This component hosts deidentified ML4 datasets for public use. To request the dataset with all variables collected (including sensitive ones that cannot be shared openly), contact Rick Klein with evidence of IRB approval to use the data. Data Files: - The dataset used in the manuscript is the merged deidentified subset ([rds][1], [csv][2]). This already excludes participants as per the pre-registration. - We also provide merged deidentified full ([rds][3], [csv][4]) which contains data from all participants with no exclusions for exploratory analyses. A description of the variables in the dataset may be found in the [Data Key][5] under the [Codebooks Component][6]. These data are released under a [CC0 license][7] into the public domain, and may be used without restriction. The following variables were removed from the full dataset to create the de-identified dataset: Summary of dropped variables: - All open-text responses (difficult to police) - Qualtrics collected variables like location and IP - Time - Potentially triangulating data, such as age, gender, political affiliation For the moment we have also deleted nearly all questions asked only at one site until we can vet them for confidentiality concerns. The full raw dataset is available in a private component and may be requested by emailing the corresponding author with appropriate ethics approval. This dataset contains raw data received from individual sites and may contain sensitive or identifying information about participants. Universities vary in their procedure, so please check with your ethics committee about this. Note that for most purposes, the deidentified (publicly available) dataset should suffice. [1]: https://osf.io/mec9g/ [2]: https://osf.io/d6qf2/ [3]: https://osf.io/92wrs/ [4]: https://osf.io/ck7xf/ [5]: https://osf.io/8zg5e/ [6]: https://osf.io/jd8xh/ [7]: https://wiki.creativecommons.org/wiki/CC0