<Minnerva>

Household Surveys are complex to organise and expensive to run - obtaining best value from them is a priority. This means that the survey manager needs to be in control of the process and to have some immediate feedback on both the quality and content of the data being returned so that corrective actions can be taken to reduce the possibilities of bias entering the data. Depending on the nature of the survey and the way in which the sample might have been stratified, the potential for bias exists with:

• Day of week the interview takes place
• Response rates by geographical area
• Interviewer performance

Interviewer performance is particularly relevant. Not only might some interviewers be better or worse at obtaining the interview in the first instance, it might be that there are inconsistencies between interviewers in how they have interpreted the nature of the survey and the way in which questions are asked. In the context of a travel survey we might want to monitor the average number of trips per household recorded by each interviewer to see if there are extremes being recorded. This lets the Survey Manager investigate how Interviewers who might be at the extremes are conducting their interviews.

HSMIS is a tool that brings together the sample and the interviewers. The Survey Manager can then:

• Allocate work to interviewers and issue them with work sheets
• Log interviewer’s returned work back into the system
• Monitor and analyse field progress
• Monitor and analyse progress by interviewer

Together with a range of additional features and tools, HSMIS increases the productivity and control of the Survey Manager, all of which contributes to higher confidence in the data being collected. The HSMIS database can be linked with GIS providing a visual understanding of the sample location. Sample addresses can be issued to the Interviewers through this medium, enhancing the dialogue between interviewer and the Survey Office.

(Contact: Martin Bach)