# Ways of Controlling Extraneous Variables

Ways of Controlling Extraneous Variables

In definition, a variable is anything that can change or be changed such as attention, memory, time taken to achieve an objective etc. (Brown, 2013). Variables are accorded different terminologies that are used in experimental investigations such as the commonly used dependent and independent variables. These two are applied when a research seeks to identify the possible effect of a given dependent variable that may occur as a result of altering an independent variable. There are also extraneous variables that are not independent, but they could affect the result of an experiment.  According to Grove, burns & Gray (2014), it is up to a research to control extraneous variables when possible because they are not usually part of the study. These authors argue that this is because they may not be Important enough to give a justified analysis for the effects.

As a result, researchers have been using different methods to control extraneous variables. They include; randomization, matching, statistical control, homogeneous sampling techniques and building the variables into design.

Randomization

Assigning subjects randomly to various treatment as well as control groups helps in controlling all possible extraneous variables. This is because the groups can be regarded as statistically equal when starting an experiment. Brown (2014) argues that this may not mean they are equal, but the probability of them being equals is usually greater than the probability of them not being equals especially if randomization is done correctly.

Matching

Matching is mostly used when randomization is impossible. It may also be used when the experimental groups contain crucial variables or may be too small leaving the option of matching the subjects for those variables.

Statistical control

This control method involves subtracting the effects of extraneous variables statistically from the overall action of the purpose. Most researchers prefer to use the technique of analysis of covariance (ANCOVA) to achieve this.  However, Grove, Burns & Gary (2014) advise that this method should be used as a last resort because it adds cost to the study in question since it requires too much data collection and data analysis.

Building extraneous variables into the design

This method is mainly used when it is not possible to adequately control the extraneous variables by randomization. As a result, the extraneous variables are built into the design as if they were independent variables. This way, they are added to the study group and tested for relevance together with other variables (Brown, 2014). Their significance and effects can then be measured and separated from any other effects of the independent variables.

References

Brown, S. J. (2013). Evidence-based nursing: The research-practice connection. Jones & Bartlett Publishers.

Grove, S. K., Burns, N., & Gray, J. R. (2014). Understanding nursing research: Building an evidence-based practice. Elsevier Health Sciences.