In: Statistics and Probability
A multinomial experiment produced the following results: (You may find it useful to reference the appropriate table: chi-square table or F table)
Category | 1 | 2 | 3 | 4 | 5 |
Frequency | 87 | 52 | 89 | 88 | 79 |
a. Choose the appropriate alternative hypothesis to test if the population proportions differ.
a All population proportions differ from 0.20.
b Not all population proportions are equal to 0.20.
b. Calculate the value of the test statistic. (Round intermediate calculations to at least 4 decimal places and final answer to 3 decimal places.)
c. Find the p-value.
a 0.01 p-value < 0.025
b p-value < 0.01
c p-value 0.10
d 0.05 p-value < 0.10
e 0.025 p-value < 0.05
d. Can we conclude at the 5% significance level that the population proportions are not equal?
a No, since the p-value is less than the significance level.
b Yes, since the p-value is less than the significance level.
c No, since the p-value is more than the significance level.
d Yes, since the p-value is more than the significance level.
Solution:
Part a. Choose the appropriate alternative hypothesis to test if the population proportions differ.
We have k = 5 categories , thus proportion for each category = 1 / 5 = 0.20
Thus alternative hypothesis is:
b Not all population proportions are equal to 0.20.
( Note under alternative hypothesis, we never say all proportions are different from stated proportion, we say at least one of the proportion is different from stated proportion)
Part b. Calculate the value of the test statistic.
Chi square test statistic for goodness of fit
Where
Oi = Observed Counts
Ei =Expected Counts
Thus we need to make following table:
Category | Oi: Observed frequency | Ei: Expected Frequency | Oi2/Ei |
---|---|---|---|
1 | 87 | 79 | 95.810 |
2 | 52 | 79 | 34.228 |
3 | 89 | 79 | 100.266 |
4 | 88 | 79 | 98.025 |
5 | 79 | 79 | 79.000 |
N = 395 |
Thus
Part c. Find the p-value.
df = k- 1 = 5 - 1 = 4
Look in Chi-square table for df = 2 row and find an interval in
which Chi-square test statistic =
fall,
then find corresponding right tail area interval ( in ascending
order) , which is p-value interval.
Chi-square test statistic = fall between 11.143 and 13.277
corresponding right tail area interval ( in ascending order) is between 0.010 and 0.025
thus p-value interval is:
0.01 < p-value < 0.025
Thus correct answer is:
a 0.01 p-value < 0.025
Part d. Can we conclude at the 5% significance level that the population proportions are not equal?
Since 0.01 p-value < 0.025 < 0.05 significance level, we reject the null hypothesis.
Thus we conclude that: the population proportions are not equal.
Thus correct answer is:
b Yes, since the p-value is less than the significance level.