A variable is a quantity that changes. You may remember talking aboutindependent and dependent variables when you were learning how to read graphs inschool. The independent variable was the x-axis (horizontal) label and the dependentvariable was the y-axis (vertical) label.
There is another way to determine the independent and dependent variables. The independent variable is the quantity that you, as the experimenter,control. The value of the independent variable depends only on what youdetermine that it should be. The dependent variable is the unknown. Itis what the experimenter measures. It is the quantitythat changes when you change the independent variable. A common research goalis to determine the relationship between two variables.
Suppose you wanted to know whether tall people are better basketball playersthan short people. You gather a sample of people of various heights and askthem to shoot baskets from the foul line. The number of baskets they make outof ten shots will be your measure of basketball ability.
The variables in this example are HEIGHT and NUMBER OF BASKETS. The heightof the people in your sample cannot be changed- by choosing these particularpeople, you have determined the values of the independent variable.
The number of baskets your subjects will make is unknown. Number of basketsis the dependent variable.
It is very important that a statistician not make inferences or predictionsbased on statistics, if other factors can influence results. For example, ateacher could not make a good prediction of future test results, based on paststatistics, if half the class will be up all night before the next testprotesting an austerity budget. If a student scores ten points below the mean ona test a statistician cannot interpret that as lack of study, since many factorscould account for a low score. These factors are called confoundingvariables. A confounding variable is a variable the experimenter did not takeinto consideration and which could account for the observed relationship betweenthe independent and dependent variables.
Going back to our basketball example above, supppose you found that tallpeople did make better basketball players than short people. A reason for thiscould be that the tall people in your sample had more practice with foul shotsthan the short people in your sample. AMOUNT OF PRACTICE is a confoundingvariable.