Materials (ie. data) are available for download in the files section of the [Data component][1]. A note from Dr. Heine: "Attached is the datafile that includes all the values that were in our paper (it's the same file that was linked to the paper in footnote #1). We didn't actually collect any of this data but instead used the values from publicly available sources - hence we don't have any questionnaire materials or lab protocols to share. here are a few points that we think are important to keep in mind in doing this replication: 1 - the goal of our paper was to question terracciano et al's claim that national stereotypes are incorrect, based on the evidence that national stereotypes don't correlate with aggregate personality traits. we questioned whether aggregate personality traits themselves are valid, because when people complete personality trait measures their responses are anchored to local cultural norms (aka, the reference-group effect). in contrast, national stereotypes are not anchored to local norms (people are aware that they're comparing countries by international norms), so we hypothesized that national stereotypes might be more valid than the aggregate personality traits when assessed with objective behavioral and demographic measures. 2 - one challenge with replicating this is that we don't report any inferential statistics. we report the average correlations between different measures of aggregate personality and our criteria in tables 1 and 2. our results were supported in that the aggregate personality traits overall didn't correlate with the criteria, whereas the national character stereotypes did. we never actually reported on whether the difference between those correlations was significant as we had received advice that inferential tests that compared average correlations involved some disputable assumptions, so we just reported the average correlations rather than conducting any statistical tests on them - the values were pronounced enough that they spoke for themselves. looking at the individual correlations between individual measures and individual criteria, reported in table 1, for the aggregate personality traits 0 of the 30 correlations with individual criteria reached conventional levels of significance in the positive direction, and 6 of the 30 were significantly correlated in the opposite direction. in contrast, 9 of the 10 correlations were significantly correlated with the individual criteria in a positive direction for the national stereotypes. so i'm not sure of what kind of evidence here would count as a replication - i guess again it would be null correlations between the aggregate personality traits and the criteria and positive and larger correlations between national stereotypes and the criteria. 3 - with regards to what datasets to use, we know of only one measure of cultural stereotypes of conscientiousness, from terracianno et al, and we used that. if there are others that would exist, they would be great to include. 4 - any aggregate measure of conscientiousness measured either by self-report or peer-report should do that has been collected in multiple cultures should do. we found 3 measures at the time that were collected across a broad range of countries. there are probably more by now. all of these measures should be equally valid for assessing their correlations with criteria. 5 - for the criteria, we were wanting to include any kind of objective behavioral or demographic measure of conscientiousness - something that wouldn't be influenced by people's self-evaluations, which we argue are always tethered to local cultural standards of conscientiousness. we knew of the levine and norenzayan dataset, which included accuracy of bank clocks, postal worker efficiency, and walking speed. conscientiousness has been linked with health and income, so we also included national measures of health (longevity), and gdp per capita. there likely are some other measures that could be incorporated here. oishi and roth, 2009, also included a measure of corruption as an index of conscientiousness (which largely replicated our findings). i think any measure that shows that people are exerting some impulse control for the greater good could be argued to qualify as an indicator of conscientiousness, e.g., measures of cleanliness, law-abidingness, tax evasion. we think this is a place where some tricky judgments calls come into question, as there is grounds for debate whether a particular index assesses conscientiousness as well as another index. i think the selection of these criteria will straddle the line between a direct replication and a conceptual replication. 6 - conceptually, other trait measures (e.g., extraversion, agreeableness) would also test our hypothesis, however, there may be factors that make self-report of other personality traits more or less linked to behavioral measures. we didn't expand beyond conscientiousness as the behavioral predictors are better documented in the literature for conscientiousness than they are for other trait measures. i would think that looking at other traits would be more of a conceptual replication of our paper rather than a direct replication." [1]: https://osf.io/u35rd/