The Nonequivalent Groups Design
The Basic Design N O X O N O O Key Feature: Nonequivalent assignment
What Does Nonequivalent Mean? Assignment is nonrandom. Researcher didn’t control assignment. Groups may be different. Group differences may affect outcomes.
Internal Validity History Maturation Testing Instrumentation N O X O N O O History Maturation Testing Instrumentation Regression to the mean Selection Mortality Diffusion or imitation Compensatory equalization Compensatory rivalry Resentful demoralization
Internal Validity Selection-history Selection-maturation N O X O N O O Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality Statistical Analysis of Basic two-group pre-post NEGD cannot use the usual ANCOVA regression model because of measurement error on the pretest which leads to the attenuation of slopes and introduces bias have to do a reliability-corrected ANCOVA model adjusts the pretest scores based on estimates of reliability of the pretest
The Bivariate Distribution 8 7 6 5 4 3 9 P r e t s o
The Bivariate Distribution 8 7 6 5 4 3 9 p r e t s P o Program Group has a 5-point pretest advantage.
The Bivariate Distribution Program group scores 15-points higher on Posttest. 8 7 6 5 4 3 9 p r e t s P o Program group has a 5-point pretest advantage,
Graph of Means pretest posttest pretest posttest MEAN MEAN STD DEV STD DEV Comp 49.991 50.008 6.985 7.549 Prog 54.513 64.121 7.037 7.381 ALL 52.252 57.064 7.360 10.272
Possible Outcome #1 Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality (CG not growing) (PG moving away, CG level) More low-score PG dropouts
Possible Outcome #2 Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality (Both growing) (Wrong direction) More low-score dropouts
Possible Outcome #3 (In PG only) (In PG) Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality (In PG only) (In PG) More high-score PG dropouts not as likely
Possible Outcome #4 Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality (In PG only) (In PG) More low-score PG dropouts
Possible Outcome #5 Selection-history Selection-maturation Selection-testing Selection-instrumentation Selection-regression Selection-mortality