The pass-along rate isn’t some number ginned out of the air. It is the result of hundreds and hundreds of surveys over many, many (many!) years. Scarbrough, Media Audit — you pick the survey entity, and the results pretty much always show a pass-along rate of between 2 and 3 readers for a daily newspaper. And how do these surveys show this? Let’s consider the Media Audit, which takes a random sample representative of a metropolitan area and then asks (among hundreds of demographic and behavioral questions) each person in that sample whether he or she has read a particular publication within the last day, week, month. Because the sample is a random sample, the answers to those reading behavior questions can be generalized to the metro area (within the confidence range of the survey, of course). And when you do that, lo and behold, you come up with somewhere between 2 and 3 readers for every newspaper that is printed/circulated. The comment on pass-along rates in this post — “without being able to actually see what people do, rather than what they claim, it’s impossible to be totally accurate” — is simply untrue. When you have hundreds of surveys over decades, all with reasonable confidence intervals, showing the same thing, then you can bet the house that what they show is “totally accurate.”