In each case, a valid research study would have to demonstrate that
changes in one variable (for example, the amount of exercise) are followed
by changes in the other variable (cholesterol level), and that no other variable
provides an alternative explanation for the results. This kind of validity
is called internal validity. Internal validity is concerned with factors in
the research study that raise doubts or questions about the interpretation of
the results. A research study is said to have internal validity if it allows one
and only one explanation of the results. Any factor that allows an alternative
explanation for the results is a threat to internal validity. For example,
suppose a clinician obtains a group of depressed clients and measures the
level of depression for each individual. The clinician then begins therapy
with the clients and measures depression again after 3 weeks. If there is a
substantial decline in depression, the therapist would like to conclude that
the therapy caused a reduction in depression. However, suppose that the
weather was cold and miserable when the study began, and changed to
bright and sunny when the study ended 3 weeks later. In this case, the
weather provides an alternative explanation for the results. Specifically, it
is possible that the improved weather caused the reduction in depression.
In this example, the weather is a threat to the internal validity of the
research study.