Reply #132: Argumentum ad numerum

[Euromutt]

A majority of Californian voters voted in favor of Prop 8; that didn't make it right.

Moreover, that piece has a few problems. The first is that it has the names of Arthur Kellermann and Garen Wintemute in it, both notorious for twisting data to support a predetermined conclusion, and writing abstracts and conclusions that conclude that Guns Are Bad, even though the research does not support that conclusion.

The second is that it's from the NEJM, which has demonstrated a sufficiently strong editorial bias against firearms that it has unquestioningly printed research (like Kellermann's) so shoddy it would have received a failing grade had it been written by an undergrad.

Note in the context of the above that the article does not include a list of the questions that the survey respondents were asked. It's well established that how you phrase a question plays a major role in what kind of answer you're likely to get. Zogby, for example, is known to specialize in producing the results the party commissioning the survey wants to get by tailoring the questions to generate the desired response.

The third is that it's from 1998. Relying on opinion polls to guide public policy is bad enough, but relying on 12+ year-old opinion polls is downright terrible.