Research is broadly divided into two categories: qualitative and quantitative. Within each of those categories are multiple designs and methods. Among the quantitative methodologies are experimental designs.

Experimental designs are those in which a controlled experiment is carried out in order to determine if there exist statistical similarities within or between groups of subjects. The process requires a pre-test and a post-test of the subjects, and the application of a treatment or condition change.  A control group must also be established to make a statistical comparison that helps determine the effectiveness of the treatment or condition applied.

Experiments can be carried out using independent samples. The independent samples design compares the reactions of two or more groups to a treatment or condition and uses statistical models such as the independent-sample t-test, the one-way analysis of variance (ANOVA), the chi-square test, and non-parametric statistical tests. For example, this design addresses a research question regarding whether a new classroom methodology is effective for students or not. Subjects are tested to record a statistical baseline. The condition (the new methodology) is applied to one group of students but not the other. Subjects are post-tested, the results are compared, and the statistical similarity or difference between the groups is noted.

Experiments can also be carried out using within-subjects designs. The within-subjects design allows consideration of two or more conditions as applied to the same group of subjects; the results of the conditions are statistically analyzed. This design uses statistical models such as the one-sample t-test, the repeated-measures analysis of variance (ANOVA), and the paired-sampling t-test. In this design, for example, the research question might be focused on whether one new classroom methodology is more effective than the other. Subjects are tested to record a statistical baseline, and then each of the new methodologies is independently applied to the same group. Subjects are post-tested after experiencing each of the new conditions. The statistical similarity or difference between the results of the conditions applied is noted.

Experimental design is but one form of research design that is suitable and meaningful for dissertation research. Additional designs will be reviewed later in this series. In all cases, learners are advised to have relevant conversations with their dissertation committee and their methodologist and chair in particular to determine the best design to address the research questions of the proposed dissertation study.