FRHD 3070 Lecture 8: Week 8 – Experimental Designs
19 Dec 2017
School
Department
Course
Professor
Document Summary
Descriptive non-experimental: observational method, case study method, survey method. Predictive non-experimental: correlational method, quasi-experimental method. Experimental control of all other variables: given the above, any difference on the dependent variable should be due to the effect of the independent variable, should try and reduce noise as much as possible. A pre-test helps ensure that groups are equivalent before the iv was introduced. With large enough groups, randomization usually ensures that groups are equivalent. The larger the sample, the less likely it is that the groups will differ in any systematic way. General rude is that you will probably need a minimum of 30 participants per condition to ensure no selection differences exist. Pretest might be necessary when: small sample size, pre-test might be necessary to select the participants. Lowest or highest alcohol consumption, once identified, then assigned randomly to experimental and control conditions: might want to measure the extend of the change in each individual, potential for drop out.
