Lesson 12 will consist of the following topics
For lesson 12, read pages 337-354 in Educational Research: Competencies for Analysis and Application, Seventh Edition by L. R. Gay and Peter Airasian (2003, Merrill/Prentice Hall, ISBN 0-13-099463-4)
or read pages 216-224 in Educational Research: Competencies for Analysis and Application, Eighth Edition by Lorrie Gay, Geoffrey Mills, and Peter Airasian (2006, Prentice-Hall, ISBN 0-13-118534-9)
Causal-comparative educational research attempts to identify a causative relationship between an independent variable and a dependent variable. However, this relationship is more suggestive than proven as the researcher does not have complete control over the independent variable. If the researcher had control over the independent variable, then the research would be classified as true experimental research.
There are many occassions while conducting educational research during which the researcher is unable to control the independent variable, or it would be unethical to control the independent variable, or it is simply too difficult to control the independent variable. For example, someone attempting to find the effect of gender (male or female) on an educational dependent variable would not be able to experimentally manipulate gender, and thus would use a causal-comparative research design rather than an experimental design for the research.
The statement or title of a causal-comparative research study takes the form: The effect of [independent variable] on [dependent variable] for [subjects], with the understanding that the independent variable is not under experimental control.
An example would be "The effect of gender on a visual alertness measure for 6th grade public school pupils." Here the independent variable is gender and the dependent variable is the visual alertness task.
Another example of a causal-comparative research design would be "Classroom behavior of good and poor readers." In this study the independent variable would be good readers versus poor readers. The researchers identified the three students with the highest scores and three students with the lowest scores on a reading achievement test for each of 18 classrooms. The dependent variable, classroom behavior, was measured by having a trained observer, measure the frequency of seven classroom behaviors during a 30 minute period of time over ten days. The observer did not know which of the pupils she was observing were good readers and which were poor readers. The complete reference for this study is:
Wasson, B., Beare, P., and Wasson, J. B. (1990). Classroom behavior of good and poor readers. Journal of Educational Research, 83, 162-165.
"An important difference between causal-comparative and correlational research is that causal-comparative studies involve two or more groups and one independent variable, while correlational studies involve two or more variables and one group." (Gay & Airasian, 2000, 364).
The basic design of a causal-comparative research study is to select a group that has the independent variable (the experimental group) and then select another group of subjects that does not have the independent variable (the control or comparison group). The two groups are then compared on the dependent variable. For example, in your junior high school, some of the seventh grade math classes use hand held calculators in their seventh grade mathematics classes, while other classes do not use calculators. You want to find the effect of calculator use on mathematics grades at the end of the year. So you select a group of students from the classes that use calculators and then select another group of the same size from the classes that do not use calculators and compare the two groups at the end of the year on their final math grades. Another variant of this study would be to take the students from one class that uses calculators and compare them with another class that does not use calculators. Both these studies would be causal-comparative research studies but they would differ in how you can generalize the results of your study. The results of the first study could be generalized to the seventh grade students taking mathematics classes in your school while the second study could only be generalized to the two classes that participated in the study.
Instead of using an experimental group and a control group as in the study considered above, you could have a causal-comparative research study in which two or more groups differ in some variable that constitutes the independent variable for the study. For example a study might wish to compare students at four different age levels (or grade levels) on their amount of participation in extra-curricular activities. The researcher could look at the number of extra-curricular activities participated in by four groups of students. The first group would be students in grades 1-3, the second group students in grades 4-6, the third group students in grades 7-9, and the fourth group students in grades 10-12. The independent variable in this study would be grade placement and the dependent variable would be participation in extra-curricular activities (the effect of grade level on participation in extra-curricular activities for public school students grades 1-12).
One of the problems with causal-comparative research is that since the pupils are not randomly placed in the groups, the groups can differ on other variables that may have an effect on the dependent variable. In experimental research we can assume that these other variables cancel out among the study groups by the process of randomization. However, in causal-comparative research if we are suspicious that some external variable might be involved, we can use some control procedure in an attempt to ameliorate the effect of the external variable.
Matching is one way to help control the effect of extraneous variables on the dependent variable in a causal-comparative study. For example, let's say you do not think that the two groups that you are using to evaluate a new approach to reading instruction are similar on verbal ability. Further you suspect that verbal ability might be related to the dependent variable in this study. The dependent variable is performance on a reading test. To overcome this difficulty you assess each of your students with a measure of verbal ability, such as a test of general intelligence, and then select pairs of subjects, one from each group, that are similar to each other in verbal abilily.
In this causal-comparative study, the independent variable is method for reading instruction. The dependent variable, is reading proficiency, and verbal ability is the matching or control variable.
Another method to control the effect of an extraneous variable on the dependent variable is to compare homogeneous subgroups. We could do this in the previously mentioned study by restricting our subject selection from each of the groups, to those with tested IQ's in the range 90-110. This would constrict the number of subjects we could use in our study, but would help control the effect of tested intelligence (verbal ability) on the dependent variable in the study.
A third method that is sometimes used to control the effects of an extraneous variable is analysis of covariance. Analysis of covariance is a statistical method of control in which the scores on the dependent variable are adjusted for the subject's initial differences in the control variable.
An inferential statistic used in both causal-comparative and experimental research designs is the t-test. Where the subjects in the two groups are independent of one another, that is no matching of subjects or other control procedures were used, the independent t-test is used to test the significance of a difference between the means of the experimental and control groups in the study. You can find additional information on the calculation and interpretation of the independent t-test online in Ed 602 Lesson 11.
In research designs where the influence of an extaraneous variable has been controlled, or in designs utilizing a pre-test-post-test procedure, the appropriate t-test to use to compare the two groups would be the dependent t-test. You can find additional information on the calculation and interpretation of the dependent t-test online in Ed 602 Lesson 12.
When you have three or more groups to compare, the appropriate inferential statistic to use would be one-way analysis of variance. This statistic shows the significance of differences in the means of three or more groups of subjects. You can find additional information on the calculation and interpretation of analysis of variance online in Ed 602 Lesson 13.
In cases where you are using frequency counts for the dependent variable, the appropriate inferential statistic to use would be the chi-square test. This statistic tests the significance of differences between two or more groups (independent variable) in frequencies for the dependent variable. For example, a high school social studies teacher wants to see if the major party political affiliation for students is similar to or different than that of the registered voters in the county where his high school is located. The teacher would ask the students (anonymously) to indicate whether they would support the democratic party or the republican party. The proportion of students selecting the democratic or republican parties would be compared with the county proportions of democratic and republican voters using the chi-square statistic. You can find additional information on the calculation and interpretation of the chi-square test online in Ed 602 Lesson 14.
Click on Lesson 12 Assignment to go to the assignment for Lesson 12.
Click on Lesson 12 Quiz to go to the Quiz for Lesson 12.
Please send electronic mail to the course instructor if you have any questions about this lesson or other concerns.