Abstract
The paper evaluates how a pre-test post-test quasi-experimental design was used to undertake a quantitative study and the various types of validity that are important in this design. From the evaluation and discussion, it is evident that quasi-experimental designs can be effectively used in situations where it is neither practical nor feasible to assign participants randomly to treatments.
Introduction
This paper reviews the article titled “Walk Texas! 5-A-Day Intervention for Women, Infant, and Children (WIC) Clients”, with the view to evaluating the choice of design used and discussing some issues related to the validity of the design.
Evaluation of Design Choice
Bartholomew, Miller, Ciccolo, Atwood, and Gottlieb (2008) chose a pre-test post-test quasi-experimental design to determine whether a brief stage-based counseling guide (Walk Texas! Clinical Counseling Guide for Nutrition) was adequate in elevating the consumption of fruits and vegetables within a clinical setting consisting of women, infants, and children. Available literature demonstrates that quasi-experimental designs are not true-experimental designs since the researcher does not have any control over the relationships by virtue of the fact that he is unable to control the causal variable of interest (Creswell, 2009; Frankfort-Nachmias, Nachmias, & DeWaard, 2014). As such, researchers had to control other variables to associate the counseling guide to an increase in consumption in the experimental group because this design lacks control over assignment of participants to the independent variable.
Rationale for Design Choice
The researchers’ choice of quasi-experimental design was fueled by the fact that they intended to have two intervention and two comparison clinics in the study settings that could be matched for size and ethnicity, hence it was neither practical nor feasible to assign participants randomly to treatments (Frankfort-Nachmias et al., 2014). The design enabled the researchers to conclude that the increased intake of fruits and vegetables noted after intervention and consequent follow-ups could be as a result of the counseling Intervention.
Types of Validity and Critical Differences
The most important types of validity in a quasi-experimental research design include “internal validity, external validity, statistical conclusion validity, as well as construct validity” (Grimshaw, Campbell, Eccles, & Steen, 2000, p. 14). Internal validity questions whether the observed outcomes are as a result of effects of some other variables on the dependent variable(s), while external validity questions whether the observed outcomes can be found in other contexts or are specific to the experimental setting and participants. Statistical conclusion validity relates to the degree to which conclusions about observed variables relationships are correct or reasonable based on the corrected data, while construct validity relates to the degree to which a test measures what it claims to be measuring (Shadish, Cook, & Campbell, 2005).
Performance of Authors
These types of validity are not well explained by the researchers in the study, though there is an attempt to show how participants were randomly assigned to intervention and control groups and also how the intervention was designed to direct counseling through a number of steps (Bartholomew et al., 2008). The researchers also describe how different tests measured the variables of interest (e.g., stage of change, fruits and vegetables consumption, self-efficacy, attitude, and perceived barriers).
Assessing the Study’s Validity and Information Required
The study’s validity can be evaluated by examining and controlling the factors that have the potential to bias the findings. As such, information about the history (events other than the treatments that may occur during the experimental period), maturation (physical or psychological shifts taking place within the subjects), testing (pre-test and post-test assessments), instrumentation (testing instruments or testing conditions), selection (subjects’ abilities and characteristics), and experimental mortality (loss of subjects during the study) is required to evaluate the study’s validity (Frankfort-Nachmias et al., 2014; Shadish et al., 2005).
Conclusion
This paper has evaluated how a pre-test post-test quasi-experimental design was used to undertake a study and the various types of validity that are present in this type of design. From the discussion, it is clear that this type of design can be effectively used in situations where it is not practical or feasible to assign participants randomly to treatments.
References
Bartholomew, J.B., Miller, B.M., Ciccolo, J.T., Atwood, R., & Gottlieb, N.H. (2008). Walk Texas! 5-a-day intervention for women, infant, and children (WIC) clients: A quasi-experimental study. Journal of Community Health, 33(2), 297-303.
Creswell, J.W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Thousand Oaks, CA: Sage Publications Inc.
Frankfort-Nachmias, C., Nachmias, D., & DeWaard, J. (2014). Research methods in the social sciences (8th ed.). New York: Worth Publishers.
Grimshaw, J., Campbell, M., Eccles, M., & Steen, N. (2000). Experimental and quasi-experimental designs for evaluating guideline implementation strategies. Family Practice, 17(1), 11-16.
Shadish, W.R., Cook, T.D., & Campbell, D.T. (2005). Experimental and quasi-experimental designs for generalized causal inference. New York: Houghton Mifflin Company.