Of some randomly selected Starburst candies in my Amazon-bought
package, 27 are orange, 8 are red,
15 are yellow, and 10 are pink.
(a) Let po, pr, py, and pp denote the proportions of orange, red,
yellow and pink candies in the
Starburst package respectively. For the test if whether or not
there are equal numbers of orange,
red, yellow and pink candies in the package, state the null and
alternative hypotheses, in terms
of po, pr, py, and pp.
(b) Calculate the test statistic and p-value for the test above. Is
there evidence to suggest that the
number of candies are not equal? If yes, what is the smallest at
which the test is signicant?
1
Here chi sqauer test will be used.
(a)
If the proportions of orange, red, yellow and pink candies in the Starburst package are equal then each proprotion should be equal to
po = pr = py = pp =0.25
The hypotheses are:
(b)
Following table shows the calculations:
O | p | E=p*60 | (O-E)^2/E | |
Orange | 27 | 0.25 | 15 | 9.6 |
Red | 8 | 0.25 | 15 | 3.266666667 |
Yellow | 15 | 0.25 | 15 | 0 |
Pink | 10 | 0.25 | 15 | 1.666666667 |
Total | 60 | 60 | 14.53333333 |
The test statistics is:
Degree of freedom: df=4-1=3
The p-value is: 0.0023
Since p-value is less than 0.01 so we reject the null hypothesis.
That is there is evidence to suggest that the number of candies are not equal.
Get Answers For Free
Most questions answered within 1 hours.