In: Statistics and Probability
Do a two-sample test for equality of means assuming unequal
variances. Calculate the p-value using Excel.
(a-1) Comparison of GPA for randomly chosen
college juniors and seniors:
x⎯⎯1x¯1 = 4.75, s1 = .20,
n1 = 15, x⎯⎯2x¯2 = 5.18, s2
= .30, n2 = 15, α = .025, left-tailed
test.
(Negative values should be indicated by a minus sign. Round
down your d.f. answer to the nearest whole number and
other answers to 4 decimal places. Do not use "quick" rules for
degrees of freedom.)
d.f.
t-calculated
p-value
t-critical
b-1) Comparison of average commute miles for
randomly chosen students at two community colleges:
x⎯⎯1x¯1 = 25, s1 = 5, n1 =
22, x⎯⎯2x¯2 = 33, s2 = 7,
n2 = 19, α = .05, two-tailed
test.
(Negative values should be indicated by a minus sign. Round
down your d.f. answer to the nearest whole number and
other answers to 4 decimal places. Do not use "quick" rules for
degrees of freedom.)
d.f.
t-calculated
p-value
t-critical
(c-1) Comparison of credits at time of
graduation for randomly chosen accounting and economics
students:
x⎯⎯1x¯1 = 150, s1 = 2.8, n1
= 12, x⎯⎯2x¯2 = 143, s2 = 2.7,
n2 = 17, α = .05, right-tailed
test.
(Negative values should be indicated by a minus sign. Round
down your d.f. answer to the nearest whole number and
other answers to 4 decimal places. Do not use "quick" rules for
degrees of freedom.)
d.f.
t-calculated
p-value
t-critical
1) df = (s1^2/n1 + s2^2/n2)^2/((s1^2/n1)^2/(n1 - 1) + (s2^2/n2)^2/(n2 - 1))
= ((0.2)^2/15 + (0.3)^2/15)^2/(((0.2)^2/15)^2/14 + ((0.3)^2/15)^2/14)
= 24
The test statistic t = ()/sqrt(s1^2/n1 + s2^2/n2)
= (4.75 - 5.18)/sqrt((0.2)^2/15 + (0.3)^2/15)
= -4.6189
P-value = P(T < -4.6189)
= 0.0001
At alpha = 0.025, the critical value is t* = -2.064
b) df = (s1^2/n1 + s2^2/n2)^2/((s1^2/n1)^2/(n1 - 1) + (s2^2/n2)^2/(n2 - 1))
= (5^2/22 + 7^2/19)^2/((5^2/22)^2/21 + (7^2/19)^2/18)
= 32
The test statistic t = ()/sqrt(s1^2/n1 + s2^2/n2)
= (25 - 33)/sqrt(5^2/22 + 7^2/19)
= -4.1504
P-value = 2 * P(T < -4.1504)
=2 * 0.0001 = 0.0002
At alpha = 0.05, the critical values are t* = +/- 2.037
c) df = (s1^2/n1 + s2^2/n2)^2/((s1^2/n1)^2/(n1 - 1) + (s2^2/n2)^2/(n2 - 1))
= ((2.8)^2/12 + (2.7)^2/17)^2/(((2.8)^2/12)^2/11 + ((2.7)^2/17)^2/16)
= 23
The test statistic t = ()/sqrt(s1^2/n1 + s2^2/n2)
= (150 - 143)/sqrt((2.8)^2/12 + (2.7)^2/17)
= 6.7290
P-value = P(T > 6.7290)
= 1 - P(T < 6.7290)
= 1 - 1 = 0
At alpha = 0.05, the critical value is t* = 1.714