Erratum

Abstract

Gorard S (2016) Damaging Real Lives Through Obstinacy: Re-emphasising Why Significance Testing is Wrong. Sociological Research Online 21(1): 2. DOI: 10.5153/sro.3857 It has been brought to the attention of the Editors and the Publishers that some corrections requested by the author while reviewing the proofs had inadvertently been missed ahead of first publication of the above article on 28 February 2016. The author’s corrections were incorporated into subsequent versions of the online article, and there was an unintentional delay in uploading the corrected PDF version of the article online. For clarity of the scientific record, the corrections are outlined in this erratum: In paragraph 4.4 The sentence ‘On the first run, 1217 p-values were below 0.05 (this represents around 5.5% of the samples).’ was corrected as follows: ‘On the first run, 1217 p-values were below 0.05 (this represents around 12% of the samples)’. In paragraph 4.5 The sentence ‘Lack of normality may reduce the so-called “power” of the test slightly, but with 268 cases (deemed a very large N in most resources), this has been shown not to matter ( http://thestatsgeek.com/2013/09/28/the-t-test-and-robustness-to-non-normality/ )’ was corrected as follows: ‘Lack of normality may reduce the so-called “power” of the test slightly, but with 200 cases (deemed a very large N in most resources), this has been shown not to matter ( http://thestatsgeek.com/2013/09/28/the-t-test-and-robustness-to-non-normality/ )’ Sociological Research Online apologises to the author and the readers for any inconvenience this may have caused. The correct and citable version of the article is accessible at the following DOI: 10.5153/sro.3857