Data analysis explained
I worked closely with Dale in analysing the data for this study and would like to point out to M. Burns that the percentages shown on the charts are percentages of the full 400 respondents, so for example in figure 1, 16% thought Apple were High Quality and Good Value this equates to 64 respondents. I would like to apologise if this is too difficult to work out for some readers.
The reason the bars do not add up to 100%, (see comment by Anonymous 16.20 GMT), is that not all respondents had an opinion on every brand therefore the plots show only those who did have an opinion.
Re the total number of respondents, while it would have been great to have had a few thousand, some polls attract more readers than others, and it wouldn’t be right not to report back results of polls just because response are lower than we would like. Having said this, 400 (it was actually 402) is not a bad sized sample to get a flavour of what’s going on provided you bear in mind the caveats mentioned in the article.