Skewed informant-report scales can be corrected with a log transform

Evidence

Prior to analysis, informant I-AB was found to be non-normally distributed (skewness 2.30, kurtosis 4.87). To reduce this skewness and kurtosis, informant I-AB was logarithmically transformed (skewness 1.82, kurtosis 2.35). — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims, 101

Categorization

• XREF: Connects to psychometric practice of reporting skewness/kurtosis cutoffs before transformation; pairs with any methodology notes on cleaning Likert-scale aggregate scores.
• QUESTION: A log transform of a scale that likely includes zero or negative values raises questions about how the offset was handled — worth checking the original data treatment.