Testing Threats

Testing can be a threat to internal validity when the fact that a group has taken a pretest or a number of pretests causes an observed effect on the posttest. In other words, a group may perform better on a posttest not because of the treament that was implemented, but because the pretest primed the group members to perform better. For example, if a class is given a pretest in science before being given a special science program, the students might remember some of the questions and correct answers. When they are given the test again, they may answer more questions correctly because of their familiarity with the test rather than an improvement in their overall knowledge of science. 

How do you think a comparison group could alleviate this threat to internal validity? 

Back

---

Home | Name That Threat | Trochim Home | Research Methods Tutorials 
History | Instrumentation | Maturation | Mortality | Regression 

---

Copyright © 1997 Wendy Martin. All rights reserved.