threat of maturation. Changes such as these are more likely to influence the
DV when a study continues over time. For example, if subjects during the study
on condom use came of age to obtain a driver’s license and get a job, it would
be difficult to determine whether an increase in condom use was a result of the
IV or of the subjects’ increased ease of purchasing condoms for use. A control
group may help limit this threat to internal validity.
Testing
The threat of testing occurs when a pretest influences the way subjects respond
on a posttest. Repeated testing can cause familiarity with the test itself, and an-
swers may reflect subjects’ abilities to remember how questions were answered
previously rather than reflecting current knowledge and beliefs.
Instrumentation
When there are changes made in the way variables are measured, the threat
of instrumentation can occur. For example, in a study measuring blood
pressure, if the original measurement is taken using an aneroid sphygmo-
manometer but later measurements are taken using an automated device, it
makes it difficult to determine whether any change in blood pressure read-
ings is a result of the IV or the change in devices. Another potential instru-
mentation threat arises when data are collected by observation or interview
using different data collectors. To control for this threat, researchers need
to ensure all data collectors are comprehensively trained. Researchers also
evaluate interrater reliability to determine the degree of consistency among
individuals collecting data.
Mortality
Mortality refers to the loss of subjects before the study is completed. Loss
of subjects may be a threat to internal validity if there is a difference in the
characteristics of the subjects who dropped out compared to those who
completed the study. Internal validity is also threatened when there is a dif-
ference in the loss of subjects between the experimental and control groups.
Mortality tends to increase the longer a study lasts. In health-related research,
emotional states such as depression and anxiety, as well as physical states
such as fatigue, may influence dropout rates. When appraising a study for
the threat of mortality, it is important to compare the number of subjects
who entered the study with the number of subjects in the final sample. In a
research article, the term attrition rate refers to the dropout rate. If the at-
trition rate is high, the author of the article should provide an analysis and
explanation for the dropout rate.
KEY TERMS
maturation: A
threat to internal
validity when
subjects change
by growing or
maturing
testing: A threat
to internal validity
when a pretest
influences the way
subjects respond on
a posttest
instrumentation:
A threat to internal
validity when there
are inconsistencies
in data collection
mortality: A threat
to internal validity
when there is a
loss of subjects
before the study is
completed; attrition
rate
attrition rate:
Dropout rate; loss of
subjects before the
study is completed;
threat of mortality6.2 What Is Validity? 157