HOME
CONCEPTS
      Confounding Variable

      A goal of conducting a randomized experiment is
to make a causal conclusion about the effect of an
independent variable on a dependent variable.  In
order to make a causal conclusion, one must be able
to rule out all other explanations for the effect on the
dependent variable.   In order to achieve the goal of
making a causal conclusion about the effect of an
independent variable on a dependent variable in an
experiment, the participants would be randomly
assigned to conditions and there would be no other
variables that could explain the effect on the
dependent variable.  However, with some
experiments, there could be a confounding variable or
confounding variables that may explain the effect on
the dependent variable.  If there are one or more
confounding variables, then one would
not be able to
make causal conclusions.  What is a confounding
variable?  Below is a definition of a confounding
variable.

Confounding Variable Definition:

A confounding variable in an experiment is a
variable other than the independent variable that
may explain the effect on the dependent variable.

Confounding Variable Example:

   An example of a confounding variable may help to
gain a better understanding of the definition of a
confounding variable.  Below is a description of a
hypothetical experiment that involves a confounding
variable example:
    Imagine that you conducted an experiment to
address the hypothesis that greater detail in a letter of
recommendation would make the letter more
persuasive.   To address this hypothesis, you create
two versions of a letter of recommendation that differ
with respect to amount of detail.   In one condition,
the letter of recommendation has many details.  In the
other condition, the letter of recommendation has few
details.  Moreover, you randomly assign participants
to conditions.  In this experiment, you find that, on
the average, the letter of recommendation is
perceived to be more persuasive in the condition in
which there are many details than in the condition in
which there are few details.  However, there may be
at least one confounding variable in the experiment
described above.  For example, the number of words
in each version of the letter of recommendation is one
confounding variable. The condition with greater
detail may also have more words.  This, it is possible
that a difference between the two conditions with
respect to persuasiveness is due to the number of
words, rather than the amount of detail.  Thus, in the
above experiment one would
not be able to make a
causal conclusion about the effect of amount of detail
in a letter of recommendation on the persuasiveness
of the letter because there is a confounding variable
that provides another explanation for the results of
the experiment.