This article describes the application of co-nomination analysis, a technique designed to map the structure of a research community. The technique was used as part of the evaluation of the UK national information technology programme, which sponsors collaborative research between firms and between firms and universities. Co-nomination networks are based upon responses to questionnaires which researchers are asked to nominate other researchers whose work is similar or relevant to their own. Researchers nominated in the same response are presumed to be linked, and where these links occur with multiple frequency, the likelihood of their being significant is increased. The article describes the extension of a network which had been previously identified and compares the citation scores of researchers identified in the networks. It is concluded that the networks represented were realistic and a useful input to the evaluation. Industrial and government researchers with low citation scores were in some cases central to the networks, suggesting that co-nomination is useful in areas where publication is restricted or considered less important. Further development of the technique is planed.