Chapter 6




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 occasions 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.


"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)


       Causal-comparative and experimental research are the final two forms of quantitative research we will examine. Both are aimed at making cause-effect statements about the performance or two or more groups, methods, or programs. The basic difference between the two is in the amount of control the researcher has over the comparisons studied. In experimental research, the alleged cause is under the control of the researcher and is manipulated by the researcher while in causal-comparative research it is not. The alleged cause, that is, the characteristic believed to make a difference, is often referred to as the treatment . The more formal name for the cause or treatment is the independent variable . It is the variable believed to cause a difference between groups. The difference, or effect , of  the independent variable is called the dependent variable because it is dependent on what happens to the independent variable. For example, in the statement “giving students assertiveness training will improve their self-confidence ,” the causal factor or independent variable is assertiveness training and the effect or dependent variable is self-confidence . What happens to self-confidence will depend on the success of assertiveness training. Try to identify the independent and dependent variables in this statement: “Careful study of the Gay and Airasian textbook will make readers high quality researchers.”   What is the cause and what is the effect? Note that experimental and causal-comparative research produce stronger relationships among variables than descriptive and correlational research because they link a cause to an effect.

      In a causal-comparative study the independent variable, or cause, has already occurred or cannot be manipulated, so  the researcher has no control over it. For this reason, causal-comparative research is also called ex post facto (after the fact) research. The independent variables in causal comparative studies either cannot be manipulated (e.g., gender, height,) or should not be manipulated (e.g., smoking, prenatal care). In causal-comparative research, at least two different groups are compared on some dependent variable or measurement of performance (the effect). For example, a causal-comparative study might involve the independent or causal variable “smoking,” with a comparison between a group of long-time smokers and a group of nonsmokers.   Causal-comparative and experimental research always involves the comparison of two or more groups or treatments. The dependent variable (the effect) might be the comparative frequency of lung cancer diagnoses in the two groups. In this example and in causal-comparative research in general, the researcher does not  have control over the independent variable. That is, the smokers and nonsmokers had already formed themselves into groups before the researcher began the study. The researcher has to select research participants from  two different, pre-existing groups. This creates a problem. Suppose that unknown to the researcher, a large number of the long-time smokers selected had lived in a smoggy, urban environment and that only a few of the nonsmoking group did. Due to the lack of control over the selection of study participants, attempts to draw cause-effect conclusions in the study would be at best tenuous and tentative. Is it smoking that causes higher rates of lung cancer? Is it living in a smoggy, urban environment? Is it some unknown combination of smoking and  environment? A clear cause-effect link cannot be obtained from this study because the researcher did not have complete control of the selection of the participant and their characteristics. 

       To conduct the smoking study as an experiment so that causal statements about smoking and lung cancer could be obtained would require the researcher’s control over  the selection of the participants for the two groups. To attain this control, the researcher would have to select a large group of participants who had never smoked and divide them into two groups, one forced to become heavy smokers and one forbidden to smoke. Obviously such a study would be unethical because of the potential harm to those forced to become heavy smokers. Thus, the only reasonable option is to conduct a causal-comparative study that approximates cause-effect results without  harming the participants. The results of causal-comparative studies sometimes lead to more rigorous experimental studies designed to confirm the causal-comparative findings. The following are examples of causal-comparative studies:

  1. The effect of preschool attendance on social maturity at the end of the first grade . The independent variable, or cause, is preschool attendance (students attending preschool and students not attending); the dependent variable, or effect, is social maturity at the end of the first grade. Two groups of first graders would be identified—one group who had attended preschool and one group who had not. The social maturity or the two groups would be compared at the end of grade one.
  2. The effect of having a working mother on school absenteeism . The independent variable is the employment status of the mother (the mother works or does not work); the dependent variable is absenteeism, or number of days absent. Two groups of students would be identified—one group who had working mothers and one  group who did not. The absenteeism of the two groups would be compared.
  3. the effect of gender on algebra achievement. The  independent variable is gender (male or female); the dependent variable is algebra achievement. The achievement of males would be compared to the achievement of females.


Causal-Comparative Research Designs

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 grade 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.


Control Procedure for Causal-Comparative Studies

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 subject, one from each group, that are similar to each other in verbal ability.

       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.


What are the Similarities and Differences between Casual-Comparative and Correlational Methods?

Causal-comparative and correlational methods (as defined in educational research textbooks) are similar in that both are nonexperimental methods because they lack manipulation of an independent variable which is under the control of the experimenter and random assignment of participants is not possible. This means, among other things, that the variables must be observed as they occur naturalistically. As a result, the key and omnipresent problem in nonexperimental research is that an observed relationship between an independent variable and a dependent variable may be spurious . That is, the relationship is not a causal relationship; it is a relationship that is the result of the operation of a third variable (see Figure 1). For an example of the third variable problem, note that self-reported "gender role identification" and high school "algebra performance" may be related. However, that relationship would probably largely be due to the joint influence of the third variable of “Gender-role socialization”.

       Because of the lack of manipulation of the independent variable and the problem of spuriousness, neither causal-comparative nor correlational research can provide as strong of evidence for causality as can a study based on a randomized experiment or a strong quasi-experimental design (such as the regression discontinuity design or the time series design). Indeed, even the more sophisticated theory testing or confirmatory approaches relying on structural equation modeling (which are "correlational") provide relatively weak evidence of causality (when based on nonexperimental data) as compared to the evidence obtained through randomized experiments.

       Causal-comparative and correlational studies are similar in that both are used to examine relationships among variables. The data from both of these approaches are typically analyzed using the general linear model (GLM), and it well known that all special cases of the GLM are correlational (e.g., Kerlinger, 1986; Tatasouka, 1993; Thompson, 1999) where the relations between variables are modeled. Given this, it is misleading to suggest, as is sometimes done in educational research texts, that only correlational research examines relationships.

       Causal-comparative and correlational studies are similar on the techniques available for controlling confounding variables. For example, one can statistically control for confounding variables in both approaches by collecting data on the key confounding extraneous variables and including those variables in the GLM. Likewise, one can eliminate the relationship between selected confounding and independent variables (regardless of their scaling) using matching or quota sampling approaches. Today, statistical control is usually preferred over individual matching (Rossi, Freeman, & Lipsey, 1999, Judd, Smith, & Kidder, 1991).

Wed, 11 May 2011 @15:22




Melawan Kemustahilan










Twitter Facebook Instagram Google Plus Youtube Channel




Copyright © 2019 bejo sutrisno · All Rights Reserved