In: Statistics and Probability
You are a social psychologist interested in the adjustment of international students who go to school at American universities. You decide to examine whether there are differences in scores on the Acceptability by Others Scale (AOS). The AOS is a scale that ranges from 1 to 20 (1 = feeling completely isolated – 20 feeling completely accepted) that measures acceptance within the college community. You collect scores on the AOS from a sample of international students (Int) at a small college in the United States and another sample of United States (U.S.) citizens attending the same college. The data is provided below. Conduct a two-tailed independent-samples t-test by hand using an alpha level of .05 to determine whether there are differences in AOS scores between the two samples. Record your answers below.
International Students |
U.S. Students |
10 |
17 |
10 |
13 |
9 |
19 |
18 |
10 |
6 |
14 |
4 |
16 |
4 |
20 |
9 |
20 |
20 |
13 |
20 |
19 |
M = |
||||||||
μ = |
||||||||
df = |
||||||||
sM = |
||||||||
t = |
||||||||
d = |
6. Step #6: Interpret the results of the statistical test in terms of the research question
Hypothesis test:
Data:
n1 = 10
n2 = 10
x1-bar = 11
x2-bar = 16.1
s1 = 6.1824
s2 = 3.4785
Hypotheses:
Ho: μ1 = μ2
Ha: μ1 ≠ μ2
Decision Rule:
α = 0.05
Degrees of freedom = 10 + 10 - 2 = 18
Lower Critical t- score = -2.100922037
Upper Critical t- score = 2.100922037
Reject Ho if |t| > 2.100922037
Test Statistic:
Pooled SD, s = √[{(n1 - 1) s1^2 + (n2 - 1) s2^2} / (n1 + n2 - 2)] = √(((10 - 1) * 6.1824^2 + (10 - 1) * 3.4785^2)/(10 + 10 -2)) = 5.016
SE = s * √{(1 /n1) + (1 /n2)} = 5.01607575750207 * √((1/10) + (1/10)) = 2.243257275
t = (x1-bar -x2-bar)/SE = -2.273479755
p- value = 0.035478669
Decision (in terms of the hypotheses):
Since 2.273479755 > 2.100922037 we reject Ho
Conclusion (in terms of the problem):
There is sufficient evidence that there are differences in AOS scores between the two samples
Confidence interval:
n1 = 10
n2 = 10
x1-bar = 11
x2-bar = 16.1
s1 = 6.1824
s2 = 3.4785
% = 95
Degrees of freedom = n1 + n2 - 2 = 10 + 10 -2 = 18
Pooled s = √(((n1 - 1) * s1^2 + (n2 - 1) * s2^2)/DOF) = √(((10 - 1) * 6.1824^2 + ( 10 - 1) * 3.4785^2)/(10 + 10 -2)) = 5.016075758
SE = Pooled s * √((1/n1) + (1/n2)) = 5.01607575750207 * √((1/10) + (1/10)) = 2.243257275
t- score = 2.100922037
Width of the confidence interval = t * SE = 2.10092203686118 * 2.24325727481268 = 4.712908643
Lower Limit of the confidence interval = (x1-bar - x2-bar) - width = -5.1 - 4.71290864300311 = -9.812908643
Upper Limit of the confidence interval = (x1-bar - x2-bar) + width = -5.1 + 4.71290864300311 = -0.387091357
The 95% confidence interval is [-9.813, -0.387]