KINE 2049 Chapter Notes - Chapter 6: Demand Characteristics, Sphericity, Statistical Power
Document Summary
The two-level design is better because subjects can be randomly assigned to a group and differences between the groups can be attributed to the bicycling program. Two group design: control group weight and bicycling program weight. Experiments provide important advantages over simple observation for developing knowledge. Experimental control means that sources of extraneous variation that may affect the results are controlled or eliminated. It is more efficient to conduct one experiment with 2 or more independent variables than to conduct a series of experiments each with a single independent variable. Experiments with multiple independent variables also stimulate the real world better than experiments with a single independent variable. The most important reason experiments with 2 or more independent variables are conducted is to allow researchers to examine how independent variables act in combination. How independent combine their effects on the dependent variable is termed interaction. Disadvantages: as the number of independent variables increases, the complexity of the experiment increases.