A STUDY into how scientists report animal welfare research has found strong evidence that they are greatly influenced by their funding agency.

An investigation by the University of Queensland's Centre for Animal Welfare and Ethics found authors of industry-funded research tended to be more negative about the welfare of animals in experiments.

Conversely, authors of charity-funded research were more positive about the welfare of animals in new methods, compared with authors of government-funded research.

The study, by Professor Clive Phillips and Agnes van der Schott found scientists working on animal industry-funded studies were more likely to report that new methods to improve welfare did not work.

Prof Phillips said this was presumably "because the new methods could increase costs" for livestock producers.

In contrast, scientists working on studies funded by animal advocacy groups were more likely to report welfare benefits, he said. "It is of some concern if scientists are putting a spin on and manipulating the results."

Such bias had been found to occur in pharmaceutical sciences, he said, but had not been tested in other areas such as climate science.

Prof Phillips called for all scientists to undergo ethical training and for organisations to prevent scientists from accepting grants that come with conditions about reporting results that were favourable to the funding agency. In separate work, Prof Phillips also found that small farmers were keenly concerned about their animals' welfare.

Victorian Farmers Federation livestock president and former Sheepmeat Council of Australia president Ian Feldtmann said he had not read the University of Queensland study but he remained confident "there is not bias in the animal welfare research that our levies help fund".