In: Statistics and Probability
HW9#8
Assume that the paired data came from a population that is normally distributed. Using a 0.05 significance level and
d=x−y,
find d overbard, s Subscript dsd,
the t test statistic, and the critical values to test the claim that μd=0.
x 10 14 6 4 7 11 16 6
y 11 11 9 9 9 12 11 7
over score d= (round three decimal places)
Sd= (Round three decimal places)
t= (round three decimal places)
Ta/2=pluse sign with a bar under it (round to three decimal places)
Sample #1 | Sample #2 | difference , Di =sample1-sample2 | (Di - Dbar)² |
10 | 11 | -1 | 0.140625 |
14 | 11 | 3 | 13.140625 |
6 | 9 | -3 | 5.640625 |
4 | 9 | -5 | 19.140625 |
7 | 9 | -2 | 1.890625 |
11 | 12 | -1 | 0.1406 |
16 | 11 | 5 | 31.6406 |
6 | 7 | -1.0000 | 0.1406 |
sample 1 | sample 2 | Di | (Di - Dbar)² | |
sum = | 74 | 79 | -5 | 71.875 |
mean of difference , D̅ =ΣDi / n
= -0.625
std dev of difference , Sd = √ [ (Di-Dbar)²/(n-1)
= 3.204
Ho : µd= 0
Ha : µd ╪ 0
std error , SE = Sd / √n = 3.2043 /
√ 8 = 1.1329
t-statistic = (D̅ - µd)/SE = ( -0.625
- 0 ) / 1.1329
= -0.552
Degree of freedom, DF= n - 1 =
7
t-critical value , t* = ±
2.365 [excel function: =t.inv.2t(α,df) ]
since, test stat=-0.552 >-2.365, do not reject Ho
There is not enough evidence to reject the claim