Session 08 - Open Data

We talk about how to publish open data and how to create a codebook.

Overview

Topic Duration Notes
Introduction Open Data 60 Slides
Wrap up homework 1: Recommendations RDM 20
Presentation: How to create a data dictionary 15 Slides
In class: Jointly create codebook for our data set 45 Google doc
Start the analysis / prepare the project folder structure ??

Optional: Create the codebook in DataWiz2

You can also use DataWiz2 for this (see the instructions in the video “- PTOS 9: Introduction to Research Data Management with D-Psy_FAIR”, starting at minute 29:27). In DataWiz2, you only need the sections “2. Upload and describe datasets” and then “5. Export”. You then upload our data set. When exporting, please select “Export as JSON” and in “Include datasets” select all datasets.

Homework (in groups)

  • Prepare the folder structure as outlined in the “Good coding practice” slides
  • Add our data dictionaries (both the human-readable xlsx or ods or pdf, and the machine-readable json file) to the /doc folder
  • Add the (preliminary) raw data files to the /raw_data folder
  • Add a README file
  • Turn the folder into a git repository and sync to Github (as data are anonymous, it can be public)
  • Create a script 01-preprocessing.R and start cleaning the data set. Things to do:
    • Remove obvious “zombie rows” (i.e., all-NA rows, or rows where participants dropped out at the second question)
    • Apply the inclusion/exclusion rules of our preregistration. See sections:
      • “M4 Participant recruitment, selection, and compensation”
      • “M5 How will participant drop-out be handled?”
      • “M7 Data cleaning and screening”
      • “AP1 Criteria for post-data collection exclusion of participants, if any”
      • “AP2 Criteria for post-data collection exclusions on trial level (if applicable)”
  • At the end of the script, you should have a cleaned data set which is ready for data analysis. Save the cleaned data frames as new .csv data files in /processed_data.