New Methods for Survey Research,
21 – 22 August 1998, Chilworth Manor, Southampton

The Compstat conference series is organised under the auspices of the International Association for Statistical Computing (IASC, with which ASC has a formal affiliation) and has been running since 1974. In 1998 Compstat returned to the UK for the first time since 1980. This seemed an ideal opportunity for ASC to hold an associated (satellite) conference, particularly trying to enhance the link between Survey Computing and the Statistical kind.

Unlike our previous international conferences, we decided to organise this one on a smaller scale, with no parallel sessions, and to make it more fully residential. In constructing the scientific programme we joined forces with the Centre for Applied Social Surveys (CASS), the Office for National Statistics (ONS) and the Market Research Society (MRS).

The conference was held at Chilworth Manor, a residential conference centre run by Southampton University, just outside the city. This set a limit of about 100 on the number of participants we could accommodate, and we almost reached the limit. Most of the participants arrived on the Thursday evening before the conference, so everyone had a chance to chat to old friends and introduce themselves to new ones. The programme was designed to present methods (generally relatively new methods) which are widely used in specialist areas but have not made their way into everyday use in survey research or analysis. We hoped that participants would gain some new ideas, together with enough insight to judge whether the methods can be useful in their own organisations. Each half–day was organised as a separate session, covering different (though related) topics. Each started with an invited (keynote) presentation setting out the main ideas and issues involved, followed by a number of presentations on the practical issues of using the method, and the benefits that result from it. Full details of the programme are available, and most of the presentations can be downloaded. All the papers are available in a printed volume.

The format and content of the conference seemed to be appreciated by the participants (based on our usual assessment questionnaire – with nearly 70% response rate). People were consistently enthusiastic about Chilworth (both the place and the staff), and the social aspects worked well. We seemed to have been successful in drawing in new people, since two thirds of the participants had not been to a previous ASC event, though we still only managed about 30% from the private sector. 80% of those who had been before thought this was better than previous events, and overall 90% thought themselves likely to come again.

Opinions about running into Saturday were mixed, with strong feelings both for and against. Numbers did drop off significantly by Saturday afternoon, though this was perhaps because some perceived this session as more difficult. This was a disappointment, since the objective was to explore new methods.

The chart shows the ratings over all 15 papers for usefulness: as a whole, 96% rated the conference as good or very good in terms of relevence to their work.

Overall, the committee was very pleased with the event, and we shall certainly try this format again, though not until after the next full international conference in Edinburgh. See you there!