27.10.2014 Views

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

118 Chapter 5<br />

the population to act as controls—that is, they would not receive <strong>in</strong>struction.<br />

The evaluation team measured blood pressure <strong>in</strong> the treatment group and <strong>in</strong><br />

the control group before the program was implemented. But when they went<br />

back after the program was completed, they found that control group members<br />

had also been chang<strong>in</strong>g their behavior. They had learned of the new diet and<br />

exercises from the members of the treatment group.<br />

Controll<strong>in</strong>g for Threats to Validity<br />

In what follows, I want to show you how the power of experimental logic<br />

is applied to real research problems. The major experimental designs are<br />

shown <strong>in</strong> figure 5.1. The notation is pretty standard. X stands for some <strong>in</strong>tervention—a<br />

stimulus or a treatment. R means that participants are randomly<br />

assigned to experimental conditions—either to the <strong>in</strong>tervention group that gets<br />

the treatment, or to the control group that doesn’t. Several designs <strong>in</strong>clude<br />

random assignment and several don’t. O stands for ‘‘observation.’’ O 1 means<br />

that some observation is made at time 1, O 2 means that some observation is<br />

made at time 2, and so on.<br />

‘‘Observation’’ means ‘‘measurement of some dependent variable,’’ but as<br />

you already know, the idea of measurement is pretty broad. It can be tak<strong>in</strong>g<br />

someone’s temperature or test<strong>in</strong>g their read<strong>in</strong>g skill. It can also be just writ<strong>in</strong>g<br />

down whether they were successful at hunt<strong>in</strong>g game that day.<br />

The Classic Two-Group Pretest-Posttest Design with Random<br />

Assignment<br />

Figure 5.1a shows the classic experimental design: the two-group pretestposttest<br />

with random assignment. From a population of potential participants,<br />

some participants have been assigned randomly to a treatment group and a<br />

control group. Read across the top row of the figure. An observation (measurement)<br />

of some dependent variable or variables is made at time 1 on the members<br />

of group 1. That is O 1 . Then an <strong>in</strong>tervention is made (the group is exposed<br />

to some treatment, X). Then, another observation is made at time 2. That is<br />

O 2 .<br />

Now look at the second row of figure 5.1a. A second group of people are<br />

observed, also at time 1. Measurements are made of the same dependent variable(s)<br />

that were made for the first group. The observation is labeled O 3. There<br />

is no X on this row, which means that no <strong>in</strong>tervention is made on this group<br />

of people. They rema<strong>in</strong> unexposed to the treatment or <strong>in</strong>tervention <strong>in</strong> the

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!