The ARC has recently released
consultation documents for the 2015 ERA submission. The process is
largely unchanged from the 2012 rules. However sharp eyed readers
will have noticed business rules 131 and 132. These rules place a
limit on the percentage of publication outputs that can be recoded
from the journal discipline codes assigned by the ARC. The recoding
percentage is capped at 40%. This presumably has been introduced to
limit gaming by institutions trying to hide supposedly poor quality
or low citing publications. It could potentially have major impacts
for smaller institutions doing multidisciplinary research who may
struggle to research assessment thresholds. Its hard to believe this
important change was accidentally omitted from the change summary in
the submission guidelines – perhaps the rule changes were
considered and discarded but accidentally left in the business rules
document.
I suspect most Australian universities
did their very best to present the best possible submissions in 2010
and 2012 but its not clear that gaming significantly changed
outcomes. Its interesting to look at the 2012 outcomes in the 15xx
codes. A number of universities submitted assessable numbers of
publications in 1599, the 'not else where classified' subdivision of
the Commerce, Management, Tourism and Services (CMTS) group. Of the
15xx codes submitted by these institutions, who were presumably hiding
publications in 1599, only one of the defined 15xx codes scored a 4
('above world standard'), and handful scored 3 ('world standard').
The only success from this strategy was probably to divert and
consolidate some poor outcomes into a single code rather than
advertising apparent shortcomings in the economically important CMTS
disciplines.