Can you make perceptual maps from fake data? (A replication study)

See the blog post for more details.

Common files

• global.r — Loads libraries and functions used in the other scripts.
• openai.key — The OpenAI key used to access their API (not included in the repo).

Study on Super Rugby teams

• Datasets:
  • super-rugby-real-data.xlsx — The real data from the original study.
  • super-rugby-real-similarities.csv — The real similarity matrix from the original study. To be read of % of responses who said that [column] was more similar to [row].
  • super-rugby-facts-about-teams.csv — A simple dataset about a team each team containing the RGB colour of the team's kit, the team's name, and the team's country (coded as Australia or New Zealand).
  • The generated fake data is in the super-rugby-fake-data folder.
• Code:
  • super-rugby-fake-data-generation.r — The R script used to generate the fake data.
  • super-rugby-similarity-plotting.r — The R script used to plot the similarity charts (graphs and MDS).
• Output charts are in the super-rugby-outputs folder.

Study about Australian cafe chains

• Datasets:
  • cafe-real-data.xlsx — The real data from the original study.
  • cafe-real-similarities.csv — The real similarity matrix from the original study. To be read of % of responses who said that [column] was more similar to [row].
  • The generated fake data is in the cafe-fake-data folder.
  • cafe-expert-based-maps.csv — Two expert-based perceptual maps.
• Code:
  • cafe-fake-data-generation.r — The R script used to generate the fake data.
  • cafe-similarity-plotting.r — The R script used to plot the similarity charts (graphs and MDS).
• Output charts are in the cafe-outputs folder.