Besoeker

That assumes you know with certainty:

1. that there was a "won't say" category
2. and that you know with certainty that some of those in that category, which may or may not have existed, were of the same opinion as the 30% who would and did say.

Do you?

Britannist

All opinion polls have a "won't say" category - opinion polling organisations can hardly tell the people they interview that they have got to give their view can they? All those answering questions for an opinion poll also have the option of choosing the "don't know" or "undecided" categories.

The "won't say" category will be a mix of those who agree with the 30% and those who do not - possibly more disagreeing with the 30% than agreeing. It is very unlikely that everyone in the "won't say" category will all be of the same view.

The poll result (referred to in posting one to this thread) says 30% believe that there is too much UK media coverage of alleged global warming. The very existence of the "won't say", "don't know" and "undecided" categories in opinion polling make it almost impossible for all of the remaining 70% to be all of the opposing view to that held by the 30%. The 70% will be divided into different categories (i.e. perhaps 55% opposing the view of the 30% with the other 15% of no view).

Besoeker

Don't you see the disparity between the two points in bold?
Certainty v probability.