Health and Social Wellbeing SurveyMethodologyThe questionnaire consists of a household interview followed by an individual interview with each person in the household aged 16 and above. The individual interview consists of core modules and modules that will recur on a regular cycle. Core items include accommodation, tenure, employment status, educational qualifications, family information, smoking and drinking and health and ill-health, Non-core items include physical activity and sexual health.
The 1997 run of the survey also included a physical measures module, completed by one respondent selected at random from each household. Qualified nurses were employed to record details of all prescribed medication that the respondent was taking, measure height, waist, hip, weight and blood pressure. A blood sample was also taken to measure the level of cholesterol (non-fasting). The Clinical Biochemistry Laboratory at the Royal Victoria Hospital carried out analysis of cholesterol levels.
Sample
The Health and Social Wellbeing Surveys in 2005/06 and 2001 were based on a systematic random sample of 5,000 addresses drawn from the Land and Property Services Agency´s property database. The LPSA addresses were sorted by district council and ward, so the sample was effectively stratified geographically.
The 1997 study involved the selection of a stratified random sample of 3520 addresses from the Land and Property Services Agency list of private addresses in Northern Ireland. Prior to selection, the list of addresses was stratified by Health and Social Services Board area and an equal number of addresses were selected from each area to facilitate analysis at HSS Board area level.
Fieldwork
Fieldwork for the 2005/06 survey was spread over a one year period from February 2005 to March 2006. Fieldwork for the 2001 and 1997 surveys was spread over a 6 month period from January to July. Each interviewer received a monthly allocation of addresses and collected the information covered in the survey by computer assisted personal interviewing (CAPI). Interviews were sought of all adult members (those aged 16 and over) of eligible addresses to yield a representative sample across Northern Ireland.
Weighting
In 2005/06 the results were weighted by age and sex to compensate for differential non response.
In the 2001 survey, no weights were used in the analysis. In 1997 the survey was designed to select equal samples from each of the Four Health and Social Services Board, which meant that overall the survey would not accurately reflect the Northern Ireland population. Individuals living in a HSS Board area with a large population had a lower chance of being included in the sample than individuals from a smaller HSS Board area. Therefore a NI weight was applied to the data to permit analysis at the NI level. This weighting process adjusts the results to those that would be achieved if the sample had been drawn as a random sample of addresses in NI.
A weight was also required for the physical appraisal element of the 1997 study. Only one adult was selected from each household for the physical measures, therefore individuals living in large households had a lower chance of being included in the sample than individuals in small households.
This weighting process adjusts the results to those that would have been achieved if the sample had been drawn as a random sample of adults rather than addresses.
| 
