13.1 Basics

  • Classic and first use: Evaluation of scholarship programs (Thistlethwaite and Campbell 1960)
  • All units receive a score, and a treatment is assigned to those units whose score is above a known cutoff and withheld from those units whose score is below the cutoff
    • Q: How have you been selected for this MA program?
  • Score (variable), cutoff and treatment define a RD design
  • Sharp RD: All units comply with treatment condition they have been assigned
  • Fuzzy RD: Some units fail to receive treatment despite having a score above cutoff (or recieve it despite being assigned control, e.g. optional participation)
  • Q: Explain sharp and fuzzy RD in the case of a scholarship program with a certain grade (1.4) being the cutoff for the scholarship?
  • Q: Why is the RDD design (or relatedly the data behind it… think of the example above) so attractiv? Do we know a lot about the assignment process or not?
  • Q: Together with your neighbour come up with 3 real life examples where the RDD may be applicable?

References

Thistlethwaite, Donald L, and Donald T Campbell. 1960. “Regression-Discontinuity Analysis: An Alternative to the Ex Post Facto Experiment.” Journal of Educational Psychology 51 (6): 309.