RBUS2900 Study Guide - Final Guide: Factorial Experiment, Artificiality, Advantageous
Document Summary
Experiments provide the greatest amount of evidence for causality. Contamination from extraneous variables can be controlled more effectively than with other designs. Projecting from non-probability samples can pose problems regardless of random assignment. The convenience and cost of experimentation can be similar or cheaper than other methods. Business research is often concerned with the study of people there are limits to the procedures which are ethical. Steps in conducting an experiment: select relevant variables, the first step is to select explanatory (independent) and dependent variables for study. Experimental research and sampling: control the experimental setting, a primary concern is holding as many variables constant during the experiment, to avoid the introduction of sources of extraneous variables. Matching: when it is not possible to randomly assign subjects to groups, matching may be used. Within-subjects versus between-subjects designs: each subject receives every treatment, and the dv is measured for each condition and compared.
