Home page of the University of the West of England, Bristol Search | Site admin

Research Methods Room

The constellations in this room are concerned with how you will be doing your research. There is a very wide range of research methods available to you, and you will have to spend some time deciding which method or methods are the most appropriate, both with regard to your topic and to the approach you have decided to take. If you are a student in the physical sciences then you will probably find you tend toward quantitative or positivist methods. If you are studying the social sciences then qualitative or interpretive methods may be more appropriate. But this is a generalization, and much of the really interesting and innovative research uses a blend of both qualitative and quantitative methods. As you study the constellations visible from this room, keep referring to the continuum on the left and see if you can spot your position on it, and if you might move along it.

Research methods > Experimental research design > Extraneous and confounding variables

Extraneous and confounding variables

camouflaged insectWhat are they and why are they important when designing an experiment?

Extraneous and confounding variables are variables other than the independent variable which may have an effect on the dependent variable. They are important when designing your experiment because they could potentially alter your results leading to misinterpretation and flawed conclusions!

Extraneous variables

Extraneous variables are often classified according to their origin:

Subject variables are inherent characteristics of the Experimental Unit that might affect outcomes. Hence examples of subject variables might include age, gender and other demographic details (among subjects) and x, y and z (among objects) although this is very much dependent on the object in the experiment.

Experimental variables are characteristics of the experimenter or the experimental team which might influence how the experiment is conducted, or how the experimental subject responds/behaves in the experimental setting. There is a wide definition for these variables and they may include age, gender, qualifications, etc.

Situational variables are characteristics of the environment in which the experiment is being conducted which may have an effect on the results. The nature of these variables is very much dependent on the nature of the experiment but temperature, time and humidity could all be situational variables.

What can be done to stop extraneous variables affecting your experiment?

It is necessary to control extraneous variables so that results are not undermined by their effect. There are two ways to do this:

  • Ensuring that an extraneous variable remains the same for all experimental units in the experiment.
    • This requires that you are aware of the extraneous variable during the design stage and that you can control it.
  • Balance the variable across experimental groups
    • This enables comparisons to be made between experimental units on the basis of the effect of the variable.

    Check your understanding of extraneous variables

    Can you identify potential extraneous variables in these example experiments? Click here

    An example of extraneous variables : The telephone box experiment

    telehpone boxIn a social psychology experiment designed to assess whether men or women give lower ratings of discomfort when 6 people are crowded into a telephone box, the independent variable would be gender and the dependent variable would be the rating of discomfort given by the 6 participants (Example adapted from Kantowitz et al., 2005). The extraneous variables might include: the size of the telephone box, the size of the participants, etc. To ensure that only the independent variable of interest is able to have an effect on the dependent variable, the experimenter may want to control the extraneous variables during the experiment for example to control the size of the telephone booth, the experimenter may ensure that the same booth is used for each replication of the experiment and to control the size of the participants, the experimenter may decide only to use participants of a specified height and weight.

  • Confounding variables

    A confounding variable or factor is also sometimes referred to as a confounder or a lurking variable. It is a "hidden" variable in a statistical or research model that affects the variables in question but is not known or acknowledged, and thus (potentially) distorts the resulting data. This hidden third variable causes the two measured variables to falsely appear to be in a causal relation. Such a relation between two observed variables is termed a spurious relationship. An experiment that fails to take a confounding variable into account is said to have poor internal validity.

    An example of a confounding variable: Ice cream

    ice creamFor example, ice cream consumption and murder rates are highly correlated. Now, does ice cream incite murder or does murder increase the demand for ice cream? Neither: they are joint effects of a common cause or lurking variable, namely, hot weather. Another look at the sample shows that it failed to account for the time of year, including the fact that both rates rise in the summertime.

    (Wikipedia 2006)

    You might find it useful to consider 'The Hawthorn Effect'. A term used to describe when an experimental effect is seen but the effect is caused by the participants knowing that they are being watched (a confounding variable when the experiments took place!) and not as a result of the experimental intervention directly. This Wikipedia article describes some of the debates surrounding The Hawthorn Effect and also descriptions of some of the original experiments conducted at the Hawthorn works of the Western Electric Company, Chicago, between 1924 and 1933.


    Add a comment:

    Note: All comments are subject to moderation and will be displayed once approved by the site administrator.

    Your name:

    Your comment: