For a good time the surveys CADEM Public Square have been transformed into part of the agenda of the media (the media are published and used). A tracking weekly public opinion has its utility well you can think of.
Now, in the methodology section -that is to recognize CADEM puts at the start of their reports, so that no one can claim that is not told, at least – the following is said about the shows (then tell us that you used two techniques to produce the information)
For the interviews through telephone sampling was probabilistic, from DATABASES with national coverage, own of CADEM, and within the home, the selection of subjects was made by quotas of sex, age, and SES (High C1-C2 Medium C3; Low D/E).
For the interviews face-to-face in fixed point with tablet is pre-defined fees for specific communes in the Metropolitan Region, Valparaíso and bio-bio, in addition to sex, age and GSE (ABC1, and D/E) as a complement to the sampling of the fixed telephone.
And on the size of the sample:
711 cases. 508 interviews were implemented by telephone and 203 interviews face-to-face inflow points. Margin of error of +/- 3,7 percentage points at 95% confidence.
To tell you the truth, it does not work. It makes sense to supplement landline phone (given the fact that your coverage is not only low but also decreases, and in addition, not only is it biased by GSE but also by age, households, more young people tend to have less fixed line). But the interviews in fixed point are not a method to control bias whatsoever, for the simple reason that it is a technique that has no type of control. Even dare to put a margin of error in these circumstances is something of an invalid.
You can make arguments that there are excuses: it is Not possible to assume that a tracking weekly keep the standards of other studies, that the information you give is not so far-fetched. And if one is in the mood magnanimous, you might even get to listen to them. But the fact remains that they are studies that do not meet standard one. The fact remains, moreover, that studies of surveys are used by the exact figures that deliver: we don’t need a study to know, for example, that the popularity of the government has fallen; what we do not know, and it would be interesting to know, is how much-precisely the reason why we do surveys. But the number is what we can’t use polls CADEM Public Square.
In other words, we need to require and to do things with a minimum of seriousness.