Experimental Designs

All experimental designs have one central characteristic: they are based on manipulating the independent variable and measuring the effect on the dependent variable. Experimental designs result in inferences drawn from the data that explain the relationships between the variables.  If your study is at Level III, an experimental design is best. “Why” questions are always experimental design.

The classic experimental design consists of the experimental group and the control group.  In the experimental group the independent variable is manipulated.  In the control the dependent variable is measured when no alteration has been made on the independent variable.  The dependent variable is measured in the experimental group the same way, and at the same time, as in the control group.  The prediction is that the dependent variable in the experimental group will change in a specific way and that the dependent variable in the control group will not change.

-	Independent Variable	Dependent Variable
Experimental Group	Changed	Measured
Control Group	Unchanged	Measured

For the classic experimental design, then, there are two groups (experimental and control), two variables (independent and dependent), and one measurement of the dependent variable.  When data are analyzed, only the measurements on the dependent variable for the two groups are contrasted.

Controlling Unwanted Influences

To obtain a reliable answer to the research question, the design should eliminate unwanted influences.  The amount of control that the researcher has over the variables being studied varies, from very little in exploratory studies to a great deal in experimental design, but the limitations on control must be addressed in any research proposal.

These unwanted influences stem from one or more of the following: extraneous variables, bias, the Hawthorne effect, and the passage of time.  These four will be discussed in turn, along with some suggestions for controlling their effects.  You will need to identify those that seem relevant to your question and show how you will control their effects.

Extraneous Variables

Extraneous variables are variables that can interfere with the action of the ones you are studying.  They could as easily have been chosen as independent variables had you been interested in them because of their known effect on your variables.  Since they are not part of your study, their influence must be controlled.

In the research literature, you will see extraneous variables also referred to as intervening variables, found in Level III experimental designs, directly affect the action of the independent variable on the dependent variables are those that occur in the study setting.  They include economic, physical, and psychological variables. 

Extraneous variables usually are not a problem in Level I studies.  Since you are studying only one variable in depth, these variables are assumed to be independent.  They could be extraneous variables, as well.  That is why an exploratory descriptive study explores in depth:  all variables must be accounted for and taken into consideration in order to be described adequately.

At Level II, when you do not know which variables are independent and which are dependent, you must assume that there may be variables that have not been accounted for that may be related dependent variable.  Some variables have been shown to be related to one of your variables in other studies.  These are known extraneous variables and must be considered when designing your study, so that their effect your variables can be controlled.  At Level II, it is sometimes difficult to establish which one is the independent variable, even after confirming that there is a significant relationship between two variables.  Unless you can demonstrate that one variable precedes the other in time, it may be impossible to determine which the independent variable is.  Therefore, it is important to control extraneous variables when you can identify what they are, so that you can isolate the relationship between the ones you are interested in studying.

At Level III, where you are predicting the relationship, your must be very careful to control all possible extraneous variables that might intervene in your test.  These variables can be identified from the literature on your topic.

Methods of controlling extraneous variables include:

• randomization
• homogeneous sampling techniques
• matching
• building the variables into the design.

Randomization

Theoretically, randomization is the only method of controlling all possible extraneous variables.  The random assignment of subjects to the various treatment and control groups means that the groups can be considered statistically equal in all ways at the beginning of the experiment.  It does not mean that they actually are equal for all variables. 

However, the probability of their being equal is greater than the probability of their not being equal, if the random assignment was carried out properly.  The exception lies with small groups where random assignment could result in unequal distribution of crucial variables.  If this possibility exists in your study, perhaps one the other method would be more appropriate.  In most instances, however, randomization is the best method of controlling extraneous variables.

The principle of randomization applied to both Level II and Level III studies.  In Level II studies, a random sampling technique results in a normal distribution of extraneous variables in the sample, this approximates the distribution of those variables in the population.  The purpose of randomization at Level II is to ensure a representative sample.

At Level III, randomization comes into play when you randomly assign subjects to experimental and control groups, thus ensuring that the groups are as equivalent as possible prior to the manipulation of the independent variable.  Random assignment assures that the researcher is unbiased.  Instead, assignment is predetermined for each subject.

Homogeneous Sample

One simple and effective way of controlling an extraneous variable is not to allow it to vary.  Choose a sample that is homogenous for that variable. 

Matching

When randomization is not possible, or when the experimental groups are too small and contain some crucial variables, subjects can be matched for those variables.  The experimenter chooses subjects who match each other for the specified variables.  One of these matched subjects is assigned to the control group and the other to the experimental group, thus ensuring the equality of the groups at the outset.

The process of matching is time consuming and introduces considerable subjectivity into sample selection.  Therefore, it should be avoided whenever possible. If you use matching, limit the number of groups to be matched and keep the number of variables for which the subjects are matched low.  Matching with more than five variables becomes extremely cumbersome, and it is almost impossible to find enough matched partners for your sample.  Matching may be used in all research designs (besides Level III) when you are looking at certain outcomes and want to have as much control as possible.

Building Extraneous Variables into the Design 

When extraneous variables cannot be adequately controlled by randomization, they can be built into the design as independent variables.  They would have to be added to the purpose of your study and tested for significance along with your other variables.  In this way, their effect can be measured and separated from the effect of the variables you wanted to study initially. 

Particularly in experimental designs, but also in descriptive surveys, the effect of these variables can be subtracted statistically from the total action of the variables.  This method adds to the cost of the study because of the additional data collection and analysis required.  Therefore, it should be used only as a last resort.

In exploratory descriptive studies where the nature of the variables is not known, extraneous variables are said to be built into the design.  The purpose in these studies is to identify the relevant variables and assess their relationship in the data analysis.  Therefore, it is essential that you treat all variables as independent during the data collection, so that no data that later might point to relationships between variables will be overlooked.

The separation of extraneous variables from independent and dependent variables is part of the analysis of data in exploratory research.