SOCW 460 Lecture Notes - Lecture 3: Regression Analysis, Internal Validity, Random Number Table
Document Summary
Internal validity (causality) - results caused by independent variable: causality: a. i. History - events occur between pre&post test; not part of study b. ii. Maturation of subjects b. iv. testing - repeated testing allows participant to practice b. v. Practitioner attributes left out (ex. if inter) a. i. 3. Reactive effects of testing - pre testing makes participant more or less responsive than if had no exposure a. iii. O: one group pre-test / post-test (non-experiment, establishes correlation and time order, doesn"t account for factors other than x that might cause y, not able to rule out spurious causes. X: post test only with non equivalent groups (static group comparison design) (quasi-experiment, doesn"t control for selection bias, not able to rule out spurious causes. Random selection (sampling) - each element has an equal chance of selection independent of any other event in the selection process: population, external validity / generalizable, not r.