[elil17]

Interesting, but the sample size discussed is very very small - 204 households - and, as the researcher notes, there are many confounding variables to consider. To me, this does not at all suggest that a change of public policy is warranted at this time, but it would definitely be interesting to see more research.

[raxxorrax]

Problem is that any source of data will have the same quality issues. Not alleging general propaganda, but I guess it always tends to confirm the initial suspicions of researchers as the data can be molded either way.