Social Research Methods: Qualitative and Quantitative Approaches

EXPERIMENTAL RESEARCH

FIGURE 1 Random Assignment and Random Sampling

Random Assignment

Random Sampling

Population (Sampling Frame)

Sample

Random

Process

Random

Process

Note: Shading indicates various skin tones.

Experimental Group

Control Group

Pool of Subjects

randomly assign. We can first sample to obtain a
smaller set of cases (e.g., 150 people out of 20,000)
and then use random assignment to divide the
smaller set into groups (e.g., divide the 150 people
into three groups of 50).

How to Randomly Assign

Random assignment is simple in practice. We begin
with a collection of cases (i.e., individuals, teams,
companies, or whatever the unit of analysis is) and
then divide the collection into two or more groups
using a random process, such as asking people to
count off, tossing a coin, or throwing dice. For
example, we want to divide thirty-two people into
two groups of sixteen. We could have each write his
or her name on a standard size slip of paper, put all

slips in a hat, mix the slips with our eyes closed, and

then with eyes still closed, draw the first sixteen

names for group 1 and the second sixteen for group

2. A specific situation can be unusual and the groups
   may differ. For example, it is possible although ex-
   tremely unlikely that all cases with one character-
   istic will end up in one group. For example, we have
   thirty-two people with sixteen males and sixteen fe-
   males, but all of the males end up in one group and
   all of the females in another. This is possible by ran-
   dom chance but extremely rare (see in Figure 2 on
   random assignment).

Matching versus Random Assignment

If the purpose of random assignment is to get two

(or more) equivalent groups, you may ask whether