error because there is no temperature change in the real world. The goal of
hypothesis testing is to accept true claims and reject false ones.
Type I and Type II Errors
The goal is to avoid the two kinds of errors that can be made when making
decisions about null hypotheses. These errors are known as type I and type II
errors. A type I error occurs when the researcher rejects the null hypothesis
when it should have been accepted. In nursing, when the null hypothesis is
wrongly rejected, the usual result is that the researcher makes false claims about
the research hypothesis. Usually this means that the researcher claims that some
treatment works or some relationship exists, when in actuality that is not the
case. For example, in the previous examples, the researcher hypothesized that
skin temperatures over the sacrum would be increased prior to the appear-
ance of redness. Now suppose that in the real world, there is no change in the
temperature; however, analysis of the data collected indicates that there is an
increase of temperature. In other words, the researcher has obtained statisti-
cally significant results. This false finding could be the result of any number of
errors, such as sampling bias and measurement error. The false finding could
have happened by chance, just as it would be possible by chance to get 20 heads
in a row when tossing a fair coin. The researcher, unaware about what is true in
the world, rejects the null hypothesis based on the statistical analysis and claims
that the research hypothesis is supported. This is a type I error. If nurses adopt
this finding into practice, they will unnecessarily spend time measuring skin
temperature because it would provide no indication of risk. If the practice of
measuring skin temperature continued without evaluation of patient outcomes,
it is possible that this type I error would never be discovered.
A type II error occurs when researchers accept the null hypothesis when
it should have been rejected. In nursing, this type of error usually means
that practice does not change when it should be changed. The opportunity
to implement an effective treatment or claim the discovery of a relationship
has been missed. Consider this variation of the previous example. The re-
searcher still hypothesizes that sacral skin temperature is increased prior to
the appearance of redness, and in the real world this is true. The researcher
completes the analysis, which has statistically nonsignificant results. These
nonsignificant results can be the result of error or chance. Based on the
statistical analysis, the researcher is forced to accept the null hypothesis,
unaware that the research hypothesis is true in the real world. A type II error
has occurred, and nurses miss an opportunity to predict which patients are
at risk for skin breakdown.
There are a few strategies for remembering type I and type II errors. One
way is through the graphic representation in Table 13-10. One axis of the table
KEY TERMS
type I error: When
the researcher
rejects the null
hypothesis when it
should have been
accepted
type II error: When
the researcher
inaccurately
concludes that
there is no
relationship among
the independent
and dependent
variables when an
actual relationship
does exist; when the
researcher accepts
the null hypothesis
when it should have
been rejected13.7 Reducing Error When Deciding About Hypotheses 357