Evidence-Based Practice for Nurses

Selection Bias

Selection bias occurs when the change in the DV is a result of differences in

the characteristics of subjects before they entered a study rather than a result of

the IV. Suppose a researcher wants to compare the effectiveness of group teach-

ing to one-on-one teaching. At the hospital where the study is being conducted,

patients are allowed to choose the kind of teaching they want. It is possible

that people who select group teaching are different on some characteristics

from those who select one-on-one teaching. Selection bias can be minimized

somewhat by the use of random assignment to groups.

History

The threat of history occurs when the DV may have been influenced by some

event other than the IV that occurred during the course of the study. In the

example of condom use, it might be difficult to attribute increased use of con-

doms to the computer intervention if during the same time as the study was

being conducted MTV or VH1 aired public service announcements encour-

aging safe sex practices. In such a situation, it is difficult for the researcher to

know whether the IV or the increased awareness from television produced

the  change in the DV. The threat of history could have

been decreased by including a control group that was

exposed to the television ads but that did not receive the

computer intervention.

Maturation

Over the course of a study, subjects may change either by

growing or becoming more mature. This is known as the

Threats to Internal Validity Threats to External Validity

‹ Selection bias

‹ History

‹ Maturation

‹ Testing

‹ Instrumentation

‹ Mortality

‹ Statistical conclusion validity

‹ Construct validity

–Bias

–Confounding

–Reactivity

‹ Effects of selection

‹ Interaction of treatment and selection of subjects

‹ Interaction of treatment and setting

‹ Interaction of treatment and history

TABLE 6-1 Summary of Design Threats

KEY TERMS

selection bias: A

threat to internal

validity when the

change in the

dependent variable

is a result of the

characteristics of

the subjects before

they entered a

study

history: A threat

to internal validity

when the depen-

dent variable is

influenced by an

event that occurred

during the study

FYI

Evaluating the quality of quantitative evidence

requires nurses to identify potential threats

to validity, evaluate the seriousness of those

threats, and determine whether results are

valid for application to patient care. The two

primary types of validity are internal validity

and external validity.

156 CHAPTER 6 Key Principles of Quantitative Designs