Application: Quasi-Experimental Designs
In the social sciences, researchers often want to conduct studies in situations where they cannot control certain aspects of the study. For example, they might want to compare critical thinking skills for students taking an online class with those taking a face-to-face class. The independent variable is the type of class, and the researcher cannot manipulate it. One group enrolled in an online class by choice and another group enrolled in a face-to-face class by choice. Therefore, there is no random assignment. Additionally, the students are in class for 9 weeks, and the researcher cannot control for any critical thinking skills students might develop outside of class during that time. These challenges cause threats to internal validity.
It would be disappointing to give up on studying an important topic just because the researcher cannot exert control over all of the variables in the study. Quasi-experimental research designs make it possible to conduct such a study. A well-designed quasi-experimental study rules out some of the threats to internal validity and provides a way for the study to still offer valuable information.
In this week's Assignment you apply concepts of quasi-experimental designs, identify the independent variables in a quasi-experimental design, and recommend alternative design for a study.
To prepare:
• You may also review and research materials (hard copies or online) outside this week's Learning Resources that are relevant to the
Assignment, if you wish.
• Read the three study descriptions below:
1. Study 1 utilized a participant-variable design and demonstrated that women perform more poorly, on average, than men perform on the mathematics section of the Scholastic Aptitude Test (SAT).
2. Study 2 utilized a comparison-group before-after design and demonstrated that students from lower socioeconomic status (SES) households benefit more over time from an academic tutoring program than do students from higher SES households.
3. Study 3 utilized a cross-sectional design and measured the IQ of individuals who are in their twenties. In Group 1 are 20-year-olds, Group 2 are 24-year-olds, and Group 3 are 29-year-olds. The mean IQ for the three groups is 103, 100, and 90, respectively. Consider whether the researcher can accurately conclude that people lose intelligence as they age.
• Select any two of the studies for this Assignment.
The Assignment
• Identify the two studies you selected for this Assignment by their corresponding numbers.
• Identify the dependent variable and independent variable in each study.
• Explain what is problematic about each of the studies you selected.
• Briefly describe at least one other type of quasi-experimental research design that might have been used for each study to gain additional or different information. Justify your rationale behind selecting that design.