Experimental Research: Pretest-Posttest Design

Experimental research design is a type of quantitative research study which tests hypotheses to establish cause and effect relationships. In experimental research design, the researcher manipulates and controls the variables. The manipulated variable is called treatment or causal or experimental variable and while conducting research, the effects of treatment or experimental variable on the dependent variable are studied controlling the other extraneous variables. 

There are different types of experimental research designs such as true experimental designs, group designs and quasi-experimental designs. These designs have many sub-designs for conducting research studies which are applied while conducting research studies looking at the nature of the research problems (Gay & Airasian, 2003). For example, for the topic; "Effect of reading exercises on creative writing skills of secondary school students" the pretest-posttest control group design is appropriate to design where reading exercises can be taken as an independent variable or treatment while creative writing skills can be the dependent variable. 

The pretest-posttest control group design requires two groups such as the control group and the experimental group formed through random assignments (Gay & Airasian, 2003). The following steps can be followed in a pretest-posttest control group design.

Step 1: First of all, two groups of secondary school students may be formed through the random assignment and they can be named as control and experimental group.

Step 2:  Both the experimental and control groups may be administered pretest.

Step 3: Experimental group will receive the treatment as reading exercises given in the research problem. However, the control group will be addressed as routine work that may be different from the reading exercises.

Step 4: Both the experimental and control groups will be administered a posttest at the end of the study. Posttest scores will be compared to find out the effectiveness of the treatment.

 In pretest-posttest control group design, random assignment, pretest, and control group are meant to control the sources of internal validity. Random assignment controls here regression and selection factors, pretest controls mortality, randomization, and control group will control maturation while control group controls the effects of history, testing, and instrumentation. Here testing on reading exercises and their effects on creative writing skills are controlled. 

Because if pretesting leads to high scores in post-test then both treatment and control groups should benefit from pretesting. One weakness in this design is the interaction between the pre-test and the treatment. This means the results will be generalized to only pretested groups. How much this weakness makes a difference depends upon the nature of pre-test, treatment, and the time period during which treatment is given.

There are a variety of ways in this design that collected data is analyzed to test the hypotheses regarding the effectiveness of the reading exercises. The purpose of administering a pretest at the start of the study is to see the essential same level regarding creative writing skills. If the level of creative writing skills during the pretest is the same then the scores of the posttest are directly compared using a t-test. If both the groups are not on the same level of scores in the pretest, then the post-test scores can be analyzed using the analysis of covariance which adjusts the post-test scores for initial differences on any variable including pretest scores. This is an effective technique to find out the effects of reading exercises.

The variation of pre-test-post test control group design involves the random assignment of the same level of subjects in the treatment group for the purpose of more closely controlling one or more extraneous variables. However, any variable that can be controlled through matching can be effectively controlled through other techniques such as analysis of covariance. Another variation in pre-test-post test control group design is to administer more than one post-test. 

It will provide the effects of the reading exercises on students’ creative writing skills immediately after the treatment and its effects on the dependent variables at a later stage. It needs to keep in mind that the interaction between time of measurement and treatment may be a threat to external validity because there will be differences between the scores of both posttests. That means the effects of reading exercise on creative writing skills will be observed differently soon after the treatment and after a break from the treatment. It may not completely solve the problem but will minimize through providing information on the effects of treatment on the dependent variable in a subsequent test after the posttest. 

Pretest-posttest control group design is flexible to have extra control groups, which minimizes the influence of extraneous variables and the researcher can test whether the pretest itself has an effect on the subjects or treatment has affected the dependent variable. This design addresses many of the internal validity issues that can affect the generalizations of research findings. This design is supportive to the researcher to exert complete control over the variables and allows the researcher to check that the pretest did not influence the results.