Chapter 5 - Introduction to Experimental Research
All research methods have advantages and disadvantages. The method of choice depends on
what you want to find.
Advantages of the experimental method: Uses random assignment which allows you to infer
cause and effect. Control vs. experimental group. Vary one or more variables while holding
Disadvantages: Generalizability. Lab versus field research. Most experimental research is done
in the lab and most of the time this research is not realistic.
2 types of variables:
Variable: anything that can be changed. Comes from the word vary.
You have to have at least one IV and one DV. In some cases, there can be more than one.
Independent variable (IV): this is what the experiment manipulates or varies to see if it has an
effect on a given behaviour. (CAUSE)
Dependant variable (DV): this is the behaviour that is measured to see if the IV has an effect.
Example: To see if alcohol impairs memory.
IV: alcohol. 0oz, 2oz and 4oz (3 levels=conditions=groups)
DV is dependent on the IV. Memory scores depend on how much alcohol is consumed.
If it is not mentioned then it is not an IV and/or DV.
Extraneous variable: variables that are not controlled and are of no interest to the experiment.
They are not problematic as long as they are held constant. For example: every subject is
exposed to it. Assumption that there are an equal number of subjects in all groups (random
assignment). An extraneous variable becomes a problem when it is not held constant.
Confounding variable: also known as a “uncontrolled extraneous variable”. A variable that you
do not want. Example: study time vs. grade. 0hrs, 61%; 5hrs, 73%; 10hrs, 61%. Now what if the
10 hour group did not get any sleep? Then sleep is a confounding variable. A confounding
variable changes with the independent variable. It can also affect the dependant variable. To
eliminate a confounding variable you can change it into an independent variable. You do this by
having half sleep and the other half not sleep.
Control Group: must have in an experiment. The group that gets the 0 level of the I.V. The
treatment is withheld. Used as the basis for comparison. Any other group besides the control group is called an experimental group. All groups must be equal in every way except for the IV.
On average these groups are equal to each other.
I.V can be a manipulated variable or subject variable.
Manipulated variable: the experimenter creates the situations that the subjects will encounter.
Subject variable: refers to already existing characteristics of subjects. Something that you
cannot take away from the subject.
Is it possible to randomly assign subjects into groups that are created? (To differentiate between
manipulated and subject variables)
Example: age is selected not assigned. It is not possible to have a real control group. 2 things:
no random assignment and you have to select your subjects. YOU CANNOT INFER CAUSE
AND EFFECT with subject variables. Quasi-experiments. All you can say is that the groups
With a manipulative variable, you can infer cause and effect.
VALIDITY: (in an experiment)