POL SCI 3 Chapter Notes - Chapter 4: Classification Of Discontinuities, Regression Analysis, Randomized Experiment
12 Oct 2016
School
Department
Course
Professor
Document Summary
Rd design is a pretest posttest program comparison group strategy. Comparable in internal validity to conclusions from randomized experiments. From an administrative viewpoint, rd design often directly usable with existing measurement efforts such as regularly collected statistical information typical of most management information systems. Does not require that the pre and post measurements be the same. Cutoff value on the preprogram measure is used to assign people or other. Program any program, treatment or manipulation whose effects we wish to examine. Wish to try a new protocol on patients considered most ill. Assume a pretest cutoff of 50 (1-100, higher healther) as the assignment criterion: below 50 given treatment protocol and 50+ given the standard treatment. If we assume the treatment has a constant effect which raise each treated. We observe a jump/ discontinuity in the regression lines at the cutoff point. Assignment is based on a cutoff value on a preprogram measure.
