Reichardt, C. S., & Henry, G. T. (2012). Regression-discontinuity designs. In H. Cooper, P. M. Camic, D. L. Long, A. T. Panter, D. Rindskopf, & K. J. Sher (Eds.), APA handbook of research methods in psychology, Vol. 2. Research designs: Quantitative, qualitative, neuropsychological, and biological (pp. 511-526). Washington, DC, US: American Psychological Association.
http://dx.doi.org/10.1037/13620-027
Regression-discontinuity (RD) designs are used to estimate the effects of treatments, programs, or interventions. The distinguishing feature of an RD design is that participants are assigned to treatment conditions on the basis of their scores on a quantitative variable. Participants with scores below a specified cutoff value on the quantitative variable are assigned to one treatment condition, whereas participants with scores above the cutoff value are assigned to no treatment or an alternative treatment. Effects of the treatments are evidenced by a discontinuity (at the cutoff score) in the relationship between the quantitative assignment variable and an outcome variable. (PsycINFO Database Record (c) 2016 APA, all rights reserved)