SLWK 380 Lecture Notes - Lecture 6: Internal Validity, Design Of Experiments, Selection Bias
Document Summary
Experimental design: a research design that attempts to provide maximum control for threats to internal validity in order to determine causality. Association of variables: variables have statistical relationship. Time order: independent variable occurs before dependent variable. Non-spuriousness: relationship between independent and dependent variables is actually caused by a third variable (or occurs randomly). The degree to which an effect observed in the dependent variable was actually caused by the independent variable and not other factors. Threats to internal validity are addressed in two ways. Something that happens outside the study causes changes to the dependent variable. A participant changes over time for reasons not related to the study. Perceived changes in the dependent variable are a result of the way participants respond to the test rather than actual changes in the variable. Measured used for the pretest and posttest are not comparable in some way. The experimental and control groups are not comparable in some way.