You are obviously misusing the statistic. First off, the question that 55% answered positively was not whether they liked the NSA, but whether "in order to prev

You are obviously misusing the statistic. First off, the question that 55% answered positively was not whether they liked the NSA, but whether "in order to prevent terror attacks, you have to live with mass spy activities". That is shameless framing because you give a negative consequence no one wants (terror attacks). It is not too surprising that a majority answers yes to the solution the interviewer proposes. But would such a majority of interviewees come up with the solution themselves? I highly doubt it. Just ask any passerby: "Please finish this sentence: 'In order to prevent terror attacks, spying ...'. Additionally, the very same interviewees are surprised just how much Germany has been spyed on (61%). Note the framing here again. Here you have no evaluation, but just a very weak sensation ("surprised"). The interviewer could also have asked: "Do you like how much the NSA spied on Germany?" - and I bet the answer would have been clear. And again, there are very few topics that we think the public cares about (taxes, pension) and again, how do we notice they might care? Media. As with privacy. As with immigration. As with any other political topic.