However, the nonexperimental and quasi-experimental strategies define (i.e., not create) the groups with a nonmanipulated variable such as a preexisting participant characteristic (e.g., age or gender, or ses) or time (e.g., before/after a treatment, at age 10 and age 20), and here of course we do not assign participants to these sorts of groups.

Prompt:
1. Now it’s your turn to design this type of study. Keep it simple. Your “defined” groups should only have 2 levels that you will use for any sort of comparison…remember you can’t assign folks to these groups just like in the examples above, so use something that reflects a preexisting group. Just be sure to describe the variable you are going to use to define your groups, and its 2 levels (e.g., like the alcohol or education studies above). Next, indicate what you would measure and use to ultimately compare the groups (i.e., your DV).
2. And finally, identify and describe at least one issue you would need to be concerned about when trying to interpret any group differences you might find. That is, why won’t you be able to know for sure what led to the group differences you might find?
Description and example
In this chapter you are going to learn all about two particular types of research strategies: Quasi-Experiments and Non-Experiments. For right now, this is what you need to know: Experimental, non-experimental, and quasi-experimental research designs are similar in that all three strategies investigate relationships between variables by comparing groups of scores. There are also important differences among them. As you already know, the experimental strategy creates the groups by manipulating an independent variable (to which we ASSIGN participants). However, the nonexperimental and quasi-experimental strategies define (i.e., not create) the groups with a nonmanipulated variable such as a preexisting participant characteristic (e.g., age or gender, or SES) or time (e.g., before/after a treatment, at age 10 and age 20), and here of course we do NOT assign participants to these sorts of groups.
Why do these type of studies? While experiments offer the most control over extraneous variables from becoming confounds, in the “real world” researchers often want to study phenomenon where random assignment or controlling the independent variable just isn’t possible. Chapter 10 is about research strategies and designs that are similar to experimental designs but fail to meet the strict standards of a “true” experiment (and you know what these are right? 1. manipulation of the IV, meaning we ASSIGN persons to groups that we create, and 2. control of extraneous variables from becoming confounds). Nonexperimental and quasi-experimental designs fall short of this standard. Sometimes the independent variable is measured but not manipulated by the researcher, for example if we want to study stress levels in people who were exposed to a tornado compared to people who were not, the researcher doesn’t control who gets the tornado (how weird would that be!). Sometimes the independent variable is a demographic characteristic like gender, or ethnicity, or education, SES level that can’t be manipulated (i.e., obviously we can’t assign persons to these sort of variables, but we might be curious if such groups are different)…let’s walk through some more examples…
For example, to perform an educational experiment, a class might be arbitrarily divided by alphabetical selection (one group is last names from A-M, the other group from N-Z) or by seating arrangement (the front three rows of students vs. the back three rows of students) and compare their GPAs. The division is often convenient and, especially in an educational setting, causes as little disruption as possible. Let’s look at another example: if we study the effect of maternal alcohol use when the mother is pregnant, we know that alcohol does harm embryos/fetuses. A strict experimental design would require that mothers get randomly assigned to drink different amounts of alcohol. This would be highly unethical because of the harm to the embryo/fetus, and the mother too of course…so, instead we could do a quasi or a nonexperiment, where researchers ask mothers how much alcohol they used during their pregnancy, and use this information about alcohol consumption to define groups that we can then compare (e.g., those who drank an average of 6 alcoholic beverages per week vs. those who drank an average of 2 per week). And one more example, as students of online learning, you might be curious if there are differences in the average course grade between students who took this course online vs. those who took it face-to-face (get the idea? you can’t assign students to either class format, but you can certainly compare them…but you will likely wonder what any difference you find between the formats might also be due to…).
Here’s the downside of these studies: they do not take into account any preexisting factors regarding the people in our groups, nor do they recognize factors outside the experiment that influence the results in the study…for example, in the alcohol study of mothers, issues here could be what made them drink or not drink, other drugs they may use in addition to alcohol, or preexisting health conditions or number of previous pregnancies…in the educational study, one group of children defined by rows, those in the “front rows” may have been slightly more intelligent or extroverted than those in the back rows; in the online vs. face to face class study, online students might be more motivated, or perhaps not!
So, we can compare our groups on something we can measure (i.e., a DV of sorts), but we can’t be sure what led to any group differences that we find…but this is what we have to live with in these sorts of studies….again, we can’t always control or manipulate our variables.

Do you need any assistance with this question?
Send us your paper details now
We’ll find the best professional writer for you!