In this exercise we examine the effect of the sample size on the significance test for comparing two proportions. In each case suppose that p̂1 = 0.65 and p̂2 = 0.45, and take n to be the common value of n1 and n2. Use the z statistic to test H0: p1 = p2 versus the alternative Ha: p1 ≠ p2. Compute the statistic and the associated P-value for the following values of n: 40, 50, 60, 80, 380, 480, and 980. Summarize the results in a table. (Test the difference p1 − p2. Round your values for z to two decimal places and round your P-values to four decimal places.)
n z P-value
40
50
60
80
380
480
980
Explain what you observe about the effect of the sample size on statistical significance when the sample proportions p̂1 and p̂2 are unchanged.
As sample size increases, the test becomes less significant.
As sample size increases, the test becomes more significant.
As sample size increases, there is no effect on significance.
There is not enough information.
SOLUTION:
From given data,
In this exercise we examine the effect of the sample size on the significance test for comparing two proportions. In each case suppose that p̂1 = 0.65 and p̂2 = 0.45, and take n to be the common value of n1 and n2. Use the z statistic to test H0: p1 = p2 versus the alternative Ha: p1 ≠ p2. Compute the statistic and the associated P-value for the following values of n: 40, 50, 60, 80, 380, 480, and 980.
p̂1 = 0.65
p̂2 = 0.45
The value of z for the sample size as 40 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 40 ) + (0.45 (1-0.45) / 40 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 40 ) + (0.45 (1-0.45) / 40 ) )
z = 0.2 / 0.10897247
z = 1.83
The p-value for two tailed ,
P(z=1.83) = 2*P(z > 1.83)
= 2* (1-P(z < 1.83))
= 2* (1- 0.96638)
= 2*0.03362
= 0.06724
The value of z for the sample size as 50 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 50 ) + (0.45 (1-0.45) / 50 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 50 ) + (0.45 (1-0.45) / 50 ) )
z = 0.2 / 0.09746794
z = 2.05
The p-value for two tailed ,
P(z=2.05) = 2*P(z > 2.05)
= 2* (1-P(z < 2.05))
= 2* (1- 0.97982)
= 2*0.02018
= 0.04036
The value of z for the sample size as 60 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 60 ) + (0.45 (1-0.45) / 60 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 60 ) + (0.45 (1-0.45) / 60 ) )
z = 0.2 / 0.08897565
z =2.24
The p-value for two tailed ,
P(z=2.24) = 2*P(z > 2.24)
= 2* (1-P(z < 2.24))
= 2* (1- 0.98745)
= 2*0.01255
= 0.0251
The value of z for the sample size as 80 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 80 ) + (0.45 (1-0.45) / 80 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 80 ) + (0.45 (1-0.45) / 80 ) )
z = 0.2 / 0.07705517
z =2.59
The p-value for two tailed ,
P(z=2.59) = 2*P(z > 2.59)
= 2* (1-P(z <2.59))
= 2* (1- 0.99520)
= 2*0.0048
= 0.0096
The value of z for the sample size as 380 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 380 ) + (0.45 (1-0.45) / 380 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 380 ) + (0.45 (1-0.45) / 380 ) )
z = 0.2 / 0.03535533
z =5.65
The p-value for two tailed ,
P(z=5.65) = 2*P(z > 5.65)
= 2*0
=0
The value of z for the sample size as 480 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 480 ) + (0.45 (1-0.45) / 480 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 480 ) + (0.45 (1-0.45) / 480 ) )
z = 0.2 / 0.03145764
z =6.35
The p-value for two tailed ,
P(z=6.35) = 2*P(z > 6.35)
= 2*0
=0
The value of z for the sample size as 980 is obtained as shown below:
z = (p̂1 - p̂2) / sqrt ( (p̂1 (1-p̂1) / n ) + (p̂2 (1-p̂2) / n ) )
z = (0.65 - 0.45) / sqrt ( (0.65 (1-0.65) / 980 ) + (0.45 (1-0.45) / 980 ) )
z = 0.2 / sqrt ( (0.65 (1-0.65) / 980 ) + (0.45 (1-0.45) / 980 ) )
z = 0.2 / 0.02201576
z =9.08
The p-value for two tailed ,
P(z=9.08) = 2*P(z > 9.08)
= 2*0
=0
Please thumbs-up / vote up this answer if it was helpful. In case of any problem, please comment below. I will surely help. Down-votes are permanent and not notified to us, so we can't help in that case.
