Related concepts - X: cause, independent variable (IV), treatment - Y: effect, dependent variable (DV), outcome variable How to make causal inference?
Classification of Experimental Designs Definition of Symbols X- exposure of group to treatment O- process of observation or measurement of DV R- random assignment of groups or test units to separate treatment conditions EG- experimental group CG- control group SB- Selection Bias TE- treatment effect Pre-Experimental designs One-shot case study Design X O 1 - Single group exposed to treatment X - A single measurement is taken (O 1 ) - No R or CG
One-group pretest-posttest Design O 1 X O 2 - Group of units is measured twice (before and after treatment) - No CG - Treatment calculated as (O 2 - O 1 ) or diff before and after - Validity questioned because there is no randomisation Static Group Design EG: X O 1 CG: O 2 - Two group experimental design - EG exposed to treatment only - Measurement made after treatment - Not R - Treatment effort is O 1 – O 2 True experimental designs Post-test only control group EG: R X O 1 exposed to treatment then measured CG: R O 2 - TE= O 1 – O 2 - Randomly assigned to EG or CG - SB eliminated by R Pretest – Posttest Control Group Design EG: R O 1 X O 2 CG: R O 3 O 4 - Randomly assigned to CG or EG - Pre treatment + post treatment measure - TE= (O 2 – O 1 ) – (O 4 – O 3 ) -
You've reached the end of your free preview.
Want to read all 27 pages?
- Fall '17
- Erica Brady