Individuals make decisions every day. They make decisions based on their assumptions about
the world, only to find out later that the assumptions were incorrect. Can you think of times
when you have made a type I error? What about a situation when a type II error was made?CRITICAL THINKING EXERCISE 13-4
The Null Hypothesis
Is True in Real
WorldThe Null Hypothesis
Is False in Real
World
Researcher Accepts
the Null HypothesisNo Error Type II ErrorResearcher Rejects
the Null HypothesisType I Error No ErrorTABLE 13-10 Type I and Type II Errors
represents whether the null hypothesis is true or false in the real world. The
other axis represents the two decisions that can be made by researchers about
the null hypothesis: to accept or to reject. The center boxes are then filled in as
appropriate. No errors are made when a decision to accept a null hypothesis is
made when it is true in the real world. Likewise, there is no error when a false
null hypothesis is rejected. A type I error is made when researchers, obtaining
statistically significant results, reject the null hypothesis when in fact it was true.
A type II error occurs when researchers fail to obtain statistically significant
results, and thus they accept the null hypothesis despite the fact that it is false.
Another way to remember type I and type II errors is to think of the acronym
RAAR (Gillis & Jackson, 2002). RAAR stands for the phrase: “Reject the null
hypothesis when you should accept it, and accept the null hypothesis when
you should reject it.” The first two letters, RA, stand for type I error. The second
two letters, AR, stand for type II error.Level of Significance: Adjusting the Risk
of Making Type I and Type II Errors
In health care, type I errors are considered to be more serious than type II errors
are (Smith, 2012). It seems much more risky to claim that a treatment works when
in reality it does not than to miss the opportunity to claim that a treatment works.
For example, a researcher invents and tests a new device for measuring blood
sugar. If a type I error is made, the researcher claims that the new device works358 CHAPTER 13 What Do the Quantitative Data Mean?