Survey data alone is no longer enough to capture the complexities of a fast-changing media environment, with more content choices and more distribution platforms from which to access them. We are demanding more and more from our respondents at a time when many are becoming less willing to participate in long or boring surveys. Employing statistical techniques allows us to collect less data from individuals and to tolerate less precision in their answers.
The more important limitation for the purposes of this paper is how we extract as much information as possible from our respondents. There are natural limits to how much time people are prepared to spend answering questions. Given that few are blessed with perfect recall, we also have to be careful to word questions clearly and not to ask for information it is unreasonable to expect people to remember.
演算法的時代 章節1/2 – 資料歸屬（Data Ascription）