SOAN 2120 Lecture Notes - Lecture 9: Dependent And Independent Variables, Internal Validity, External Validity
Document Summary
Test program in one community verses other communities as control groups : aside from the absence of random assignment, these designs look" much like a regular. Research designs: experiment: o1 x o2, o3 o4, not true experiments (quasi-experiments, internal validity issues compare o1 and o3 to see a selection bias problem. People selected themselves to go to university on their own, cannot be randomly assigned. So cannot determine whether university affects income: uncontrolled (confounding) variables personal attributes, x in this case is just a descriptor of other characteristics any differences between x and o are due to selection bias. Survey research: survey can be administered in a variety of different methods telephone, online, interview, questions are asked, which then represent variables: Intentions: standard response options must be given as answers to questions (variables) Mutually exclusive (each person fits into one category only) and exhaustive (everyone fits into a category)