If you consider that the survey indicates that 57% of the country watch broadcast via a satellite provider, you can see the massive bias. DStv and StarSat would be over the moon if they could claim more than 28 million viewers between them.
It would be far better to indicate the limitations of the survey upfront, and to ensure that any analysis and results take those limitations into account. The survey itself has some merit, but not when generalisations make the results appear meaningless. This is not what I would have expected of a company like Nielsen (unless this is just bad reporting).
Dezon of Eat My Words | http://eatmywords.com/portfolio/before-and-after/