Treatment of missing data determined conclusions regarding moralizing gods
| Authors | |
|---|---|
| Year of publication | 2021 |
| Type | Article in Periodical |
| Magazine / Source | Nature |
| MU Faculty or unit | |
| Citation | |
| Web | https://www.nature.com/articles/s41586-021-03655-4 |
| Doi | http://dx.doi.org/10.1038/s41586-021-03655-4 |
| Keywords | missing data; moralizing gods; cultural evolution |
| Description | Whitehouse, et al.1 used the Seshat archaeo-historical databank2 to argue that beliefs in moralizing gods appear in world history only after the formation of complex ‘megasocieties’ of around one million people. However, inspection of the data they used shows that 61% of the data points on moralizing gods in the Seshat databank are missing values, mostly from smaller populations of less than one million people. In their analysis, the authors re-coded these data points to signify the absence of belief in moralizing gods. When we confine the analysis to only the extant data, or instead use various standard imputation methods, the reported finding is reversed: moralizing gods precede increases in social complex- ity. Our reanalyses suggest that the reported ‘megasociety threshold’ for the emergence of moralizing gods is an artefact of the decision to re-code all missing data as known absences of moralizing gods. |
| Related projects: |