CRIM 220 Chapter Notes - Chapter 13: Scientific Realism, Random Assignment, Quasi

• School: Simon Fraser University
• Department: Criminology
• Course: CRIM 220
• Professor: Ted Palys

Document Summary

Topics appropriate for evaluation research and problem analysis. Learning policy goals is a key first step in doing evaluation research. Evaluability assessment: sort of (cid:862)pre e(cid:448)aluatio(cid:374)(cid:863) i(cid:374) (cid:449)hi(cid:272)h resear(cid:272)her deter(cid:373)i(cid:374)es (cid:449)hether (cid:272)o(cid:374)ditio(cid:374)s (cid:374)e(cid:272)essar(cid:455) for (cid:272)o(cid:374)du(cid:272)ti(cid:374)g an evaluation are present. Problem formulation to conduct evaluation research we must be able to operationalize, observe and recognize the presence or absence of what is under study: process normally begins by identifying and specifying program goals. Measurement: after we identify program goals, our attn. turns to measurement, considering first how to measure a progra(cid:373)"s su(cid:272)(cid:272)ess i(cid:374) (cid:373)eeti(cid:374)g goals. If a program pursues multiple goals, researchers may have to either divide multiple outcome measures or select a subset of possible measures to correspond with a subset of goals. Designs used in basic research are readily adapted for use in evaluation research. Randomized evaluation designs randomization is the great equalizer: through probability theory, we can assume that groups created by random assignment will be statistically equivalent.