In: Statistics and Probability
Use the information below to determine whether. State the statistical hypotheses and report the P-value. Find the 95% confidence intervals.
Sample 1: N1=48, X1= 4.99, ss1= 13.2
Sample 2: N2= 32, X2= 5.83, ss2= 29.2
Hypothesis testing:
The null hypothesis, ho: there is no significant difference in the mean of sample 1 and 2. u1 =/= u2
An alternative hypothesis, ho: there is a significant difference in the mean of sample 1 and 2. u1=u2
SAMPLE 1 | SAMPLE 2 | |
n= | 48 | 32 |
mean= | 4.99 | 5.83 |
s= | 0.5300 | 0.9705 |
s^2/n | 0.0059 | 0.0294 |
test statistic, t = = (Xbar1 - Xbar2) /
sqrt(s1^2/n1 + s2^2/n2)
t = = (4.99 - 5.83) / sqrt(0.0059+0.0294)
t = -4.4709
df= (s1^2/n1 + s2^2/n2)^2 / ((s1^2/n1)^2/(n1-1) +
(s2^2/n2)^2/(n2-1))
df= (0.0059+0.0294)^2 / ( 0.0059^2/(48-1) +
0.0294^2/(32-1) )
df= 44
p-value = 2*(1-P(T<|t|)
p-value = 2*(1-P(T<abs(-4.4709))
p-value = T.DIST.2T(abs(-4.4709),44)
p-value = 5.4105E-05 = 0.00005
With t=-4.47, p<5% i reject ho and conclude that there is a significant difference in the mean of sample 1 and 2. u1=u2
95% confidence interval:
t(a/2, df) =t(0.05/2,44)
t(a/2, df) =t.INV.2T(0.05,44) =
t(a/2, df) 2.0154
CI = (xbar1 - xbar2) +- t(a/2,dF)*Sqrt(s1^2/n1 + s2^2/n2)
lower = (4.99-5.83)-2.0154*SQRT(0.0059+0.0294) = -1.2187
upper = (4.99-5.83)+2.0154*SQRT(0.0059+0.0294) = -0.4613