Tom K. answered 11/07/20
Knowledgeable and Friendly Math and Statistics Tutor
We are trying to see if the 2010 values are different from the 2001 values when we know the true 2001 values and have a sample of the 2010 values.
Thus, the first choice in a is the solution; we are testing if the 2010 value is significantly different from the 2001 values.
b - c We get the value in all 8 cells of the table by multiplying the percentages in the top table by 1019, the sample size from the 2010 study.
Values in column 2 are 387.22; 163.04; 407.6; 61.14
Values in column 3 are 377.03; 122.28; 458.55; 61.14;
d The value of the test statistic is calculated by, in each row of the b-c table, calculating (actual - expected)^2/expected, and adding these 4 values. For example, from the first row, we get (387.22 - 377.03)^2/377.03 = .2754 (values are calculated from exact values).
The values are 0.28; 13.59; 5.66; 0
The sum is 19.52
Since we have 4 groups, df = g - 1 = 3
If we use the rounded 19.52, the p-value is, from Excel, =CHISQ.DIST.RT(19.52,3) = .0002
When not stated, one assumes that α =.05
p < α, so we reject the null hypothesis that the 2010 population follow the 2001 distribution.
a is the answer.
Sierra S.
Thank you so much for your help!11/07/20