 |
 |
Dear Allstaters, Please find below a final reminder about a half-day RSS Leeds/Bradford Meeting on "Evidence and Policy" to be held next Monday 11th December. Further details can be found on our webpage: http://www.maths.leeds.ac.uk/statistics/rss/current.html#POLICY Regards, Paul =================================================================== Dr. Paul D. Baxter Secretary/Treasurer, RSS Leeds/Bradford Local Group, Department of Statistics, University of Leeds, Leeds, LS2 9JT, UK. ------------------------------------------------------------------- Leeds/Bradford: Monday 11 December, 2pm-5pm, Room ALG07, Leeds Metropolitan University (City Site) Evidence and Policy; How close do they get? Phil Willis MP (Harrogate and Knaresborough) (Chair of the House of Commons Science and Technology Committee) The politics of statistics In November, the House of Commons Science and Technology Select Committee will publish the final report of a series of publications entitled "Scientific Advice, Risk and Evidence: How the Government handles them". It will test the extent to which policies are "evidence-based". Effective policy development requires both a healthy scientific advisory system and appropriate use of advice on science and risk by Government. In light of the publication of this report I shall discuss the conclusions and implications. Statistics are constantly being used as evidence to formulate and refute policy and my talk, I hope, will provoke the audience to discuss the issues surrounding evidence, policy and the politics of statistics. Sheila Bird (MRC Biostatistics Unit, Cambridge and University of Strathclyde) Aftermath to Royal Statistical Society's report 'Performance Indicators: Good, Bad, and Ugly' After a brief introduction to the report and recommendations of RSS's Working Party on Performance Monitoring in the Public Services, I'll chart the aftermath (roadshows - political and media, consultations, and methodological) with specific reference to criminal justice and 'war on drugs'. Aubrey Blumsohn (formerly University of Sheffield) Statistics and ghosts Effective medicine requires honest data analysis. Following several "scandals", the credibility of pharmaceutical research is being questioned. Industry "ghosting" of science has become standard practice - statistical analyses are performed by industry and presented in the scientific literature by academic authors who have not seen raw data, nor written the publications. Drug regulators and NICE use these data, and accept tabulated company summaries "with blind faith". The factors which have perpetuated this state of affairs, and the consequences for research and patients will be discussed in the light of recent cases, including a personal one in Sheffield. David Torgerson (Trials in Public Policy) The need for more RCTs in the social sciences Randomised controlled trials, the gold standard evaluative method, can be undertaken in non-health care settings, such as education and criminal justice but are under-utilised in such areas. In this talk I will look at several examples of recent RCTs undertaken in the social sciences. Despite record levels of investment, few rigorous RCTs of computer technology have been attempted. One such RCT found no evidence of benefit to teach literacy skills. There are other examples of where some RCTs have been used and have produced interesting and perhaps surprising results. In summary, RCTs are required not only to demonstrate benefit but also to prevent implementation of ineffective policies.