Experimental Psychology: Within-Subjects Approach

1.John was very excited about the within-subject approach. “Now I’ll never need to collect data for large numbers of subjects again.” He said. Do you think that John is correct? Explain your answer with support.
Generally, John is not correct. However, it is dependent on the nature of his study. A within-subject approach is generally used in cases where large samples are not available (Lecture 7, n.d.). However, it should not be used as a substitute when there is a large sample ready, but the researcher does not want to follow proper procedure. Within-subject designs are used when the same subjects will be subjected to different conditions or treatments (Lecture 7, n.d.). Since the small number of N also has its disadvantages, John should also consider the precision and reliability of results as required by his study. For example, if a single variable or condition will be tested within-subjects, a large number of participants might not be needed. But in cases where there are multiple conditions or treatments, such as in factorial experiments, a small N would not be desirable at all times due to the possibility of not being able to control extraneous variables as one could relatively so with much ease when considering only one condition/treatment. Because of this, John should not expect that within-subject approaches are always applicable to all experiments. The need to collect data for large numbers of subjects might be required for different types of experiments/studies.
2. Following the questions below to outline a small N experiment to test this hypothesis: People who eat veggies only for their diets are healthier.
a. What are the independent and dependent variables?
The identification of the independent variable and the dependent variable would depend on the nature of the experiment. Generally, the independent variable would be whether the participants are eating veggies only or if they eat other food groups like meat or dairy. Ideally, to determine a spectrum or identify a trend, it would be better if the experiment considers three or more independent variables such as “person who eats vegetables only,” “person who eats both vegetables and meat but eats more vegetables,” and “person who eats both vegetables and meat but eats more meat,” and “person who does not eat any vegetables.”
In either case, the dependent variable would be the measure of how healthy/healthier the person is and the independent variable is the consumption of veggies.
b. How would “healthier” be operationally defined?
In the experiment described in the last sub-question, “healthier” could be operationally defined depending on the nature of the study. For example, if the health outcome being measured/targeted is blood pressure, a “healthier” individual would have lower blood pressure. This is applicable to other measures of health as well.
c. What is the experimental procedure to test the hypothesis?
In this case, the experimental procedure to test the hypothesis will be a between-subjects multi-factorial design utilizing simple regression analysis. This design allows for the comparison of different treatments or conditions (Lecture 7, n.d.). This statistical analysis method will allow the researcher to view the effect of eating vegetables (and possibly the volume of vegetables consumed) on the dependent variable. For example, if the health outcome being measured is blood pressure, a regression analysis can determine if the hypothesis “People who eat veggies only for their diets are healthier” should be rejected or if it warrants a failure to reject decision.
d. What are the disadvantages of using a small N design for this experiment?
In this type of experiment, a small N would not properly reflect the general trend that is desired to be observed. Hence, the rel...