In: Statistics and Probability
State the null hypothesis.
Sally was interested in whether the number of m&m's was uniform over all 6 colors (the same number of m&m's for each color). The table categorizes the number of m&m's of each color found in an 1-pound bag.
Color | ||||||
---|---|---|---|---|---|---|
Brown | Red | Yellow | Blue | Orange | Green | |
Number | 70 | 85 | 81 | 79 | 93 | 96 |
State the appropriate null hypothesis for a Chi-square goodness-of-fit test.
Question 21 options:
H0: μ=1/6 |
|
H0: The proportion for at least one of the colors is not 1/6 |
|
H0: The proportion of each color is 1/6 |
|
H0: The number of m&m's is independent of color |
Solution:
Given: Sally was interested in whether the number of m&m's was uniform over all 6 colors (the same number of m&m's for each color).
We have to state the appropriate null hypothesis for a Chi-square goodness-of-fit test.
Under null hypothesis for a Chi-square goodness-of-fit test, we assume all categories are equally distributed or distributed according to specific distribution or distributed according to expected proportions.
Since we are expecting the number of m&m's was uniform over all 6 colors, that is the proportion of each of 6 color is 1/6.
thus correct answer is:
H0: The proportion of each color is 1/6
Option 1 is incorrect , since it states about mean = 1/6 instead of proportions.
Option 2 is incorrect, since Under null hypothesis for a Chi-square goodness-of-fit test, we assume all categories are equally distributed, not the different.
Option 4 is incorrect , since independence is assumed in Chi-square independence of attributes, not in Chi-square goodness-of-fit test.