Regression Discontinuity

Using a known mechanism that has cutoffs to determine your sample, and who is treated. Utilizes regression to the mean. Compare people below and above the mean because they’ll average out.

A simplified explanation of regression discontinuity:

#researchlandscape

Regression discontinuity is a statistical method used in research when subjects are assigned to different groups based on a certain cutoff point. This method is often used in fields such as economics, psychology, and social sciences.

The idea behind regression discontinuity is that those who are just above and just below the cutoff are likely to be very similar, except for the treatment they receive due to their group assignment. Therefore, any significant differences in outcomes between these two groups can be attributed to the treatment rather than other variables.

For example, let’s say we want to measure the impact of a scholarship program on student performance. We could use regression discontinuity by assigning the scholarship to students based on a certain GPA cutoff. Those who are just above the GPA cutoff would receive the scholarship (treatment group), while those just below would not (control group). Despite their similar GPAs, if there’s a significant difference in performance between these two groups, we can attribute it to the scholarship program.

This method helps researchers isolate the effects of a particular treatment or intervention while controlling for other variables that might influence the outcome. It’s like creating a quasi-experimental setting where one can observe causal relationships more clearly.

This ties into ungrading - what’s the difference, really between an 89% and a 90%. Just the letter grade, which really has an impact on GPA.