You are on page 1of 1

Laura Bess

Critical Reflection 3
STA 301
1. To support hypothesis testing, it would be helpful to include the
value. The value gives the rate of type I error, which is rejecting the
null hypothesis when the null is true. The value is already typically
involved in the calculations and when the calculated p-value is found to
be less than , the null hypothesis is rejected. If the type I error is very
low, that means there is little chance of accidentally rejecting the null
hypothesis when it is actually true, which may also indicate that the
null hypothesis is almost always true even if the value is not the same
exact number as the null hypothesis. For example, suppose a person
claims that the average length of time that a pacemaker functions
correctly is 5.6 years. However, an independent study of this
pacemaker finds that this is not true and thus the null hypothesis is
rejected. It is later found out that the study was conducted by a rival
company and the value was .4, indicating a high likelihood of false
rejection. If the value is not included in calculations and it is claimed
that the null hypothesis is rejected, it would be misleading because it
does not indicate the likelihood of type I error.
2. Another type of error rate is the rate. This rate indicates type II error,
or the likelihood of failing to reject the null hypothesis when it is false.
More importantly, the rate is linked to the power, or the probability
of correctly rejecting the null hypothesis. The power is directly linked to
the sample size, in that as the sample size increases, the power also
increases, giving a more likely result. As your sample size increases,
power increases, and decreases. decreasing means that it is much
more likely that the null hypothesis will be correctly rejected or
accepted. For instance, a company would like to claim that their drug is
more effective at treating a disease than the current drug on the
market, which could be say, 70%. Their study finds that their drug can
cure a disease at a rate of 90%, but they fail to include the power and
thus the value. It is later found out that their power was very low,
and thus a high value. So while they initially claimed that the null
hypothesis was true, that their drug is more effective in curing the
disease, it cannot be assumed that this is the case because we have
no idea at what values these studies were performed. If they were
high values, it may be safer to assume that the drug probably is not
more effective and may even be less effective.

You might also like