School

Kent State UniversityDepartment

MathematicsCourse Code

MATH 10041Professor

Rajeev RajaramChapter

8 .3This

**preview**shows half of the first page. to view the full**1 pages of the document.**Week 12

Wednesday

11/12/18

Chapter 8 cont.

8.3- Hypothesis Tests in Detail

ā Small p-value is extreme

ā Extreme value means something unusual has happened

ā Cannot use normal curve if the sample size is too small and/or if not randomly selected

ā We cannot make the significance level so super small (ex. 0 = 0%) because that

increases the probability that we will mistakenly fail to reject the null

ā Ex. if the probability of convicting an innocent man is 0%, everyone goes free.

But that means freeing a guilty person is 100%

True Reality

Ho

Ha

Ho

correct

Wrong (type 2)

Ha

Wrong

Correct

Decisionā

ā Statistically significant = reject null

ā Statistically significant findings do not necessarily mean the results are useful

ā Type I error

ā Reject the true null

ā Ex. Hoā¦ānotā

ā Type II error

ā Fail to reject the null that you should actually reject

ā Ex. Ho is true, when actuallyā¦

ā Don't say you āprovedā something --- say you ādemonstratedā or āsolvedā

ā Don't say āaccept the nulllā ---say ācannot rejectā or āfailed to rejectā or āinsufficient

evidence to reject nullā

###### You're Reading a Preview

Unlock to view full version