Reader Comments

Post a new comment on this article


Posted by SLewandowsky-849 on 14 Jan 2015 at 18:06 GMT

The Age variable contained two outliers that represent typos or software errors in otherwise correct records. Those outliers did not affect the summary statistics reported in the article which are correct as stated. However, they did affect the reported correlations with the indicator variables, which requires correction of a single sentence. In particular, the statement on page 9 that "age turned out not to correlate with any of the indicator variables" is incorrect and it should read instead "age correlated significantly with 3 latent indicator variables (Vaccinations: .219, p<.0001; Conservatism: .180, p < .0001; Conspiracist ideation: .140, p < .0001), and straddled significance for a fourth (Free Market: .08, p .05). Inclusion of Age (minus outliers) as a covariate in the structural equation model in Figure 2 does not notably affect any of the weights and has no impact on any of the conclusions.

No competing interests declared.

RE: Corrigendum

SLewandowsky-849 replied to SLewandowsky-849 on 05 Feb 2015 at 09:24 GMT

One of the correlations in the above corrigendum is missing a minus sign (the .140 involving conspiracist ideation should be a -.140).

The minus sign was dropped during the copy and paste process from a Word document (which contained the correct numbers) as is shown in this video: .

PLOS ONE has been alerted to this apparent glitch and they have forwarded it to their IT folks. It is not entirely unusual for "special" characters to create problems during pasting, but I have never seen a minus sign disappear without a trace.

No competing interests declared.