Experimental Method (cont’d)
Study time: independent variable
Grade: dependent variable
Control group studies 0 hours; other groups study 5 and 10
Study time Grade
Hypothesis was not supported; more study time didn’t equal higher grade
Another variable not the same – last group got no sleep the night before
If you fix that your study can be repeated
Confounding variable – a variable that interferes with the results of the study. It affects the dependent variable too
(alongside the independent variable), so you don’t know whether the effects are caused by the independent variable,
the confounding variable, or both.
In an experiment, you want to be able to say that the I.V. is the cause for the D.V.
A confounding variable interferes with this.
Ex: expectation can be a confounding variable in a study with a control group which is not a placebo control group
Extraneous variable – also applies to the experiment, but isn’t an issue.
Ex: noise in the background of a reaction time study; it can become confounding when only one group is exposed to
There can be a problem if you put people into groups according to what they prefer. There may be a factor which
separates them from the rest of the population which will bias the results of the study.
Random assignment must be used to distribute subjects.
Random Assignment to Groups
Random selection is not the same as random assignment;
By using random assignment, you are assuming that all “individual difference” variables are being evenly distributed
among groups so that they are essentially equal.
Thus, not every person taking part in a study is equal; the groups average out to be (about) equal.
Individual difference variables can make groups unequal; these variables can potentially become confounding
variables. (You make the assumption that random assignment means equal groups)
If this doesn’t work out (and you find out after the fact), you can a) control the variable in question or b) introduce
the confounding variable as an additional independent variable.
Correlational Method of Research
Experimental method – IV, DV, Control group, random assignment
www.notesolution.com Studies such as effects of child abuse on behaviour, effects of cigarette smoke on lung cancer– no way you can use
the experimental method.
Thus you have to pick an alternative – Correlational method
Correlational method - measures the degree of relationship between 2 variables. There is no attempt to manipulate
or control variables. Rather, naturally occurring variations are measured to see if they are related. This method is
useful for making predictions.
**However, you cannot infer cause and effect the way you can from the experimental method**
Ex: exercise and psychological well being – which causes which?
[Co = 2 Relation = between variables]
Scatter plot/diagram – shows correlation between 2 variables graphically
When points have no strong relation, there is no relation between the variables
The stronger th