In: Statistics and Probability
Gretchen wants to know whether gender has a relationship to beverage choice among her coworkers. She defines beverage choice as the first beverage consumed upon arrival at work and selects a random sample of her coworkers. She records her coworkers' consumption habits one morning and the results are shown in the table. Is there sufficient evidence at the 0.05 level to say that gender and beverage choice are related?
Observed Sample of 90 Coworkers:
Coffee | Tea | Water | Other | Total | |
Male | 10 | 7 | 18 | 9 | 44 |
Female | 8 | 12 | 11 | 15 | 46 |
Total | 18 | 19 | 29 | 24 | 90 |
1. State the hypotheses for this test
2. Determine the number of females whose first choice was
expected to be water.
Possible Answers: A: 9.92 B: 20.3 C:14.82 D:11.0
3. Compute the value of the X2 test statistic.
Possible Answers: A: 0.196 B: 4.686 C: 10.773 D:
11.212
4. State the conclusion of this hypothesis test.
Possible Answers: A: Reject H0. There is enough evidence to support
the claim that gender and beverage choice are related.
B: Fail to reject H0. There is not enough evidence to support the
claim that the means of the two genders are the same.
C: Reject H0; There is enough evidence to state that females prefer
to drink tea.
D: Fail to reject H0. There is not enough evidence to support the
claim that gender and beverage are related.
I need help walking through the steps - I am making errors along the way, don't have too many issues with 1&4 but the middle is a little iffy.
Chi-Square Test |
Observed Frequencies(fo) | ||||||||||
Column variable | Calculations | |||||||||
Row variable | coffee | tea | water | other | Total | fo-fe | ||||
male | 10 | 7 | 18 | 9 | 44 | 10-8.8=1.2000 | -2.2889 | 3.8222 | -2.7333 | |
female | 8 | 12 | 11 | 15 | 46 | -1.2000 | 2.2889 | -3.8222 | 2.7333 | |
Total | 18 | 19 | 29 | 24 | 90 | |||||
Expected Frequencies(fe) = Σrow*Σcolumn/Σtotal | ||||||||||
Column variable | ||||||||||
Row variable | coffee | tea | water | other | Total | (fo-fe)^2/fe | ||||
male | 18*44/90=8.80 | 19*44/90=9.29 | 29*44/90=14.18 | 24*44/90=11.73 | 44 | 0.1636 | 0.5640 | 1.0304 | 0.6367 | |
female | 18*46/90=9.20 | 19*46/90=9.71 | 29*46/90=14.82 | 24*46/90=12.27 | 46 | 0.1565 | 0.5395 | 0.9856 | 0.6091 | |
Total | 18 | 19 | 29 | 24 | 90 |
1)
Ho:gender and beverage choice are not related
H1:gender and beverage choice are related
2)
Expected Frequencies(fe) = Σrow*Σcolumn/Σtotal
the number of females whose first choice was expected to be water =29*46/90=14.82
3)
X2 test statistic.=Σ(fo-fe)^2/fe =0.1636 + 0.5640 + 1.0304 + 0.6367 +0.1565 + 0.5395 + 0.9856 + 0.6091 = 4.686
4)
DF=(row-1)(column-1)=1*3=3
chi square critical value=7.81473 [excel formula =chisq.inv.rt(0.05,3)
since, test stat <critical value, do not reject Ho
hence, answer is
Fail to reject H0. There is not enough evidence to support the claim that gender and beverage are related