Lesson #2: Random Assignment of 40 Subjects to 4 Conditions
Research Randomizer can be used just as easily for random assignment as for random sampling. Although people sometimes confuse random assignment with random sampling, the two are really quite different. With random sampling, the goal is to choose a representative set of cases from the full population under consideration. With random assignment, the goal is usually to give all participants an equal chance of being assigned to each experimental condition (regardless of how representative the participants are).
By assigning participants at random, an experimenter equalizes pre-existing individual differences across experimental conditions. This makes it easier to infer that any observed group differences are due to the independent variable. Thus, random assignment tends to increase the internal validity of a study, whereas random sampling increases both internal validity and external validity (generalizability).
To see how Research Randomizer can be used for random assignment, suppose you're a medical researcher running a small clinical trial to test a risky new drug. Your goal is to determine the optimal dose of the drug, maximizing benefits while minimizing side effects. To do this, you want to randomly assign 40 volunteers to 4 experimental conditions:
Condition 1 = Wonderdrug 5%
Condition 2 = Wonderdrug 10%
Condition 3 = Wonderdrug 15%
Condition 4 = Placebo
As with the earlier example, you would begin this process by giving each volunteer an identification number (Participant #1, Participant #2, Participant #3, and so on). Then you would go to the Randomizer form and generate 1 set of 40 non-unique, unsorted numbers with a range from 1 to 4 (representing the condition numbers).
Random Sample of 50 Subjects from a Population of 643
Random Assignment of 40 Subjects to 4 Conditions
Random Assignment of 40 Subjects in Blocks of 4
Random Sample of 100 Telephone Numbers
Random Ordering (Assignment) of 25 Items on a Test
