Guest Post — Concrete Models for Educational Data Sharing
peter.suber's bookmarks 2023-12-05
"Deciding how to structure the data frames and other accoutrement in a manner that could be comprehended by others for use in secondary analysis or to reproduce our findings is a challenge, and we are still working to improve the model. It takes a lot of time and effort to create the structure, generate the documents, and review everything for accuracy and coherence. We aspire to begin using R Markdown or Quarto to automate some of the data visualization and reporting in the near future. Recent efforts also include development of Interpretation and Use Statements for various measures, and we might return to the published projects and add those at a later date.
Learning to share data has been a long and arduous process, but we remain steadfast in our commitment to data sharing. For the Schoen Lab, data sharing is a way to enhance transparency and magnify the potential legacy of the effort we put forth in primary data collection. It is exciting to know that we are contributing to the movement to enhance the transparency and value of educational research. We sincerely hope that the measures, the data, the replication code, and the model for sharing the data can be useful to other researchers. We encourage readers to examine the structures we have used to share data. We hope these examples can provide others with examples to follow (and improve upon). If you find the model (or the data) to be useful, or if you have suggestions for ways to improve the model, we will be happy to hear from you!"