Standardisation
A report on problems encountered in standardising the results for the 1992/93 Household Health survey and possible alternative methods
Research Report 5
The Household Health Survey was the first national health survey to be carried out in New Zealand. It was conducted between April 1992 and March 1993 by the then Department of Statistics and funded by the former Department of Health and the Department of Statistics. The objectives of the survey were to investigate the state of health of New Zealanders, to investigate differences in the level of health between different population groups and to look at lifestyle factors that influence health. It also looked at the use of health services and other responses to illness. The results were presented in ‘A Picture of Health’.
The analysis of the Household Health Survey results, therefore, involved comparison of the rates of occurrence of health related topics across different subgroups of the population, such as Europeans and Mäori, or people defined by marital status. This sort of comparison is not a problem if the subgroups are similar with regard to other factors that are associated with the topic being analysed. However, most health related topics are closely related to the age and sex of the respondent - older people typically have less good health than younger people and females tend to live longer than
males - and different subgroups of the population have very different profiles by age and sex -widowed people as a group would tend to be older than those never married while the median age of Pacific Islands people is much lower than that of other ethnic groups. Thus, a direct comparison of the proportion of single people having good health with that of widowed people may be misleading as the difference may owe more to the fact that the widowed people were older
rather than that their state of widowhood was contributing to their less good health, which was what the analysis was trying to assess.
As a result, it was decided to look at the possibility of removing the effects of age and sex from comparisons of subgroups of the population. This process is known as standardisation.
(124Kb)
If you do not have the Adobe Acrobat Reader you may download the reader to view or print the contents of this file.
back to top
For enquiries and information requests
This page last modified on: 18 June 2003