Experimental Manipulation

Qualitative

Qualitative variables represent experimental manipulations that differ in kind or type. With qualitative experimental manipulations, participants are randomly assigned to specific research conditions, or treatment and control groups that vary in characteristics. An example of a qualitative variable in an experimental communication research project might be to randomly assign participants to one of two groups to examine the effectiveness of a diversity training protocol. The first group is the control group. Participants in this group receive no manipulation. The second group is the experimental group. This group receives the experimental manipulation (e.g., diversity training, or other type of communication training). The researchers in this example would then be able to compare whether or not the group that received the training is significantly different from the control group.

Quantitative

Quantitative variables represent manipulation of the levels or amounts of the IV. With quantitative experimental manipulations, participants are randomly assigned to a range or degree of exposure to the IV. An example of a quantitative variable in an experimental communication research project might be to randomly assign participants to varying degrees of exposure to diversity training (or other types of communication training). Some participants receive no training (the control group), another group of participants receives a single hour of training, another receives a full 8-hour-day of training, and the final group receives a full 40-hour-week of training. The researchers in this example would then be able to explore whether or not increased time in training has increased levels of success.

Classification

Classification variables group research participants by characteristics that are already present in the participants prior to the start of the study. [Page 475]This type of variable organizes participants based on their membership in naturally occurring groups. Examples of classification variables in communication research might include gender, whether or not a student has studied abroad in the past, or enrollment in an online section of a public speaking course compared with a face-to-face section of the same course. It is important to note that classification variables are not part of true experimental designs as the requirements of random assignment and control are not present. Instead, classification variables are used within quasi-experimental designs. With classification variables, one is able to use statistical tests to determine whether or not there are differences between groups, but without random assignment to these groups or control of extraneous variables, researchers cannot be certain that the IV is the cause of a change in the DV, or if an external factor is partially or fully influencing the DV.

...