A THIRD of United States patents filed in 1996 cited scientific papers, 75 per cent of which were produced by public-funded research centres, such as universities or government laboratories, writes Julia Hinde.
"This is very strong evidence for publicly funded science as a means of generating corporate technological innovation," Diana Hicks, one of the researchers at CHI Research which did the analysis, told a conference in Cambridge this week.
The analysis showed that only a quarter of the papers cited came from industrial scientists. It also found a 15 per cent increase in patents citing scientific papers since 1985.
Dr Hicks told the Fifth International Conference on Science and Technology Indicators that she believes the analysis influenced the US decision to raise its science budget this year.
Grant Lewison, of the PRISM unit at the Wellcome Trust, said: "It's an important message, but this was for US industry. We have not done the same for British industry."
Dr Hicks said each country's patents preferentially cited papers from their own country by a factor of between two and four. For example, Britain produces 8 per cent of the world's scientific literature, but British companies' US patents citing scientific research called on UK work 23 per cent of the time. "It shows you have to do the research yourself," she said.
Sylvan Katz, from the Science Policy Research Unit at Sussex University, told delegates that as the number of scientific papers a country publishes rises, the number of citations it gets rises faster.
Ben Martin, unit head, said that Dr Katz's results contradicted research last year by Britain's chief scientific adviser Sir Robert May. This showed small countries, such as Switzerland and Israel, doing disproportionately well on citations. He said this could be because smaller countries tend to publish in international journals.
Sir Robert was also challenged by the National University at Canberra. Last year Sir Robert wrote that the international cost-effectiveness of UK research may be related to the high proportion carried out in universities rather than research institutes.
But Australian research found the highest impact was from institutes where staff researched full-time rather than from universities where people teach and research.