In this exercise we examine the effect of the sample size on the significance test for...
The null and alternative hypotheses for a test are given, as well as some information about the actual sample and the statistic that is computed for each randomization sample. Indicate where the randomization distribution will be centered. Ho = p1 = p2 vs. ha p1 < p2 Sample: p̂1 = 0.3, n= 20 p̂2 = 0.167, n=12 Randomization statistic: p̂1 - p̂2 Answer _______________
The null and alternative hypotheses for a test are given, as well as some information about the actual sample and the statistic that is computed for each randomization sample. Indicate where the randomization distribution will be centered. Ho: p1 = p2 vs. Ha: p1>p2 Sample: p̂1 = 0.3, n=20, p̂2 = 0.467 n=12 Randomization statistic p̂1 - p̂2 P.S. The answer is not -0.167
Gun Murders - Texas vs New York - Significance Test In 2011, New York had much stricter gun laws than Texas. For that year, the proportion of gun murders in Texas was greater than in New York. Here we test whether or not the proportion was significantly greater in Texas. The table below gives relevant information. Here, the p̂'s are population proportions but you should treat them as sample proportions. The standard error (SE) is given to save calculation time...
The sample size needed to estimate the difference between two population proportions to within a margin of error m with a significance level of α can be found as follows. In the expression m=z∗p1(1−p1)n1+p2(1−p2)n2−−−−−−−−−−−−−−−−−−−−√ we replace both n1 and n2 by n (assuming that both samples have the same size) and replace each of p1, and p2, by 0.5 (because their values are not known). Then we solve for n, and get n=(z∗)22E2. Finally, increase the value of n to...
For one binomial experiment, n1 = 75 binomial trials produced r1 = 60 successes. For a second independent binomial experiment, n2 = 100 binomial trials produced r2 = 85 successes. At the 5% level of significance, test the claim that the probabilities of success for the two binomial experiments differ. (a) Compute the pooled probability of success for the two experiments. (Round your answer to three decimal places.) (b) Check Requirements: What distribution does the sample test statistic follow? Explain....
I need help with the last part of this problem. I posted the first parts that I already have the answers to also but only need help with the final part. Thanks. In 1950, an organization surveyed 1100 adults and asked, "Are you a total abstainer from, or do you on occasion consume, alcoholic beverages?" Of the 1100 adults surveyed, 352 indicated that they were total abstainers. In a recent survey, the same question was asked of 1100 adults and...
(1 point) The sample size needed to estimate the difference between two population proportions to within a margin of error E with a significance level of α can be found as follows. In the expression E=z∗p1(1−p1)n1+p2(1−p2)n2‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾√ we replace both n1 and n2 by n (assuming that both samples have the same size) and replace each of p1, and p2, by 0.5 (because their values are not known). Then we solve for n, and get n=(z∗)22E2. Finally, increase the value of...
Quel He Rhino viruses typically cause common colds. In a test of the effectiveness of echinacea, 39 of the 46 subjects treated with echinacea developed rhinovirus infections. In a placebo group. 73 of the 90 subjects developed rhinovirus infections. Use a 0.01 significance level to test the claim that echinacea has an effect on rhinovirus infections. Complete parts (a) through (c) below. a. Test the claim using a hypothesis test. Consider the first sample to be the sample of subjects...
What can be said about the relationship between the significance and the effect size of a hypothesis test? Group of answer choices The significance of a test is not determined by the effect size alone. A significant test result leads to a large effect size while a non-significant test result leads to a small effect size. The effect size is determined by the alpha level while the significance of a test is not. The effect size is always the same...
Conduct the following test at the a = 0.01 level of significance by determining (a) the null and alternative hypotheses, (b) the test statistic, and (c) the P-value. Assume that the samples were obtained independently using simple random sampling. Test whether P1 * P2. Sample data are xy = 28, n = 254, X2 = 38, and n2 = 301. (a) Determine the null and alternative hypotheses. Choose the correct answer below. O A. Ho: P1 = P2 versus Hy:...