Second-hand bogus poll
Headline: 1 in 3 women watch porn – survey
Opening sentence: One in three young women regularly view porn, with many watching it on their smartphone, it has emerged.
It turns out this is “Some 31 per cent of participants in the survey by magazine Marie Claire.” If you Google, you can find the Marie Claire invitation to take the survey, with a link. There are also Facebook and YouTube versions of the invitation. It’s a self-selected internet survey; a bogus poll.
Considered in the context of its original purpose, this survey isn’t so bad. It’s part of a major project(possibly NSFW) by the magazine, and its contributing editor Amanda de Cadenet, to discuss women’s use of pornography. The survey provided a way for them to involve readers, and a context for telling readers, however they responded, “there are lots of other women like you”. From that point of view the quantitative unreliability and poorly-defined target population isn’t such a problem, though it would presumably be better to have the right numbers.
Disconnected from the magazine and presented as data-based news, the survey results have very little going for them.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »