In: Statistics and Probability
Is there a relationship between confidence intervals and two-tailed hypothesis tests? Let c be the level of confidence used to construct a confidence interval from sample data. Let α be the level of significance for a two-tailed hypothesis test. The following statement applies to hypothesis tests of the mean.
For a two-tailed hypothesis test with level of significance α and null hypothesis H0: μ = k, we reject H0 whenever k falls outside the c = 1 – α confidence interval for μ based on the sample data. When k falls within the c = 1 – α confidence interval, we do not reject H0.
(A corresponding relationship between confidence intervals and two-tailed hypothesis tests also is valid for other parameters, such as p, μ1 − μ2, or p1 − p2, which we will study in later sections.) Whenever the value of k given in the null hypothesis falls outside the c = 1 – α confidence interval for the parameter, we reject H0. For example, consider a two-tailed hypothesis test with α = 0.01 and
H0: μ = 21
H1: μ ≠ 21
A random sample of size 22 has a sample mean x = 19 from a population with standard deviation σ = 6.
(a) What is the value of c = 1 − α?
Using the methods of Chapter 7, construct a 1 − α
confidence interval for μ from the sample data. (Round
your answers to two decimal places.)
lower limit | |
upper limit |
What is the value of μ given in the null hypothesis (i.e.,
what is k)?
k =
Is this value in the confidence interval?
Yes No
Do we reject or fail to reject H0 based on this
information?
Fail to reject, since μ = 21 is not contained in this interval.
Fail to reject, since μ = 21 is contained in this interval.
Reject, since μ = 21 is not contained in this interval.
Reject, since μ = 21 is contained in this interval.
(b) Using methods of Chapter 8, find the P-value for the
hypothesis test. (Round your answer to four decimal places.)
Do we reject or fail to reject H0?
Reject the null hypothesis, there is sufficient evidence that μ differs from 21. Fail to reject the null hypothesis, there is insufficient evidence that μ differs from 21. Fail to reject the null hypothesis, there is sufficient evidence that μ differs from 21. Reject the null hypothesis, there is insufficient evidence that μ differs from 21.
Compare your result to that of part (a).
We rejected the null hypothesis in part (a) but failed to reject the null hypothesis in part (b). These results are the same. We rejected the null hypothesis in part (b) but failed to reject the null hypothesis in part (a).
Is there a relationship between confidence intervals and two-tailed hypothesis tests? Let c be the level of confidence used to construct a confidence interval from sample data. Let α be the level of significance for a two-tailed hypothesis test. The following statement applies to hypothesis tests of the mean.
For a two-tailed hypothesis test with level of significance α and null hypothesis H0: μ = k, we reject H0 whenever k falls outside the c = 1 – α confidence interval for μ based on the sample data. When k falls within the c = 1 – α confidence interval, we do not reject H0.
(A corresponding relationship between confidence intervals and two-tailed hypothesis tests also is valid for other parameters, such as p, μ1 − μ2, or p1 − p2, which we will study in later sections.) Whenever the value of k given in the null hypothesis falls outside the c = 1 – α confidence interval for the parameter, we reject H0. For example, consider a two-tailed hypothesis test with α = 0.01 and
H0: μ = 21
H1: μ ≠ 21
A random sample of size 22 has a sample mean x = 19 from a population with standard deviation σ = 6.
(a) What is the value of c = 1 − α?
0.99
Using the methods of Chapter 7, construct a 1 − α
confidence interval for μ from the sample data. (Round
your answers to two decimal places.)
lower limit |
15.70 |
upper limit |
22.30 |
What is the value of μ given in the null hypothesis (i.e.,
what is k)?
k =21
Is this value in the confidence interval?
Yes
Fail to reject, since μ = 21 is contained in this interval.
(b) Using methods of Chapter 8, find the P-value for the
hypothesis test. (Round your answer to four decimal places.)
P=0.1179
Do we reject or fail to reject H0?
Fail to reject the null hypothesis, there is insufficient evidence that μ differs from 21.
Compare your result to that of part (a).
These results are the same.
Confidence Interval Estimate for the Mean |
|
Data |
|
Population Standard Deviation |
6 |
Sample Mean |
19 |
Sample Size |
22 |
Confidence Level |
99% |
Intermediate Calculations |
|
Standard Error of the Mean |
1.2792 |
Z Value |
-2.5758 |
Interval Half Width |
3.2950 |
Confidence Interval |
|
Interval Lower Limit |
15.7050 |
Interval Upper Limit |
22.2950 |
Z Test of Hypothesis for the Mean |
|
Data |
|
Null Hypothesis m= |
21 |
Level of Significance |
0.01 |
Population Standard Deviation |
6 |
Sample Size |
22 |
Sample Mean |
19 |
Intermediate Calculations |
|
Standard Error of the Mean |
1.2792 |
Z Test Statistic |
-1.5635 |
Two-Tail Test |
|
Lower Critical Value |
-2.5758 |
Upper Critical Value |
2.5758 |
p-Value |
0.1179 |
Do not reject the null hypothesis |