Sveriges mest populära poddar

Nullius in Verba

Episode 19: Quantifauxcation

79 min • 20 oktober 2023

In this episode, we discuss Quantifauxcation, described by statistician Philip Stark as “situations in which a number is, in effect, made up, and then is given credence merely because it is quantitative.” We give examples of quantifauxcation in psychology, including errors of the third kind. We spend the second half of the podcast discussing how to develop quantitative measures that are meaningful and bridge the divide between qualitative and quantitative observations.

 

Shownotes

  • Burgess, E. W. (1927). Statistics and case studies as methods of sociological research, Vol 12(3), 103-120. (Thanks to Andy Grieve!)
  • Blog by Tania Lombrozo on nonsensical formulas in abstracts.
  • Type III errors: Philip Stark’s post of Deborah Mayo’s blog
  • Wilson, M. (2023). Constructing measures: An item response modeling approach. Taylor & Francis.
  • Wilson, M., Bathia, S., Morell, L., Gochyyev, P., Koo, B. W., & Smith, R. (2022). Seeking a better balance between efficiency and interpretability: Comparing the likert response format with the Guttman response format. Psychological Methods. Advance online publication. https://doi.org/10.1037/met0000462
  • Bhatti, H.A., Mehta, S., McNeil, R., Wilson, M. (2023). A scientific approach to assessment: Rasch measurement and the four building blocks. In X. Liu & W. Boone (Eds.), Advances in Applications of Rash Measurement in Science Education. Springer Nature. 

 

Kategorier
Förekommer på
00:00 -00:00