The first issue is that the researcher does not discuss what happened to the 21 participants that did not complete the study. The researcher should discuss if the participants left for personal reasons or if it was due to side effects from the therapy technique, and if so, what side effects caused them to discontinue their participation in regards to the research study. This is known as the history effect because the dependent variables had to change because the number of subjects changed (Jackson, S., 2012). These changes could have been due to outside events. Attrition occurred because many of the subjects dropped out(Jackson, S., 2012). Attrition of the research subjects would cause an inequality between the experimental and control groups.
4. Internal validity occurs when experimental results can be accredited to the manipulation of the independent variable instead of a confounding variable. In other words, good internal validity actually measures what the researcher intends for it to measure (Jackson, C., 2012). Internal validity is important because if an experiment has good internal validity without any confounds, then there is only one explanation for the results of the experiment. External validity is occurs when experimental results can be generalized to other situations and subjects or behaviors correlate to those behaviors external to the study (Jackson, C., 2012). External validity helps determine if findings can be generalized to different persons, environments, measures, settings, or times.
6. Within-subject design uses the same subjects in each condition of the experiment. These subjects do not have to be randomly assigned to a group because the researcher(s) are repeatedly taking measures on the same group of people. Within-subjects design requires fewer participants and it requires less time to conduct experiments (Jackson, S., 2012). This design also increases statistical power because variability is decreased (Jackson, S., 2012). Within-subject design is beneficial in settings in which the participants within the study are diverse and when few subjects are available (Logan, Hickman, Harris & Heriza, 2008).
Matched-subject design uses different subjects instead of using the same subject/participant in all conditions; however, the design matches subjects on one or more variables. This design tries to maintain similar equivalences between groups, while assigning different experimental groups for each treatment or condition (Mahdjoubi & A-Rahman, 2012). Matched-subject design reduces demand characteristics and testing effects because different subjects are being used (Jackson, S., 2012). Since participants are being matched on variables of importance, statistical power increases as it does in the within-subject design (Jackson, S., 2012). However, it is harder to find participants for an experiment that match one another. Within-study design does not have to worry about this issue because fewer subjects are used...