2 Statistics Lab
Bring a calculator.
Null hypothesis (H
0
): hypothesis of no effect, can be
falsified
.
Sample mean, standard deviation. See Table 2.1 (p. 13)
Use Student-t-test to compare samples with a normal (bell-shaped) distribution.
Use equation on page 11 to compute t
0
Compare with t
c
(Appendix 4, 2-tailed test with probability of 0.05).
If t
0
>= t
c
, then H
0
can be rejected.
Section A, Part 1
:
Measure student height (Item 1) and wingspan (Item 2).
H
0
is that there is no difference between the data pairs.
Section A, Part 2
:
Measure student height and wingspan.
Homogeneity of variance (s
2
) between 2 sample means. See Table 2.2 (p. 14)
Compare F
0
to F
c
from Appendix 5 (p. 204).
If F
0
>= F
c
, then H
0
(homogeneity) is rejected, Student-t-test cannot be used.
H
0
is that there is no difference between the 2 samples.
Section C
:
Rank test of male and female right arm lengths.
Compute U and U', compare to U
c
from Appendix 7 (p. 206).
Index
Lab notes made Aug 29, 2006 by
Peter Chen