PSY 250 Chapter 4: PSY250ch4notes.doc
Document Summary
Low internal validity= little or no evidence of causality: threats to internal validity, compromise our confidence in saying that a relationship ends between the variables of interest, experimenter bias, researcher treating conditions differently due to hypothesis. Those who drop out of your study could be qualitatively different than: mortality, controlling threats those who remain, use several repetitions of the test for within subject variables, random assignment should spread regression toward the mean across the different groups so that it does not cause group differences. Procedure: administer the new test, administer the established test, compute correlation, context, does the measure include a representative sample of the behaviors that are relevant to the variable, construct, measures what it was designed to measure. Problems with test retest reliability: overestimate reliability if remember how respond first time, alternate form on second test, underestimate reliability if measured trait changes between, split half reliability administrations, procedure.
