HSS 3101 Lecture Notes - Lecture 5: Univariate, Dependent And Independent Variables, Analysis Of Covariance
17 Feb 2016
School
Department
Course
Professor
Document Summary
Whether or not independent variable can be controlled: ex: students who come to class get better grades can"t control type of comparison made, between-subject, within-subjects, cross-sectional design, longitudinal design. Data collected at single point in time. Data collected 2+ times over an extended period of time trend study. Cross-sectional: measure year 1, 2, 3 in one sitting. Longitudinal: measure same people over years 1, 2, 3 follow-up study. Observance of independent variable: prospective or retrospective setting, laboratory = artificial setting. Manipulation: researcher does something to some subjects (introduces intervention or. Control: researcher introduces controls including a control group (must have minimum. Randomization: researcher assigns subjects to groups at random (experimental group or local standard of care) control group) Characteristics of experimentation: as much equivalence as possible among the two groups, to ensure equal groups: Random assignment: every member of population has equal chance of either. Random matching: matched on as many extraneous variables and matched.
