Statistics in research and development

In this article we report the results of a survey of National Statistical Institutes concerning two activities: (i) Research and Development (R&D) work within an agency, and (ii) scientific co-operation of a National Statistical Institute with the universities. A prerequisite for R&D is’ the existence, within a statistical agency, of a certain R&D infrastructure. It includes such components as a well-documented research plan, a scientific or professional board with representation from the academic community, and funds and procedures to support scientific research by staff members. Networking is another key element. Forms of networking with universities include long-term frame contracts, joint academic posts, various fellowship schemes, and joint research projects. Our results indicate that there was large variation between National Statistical Institutes (and groups of such institutes) in the levels of R&D infrastructure and of R&D networking. A high level of infrastructure often accompanied a high level of networking. When both levels are high, the chances for a successful implementation of research results into the production of statistics are improved. The two levels are related to agency size: both tended to be higher in large National Statistical Institutes than in smaller ones. The survey was carried out in 1999–2000. The target group consisted of 52 statistical agencies around the world, covering most European countries and selected countries outside Europe. A total of 41 agencies responded to the questionnaire.