This component contains
- [Visualizations of all 11 specification curves that we report in our manuscript][1]
- Tables with details about the sample sizes, [breakdown by sibship size and birth order position][2], [breakdown by analysis][3]
- An additional [re-analysis of birth-order effects on the Big Five traits][4] including a very brief discussion and visualizations of the specification curves
- Analysis scripts (starting from [here][5]) and an R workspace file, see folder within this component
[All SOEP questionnaires can be found on the DIW page.][6]
Notice that the original data files from the SOEP can be accessed via the German Institute for Economic Research, see [https://www.diw.de/en/diw_02.c.222836.en/access.html][7] for details.
You should be able to run the specification curve analyses including the permutation test based on our analysis scripts and the SOEP-CORE distribution (v31.1). However, I don't necessarily recommend that because the implementation is inefficient and it might take multiple days to run the permutation tests (which basically consist of running >= 720 models 500 times for each of the 11 traits).
The workspace file contains the "results frames" for all traits from which you can get the p-value and the difference between first- and laterborns for all specifications. It furthermore contains the p-values based on the shuffled samples. Thus, the workspace should contain all relevant results, but please don't hesitate to contact me (Julia Rohrer) in case you have any questions.
[1]: https://osf.io/vg2un/
[2]: https://osf.io/dtwz9/
[3]: https://osf.io/vqsh7/
[4]: https://osf.io/wu7h3/
[5]: https://osf.io/y6dtu/
[6]: https://www.diw.de/en/diw_02.c.222729.en/questionnaires.html
[7]: https://www.diw.de/en/diw_02.c.222836.en/access.html