÁñÁ«ÊÓƵ

john.ross@timeshighereducation.com