Search

South African television ratings are obtained from the AMPS meter panel. This panel must be viewed as a complex non random sample. For such samples the effective sample size differs from the actual sample size. It has been found that, when equal weights are assigned to strata, the most reliable estimate for effective sample size can be obtained by considering every household as a sample cluster. This estimate of effective sample size can be incorporated directly into a test for significant rating change. For convenience this test is implemented graphically.

A graphic procedure for determining whether a statistically significant change in television ratings has occurred is proposed. The method takes into account the fact that the ratings are generated from the same sample of viewers rather than from two independent samples. In addition it incorporates the effect of sample size. The method is applicable for the analysis of all market research data collected by such panel methods.