A well-designed experiment copes with the potential effects of extraneous variables by using random assignment to experimental conditions and sometimes also by incorporating direct control and/or blocking into the design of the experiment. Each of these strategies—random assignment, direct control, and blocking—is described as follows; A researcher can directly control some extraneous variables. In the calculus test example, the textbook used is an extraneous variable because part of the differences in test results might be attributed to this variable.
We could control this variable directly, by requiring that all sections use the same textbook. Then any observed differences between temperature groups could not be explained by the use of different textbooks. The extraneous variable time of day might also be directly controlled in this way by having all sections meet at the same time. The goal is to design an experiment that will allow us to determine the effects of the explanatory variables on the chosen response variable.
Only $13.90 / page
To do this, we must take into consideration any extraneous variables that, although not of interest in the current study, might also affect the response variable.
The effects of some extraneous variables can be filtered out by a process known as blocking. Extraneous variables that are addressed through blocking are called blocking variables. Blocking creates groups (called blocks) that are similar with respect to blocking variables; then all treatments are tried in each block. In our example, we might use instructor as a blocking variable. If five instructors are each teaching two sections of calculus, we would make sure that for each instructor, one section was part of the 65° group and the other section was part of the 75° group.
With this design, if we see a difference in exam scores for the two temperature groups, the extraneous variable instructor can be ruled out as a possible explanation, because all five instructors’ students were present in each temperature group. (Had we controlled the instructor variable by choosing to have only one instructor that would be an example of direct control? Of course we can’t directly control both time of day and instructor. )
If one instructor taught all the 65° sections and another taught all the 75° sections, we would be unable to distinguish the effect of temperature from the effect of the instructor. In this situation, the two variables (temperature and instructor) are said to be confounded. If an extraneous variable is confounded with the explanatory variables (which define the treatments), it is not possible to draw an unambiguous conclusion about the effect of the treatment on the response. Both direct control and blocking are effective in ensuring that the controlled variables and blocking variables are not confounded with the variables that de fine the treatments. Two variables are confounded if their effects on the response variable cannot be distinguished from one another.