You've reached our premium or archival content
To access this page, and more great content just like it, you need to become a paid subscriber.
If you already have an account, please login.
Otherwise, registration is quick and you'll have access instantly after payment.
A survey will be unbiased if the sample is truly random and that sample is truly representative of the population. That is taught in first year university statistics, for example STAT170 at Macquarie University. I know because I taught it there.
Measuring ratings by people just "rocking up" to the GFK website may not be representative of the population. The results may be skewed for the following reasons:
(1) People may have their favourite radio station and will participate.
(2) People may have their non-favourite radio station and want to favour another radio station.
(3) People may not be tech savvy and not participate at all - it affects all demographics.
The above list is not exhaustive with the consequence that the ratings results may not truly reflect the populations' listening habits.
Thank you,
Anthony of thinking Belfield