In: Statistics and Probability
In 1898, Hermon Bumpus studied a number of house sparrows that
were brought to the Anatomical Laboratory of Brown University after
an uncommonly severe winter storm. 35 birds survived and 24
perished.
In the spreadsheet, the weights of birds in grams (g) are
given.
Is there a significant difference between the weights of the
sparrows that perished and the weights of the sparrows that
survived?
Using a t-test of independent samples (let alpha = 0.05), answer
the following.
WEIGHT | STATUS |
24.50 | survived |
26.90 | survived |
26.90 | survived |
24.30 | survived |
24.10 | survived |
26.50 | survived |
24.60 | survived |
24.20 | survived |
23.60 | survived |
26.20 | survived |
26.20 | survived |
24.80 | survived |
25.40 | survived |
23.70 | survived |
25.70 | survived |
25.70 | survived |
26.30 | survived |
26.70 | survived |
23.90 | survived |
24.70 | survived |
28.00 | survived |
27.90 | survived |
25.90 | survived |
25.70 | survived |
26.60 | survived |
23.20 | survived |
25.70 | survived |
26.30 | survived |
24.30 | survived |
26.70 | survived |
24.90 | survived |
23.80 | survived |
25.60 | survived |
27.00 | survived |
24.70 | survived |
26.50 | perished |
26.10 | perished |
25.60 | perished |
25.90 | perished |
25.50 | perished |
27.60 | perished |
25.80 | perished |
24.90 | perished |
26.00 | perished |
26.50 | perished |
26.00 | perished |
27.10 | perished |
25.10 | perished |
26.00 | perished |
25.60 | perished |
25.00 | perished |
24.60 | perished |
25.00 | perished |
26.00 | perished |
28.30 | perished |
24.60 | perished |
27.50 | perished |
31.10 | perished |
28.30 | perished |
The null and alternative hypothesis:
; the true mean weight of sparrow that perished and survived are not different.
; the true mean wright of sparrow that perished and survived are different.
true mean weight of sparrow that perished
true mean weight of sparrow that survived
Based on the sample we have calculated the sample mean and sample variance as follows:
Perished | Survived | |
Sample mean | ||
Sample variance | ||
sample size |
To test this hypothesis we use a Two sample t-test with EQUAL VARIANCE.
The formula for test-statistic is:
; with
where Pooled variance
Calculation for :
Calculation for test-statistic:
The test-statistic is calculated as
Critical value:
,
Decision:
Since,
Conclusion:
At the data does provide enough evidence to support the alternative hypothesis, i.e.,
So we conclude that the true mean weight of sparrow that perished and survived is significantly different.