Goodness of Fit Test

Example 1. 04 Handout 1 - Page 6 of 8

Researchers have conducted a survey of 1,600 coffee drinkers

asking how much coffee they drink in order to confirm previous studies.
Previous studies have indicated that 72 % of Americans drink coffee. Below
are the result of previous studies( left ) and the survey ( right ). At α = 0.05,
is there enough evidence to conclude that the distributions are the same?

Previous studies
Response % of Coffee Drinkers
2 cups per week 15%
1 cup per week 13%
1 cup per day 27%
2+ cups per day 45%

Survey
Response Frequnecy
2 cups per week 206
1 cup per week 193
1 cup per day 462
2+ cups per day 739

Step 3:
a. The null hypothesis Ho: the population frequencies are equal to
the expected frequencies.
b. The alternative hypothesis Ha: the null hypothesis is false.
c. Level of the significance: α = 0.05
d. The degree of freedom: k -1 = 4 – 1 = 3
 e. The test statistic can be calculated using the table below:
% of (O - E¿2
Expected Observe O-E (O -E¿ 2
Respons coffee E
e Drinkers E d
O
2 cups 15% 0.15x1600 206 206-240 (-34¿2= 1156/240
per week = 240 = - 34 1156 = 4.817
1 cup per 13% 0.13x1600 193 193-208 (-15¿ = 225/208
2

week = 208 = -15 225 = 1.082

1 cup per 27% 0.27x1600 462 462-432 (30¿ = 2
900/432
day = 432 = 30 900 = 2.083
2+ cups 45% 0.45x1600 739 739-720 (19¿ = 361/720
2

per day = 720 = 19 361 = 0.5014

8.483

Therefore:
( observed – expected ¿2 (O -E¿2
Test Statistic, X 2 = ⅀ ------------------------------- = ⅀ ----------------
Expected E
X = ⅀ (4.1817 + 1.082 + 2.083 + 0.5014 )
2

X = 8.483
2

f. From α = 0.05 and k = 3, the critical value is 7.815. ( see Chi-square

Distribution Table, 04 Handout 2 )
g. Is there enough evidence to reject Ho?
Since X 2 ≈ 8.483 > 7.815, there is enough statistical evidence to
reject the null hypothesis and to believe that the old percentage no
longer hold.