Subjects are tested for reactions times with their left and right hands.

(Only right-handed subjects were u

results (in thousandths of a second) are given in the accompanying table. Use a 0.05 significance level
claim that there is a difference between the mean of the right- and left-hand reaction times. If an engineer
a fighter-jet cockpit and must locate the ejection-seat activator to be accessible to either the right or the lef
it make a difference which hand she chooses?
Subject Right Left
A 191 224
B 97 171
C 116 191
D 165 207
E 116 196
F 129 165
G 171 177
H 155 165
I 112 140
J 102 188
K 188 155
L 158 219
M 121 177
N 133 174

t-Test: Paired Two Sample for Means

Right Left
Mean 139.571428571429 182.071428571429
Variance 984.417582417583 567.456043956045
Observations 14 14
Pearson Correlation 0.302012136956483
Hypothesized Mean Differenc 0
df 13
t Stat -4.79372485441102
P(T<=t) one-tail 0.00017538324794222
t Critical one-tail 1.77093339598687
P(T<=t) two-tail 0.00035076649588444
t Critical two-tail 2.16036865646279
 nds. (Only right-handed subjects were used.) The
ying table. Use a 0.05 significance level to test the
left-hand reaction times. If an engineer is designing
be accessible to either the right or the left hand, does
nd she chooses?

Conclusion/Interpretation:
Because the test statistic does fall in the critical region, we reject the null hypothesis. There is sufficient evidence to support t
Because there does appear to be such a difference, an engineer designing a fighter-jet cockpit should locate the ejection-se
s sufficient evidence to support the claim of a difference between the right- and left-hand reaction times.
kpit should locate the ejection-seat activator so that it is readily accessible to the faster hand, which appears to be the right hand with see
s to be the right hand with seemingly lower reaction times. (We could require special training for left-handed pilots if a similar test of left-
d pilots if a similar test of left-handed pilots show that their dominant hand is faster.)
 Example: Testing the Difference Between Two Population Means (Paired Samples
A public school is considering the revision of its Reading course. The school believes
that the revised course will improve the reading comprehension skills of its students.
The school has decided to conduct an experiment to evaluate the revised course before
it is offered to the general student body. Pairs of students were formed based on their
grades in the Reading course the previous semester.

A random sample of 10 pairs was then selected. A randomization process was again used to
determine which student in the pair would be taught using the revised course. At the end of the
course, all the students in the sample were given the same exam. Based on the scores below,
can it be concluded at a=0.05 level of significance that the revised course has improved the
reading comprehension of the students?

Revised
Pair Existing Course
Course
1 211 221 H_0: mu1=mu2
2 231 216 H_A: mu2>u1
3 191 203
4 216 224
5 207 201
6 203 178
7 201 188
8 179 159
9 179 177
10 211 197

t-Test: Paired Two Sample for Means Conclusion:

Existing Course Revised Course

Because the test statistic does not fall in the criti
Mean 202.9 196.4 null hypothesis. There is no sufficient evidence
Variance 266.322222222222 444.4888888889 revised course has improved the reading com
Observatio 10 10
Pearson Co 0.784230190284965
Hypothesiz 0
df 9
t Stat 1.57109975188711
P(T<=t) one0.075303909656649
t Critical o 1.83311293265624
P(T<=t) two0.150607819313297
t Critical t 2.26215716279821
Paired Samples)

was again used to

At the end of the
he scores below,
s improved the

atistic does not fall in the critical region, we fail to reject the
here is no sufficient evidence to support the claim that the
as improved the reading comprehension of the students.
A pharmaceutical company wants to compare the effectiveness of two drugs, Drug A and Drug B, in lower
to one of the drugs, and their reductions in blood pressure (in mmHg) are re
Drug A Drug B
15.1 11.2
12.8 10.5
13.4 9.8
11.9 12.1
10.7 9.6
14.3 11.8
12 10.3
13.5 11
11.1 8.9
12.6 10.4
13.7 9.5
14.8 12.2
15 10
10.9 9.7
13.3 10.9
14.2 9.6
12.9 11.7
11.4 10.2
13.8 10.8
14.5 12.3

Is there evidence at the α=0.05 significance level to conclude that Drug A is more effective than Drug B in

H_0: mu1=mu2
H_A: mu1 > mu2

t-Test: Two-Sample Assuming Equal Variances

Drug A Drug B Conclusion:

Mean 13.095 10.625
Variance 1.935237 1.010395
Because the test statistic does fall in the cr
Observations 20 20
null hypothesis. There is sufficient evide
Pooled Variance 1.472816 that there Drug A is more effecive than D
Hypothesized Mean Difference 0 pressure.
df 38
t Stat 6.436099
P(T<=t) one-tail 7.193E-08
t Critical one-tail 1.685954
P(T<=t) two-tail 1.439E-07
t Critical two-tail 2.024394
 and Drug B, in lowering blood pressure. Patients are randomly assigned
sure (in mmHg) are recorded after 8 weeks.

ective than Drug B in lowering blood pressure?

atistic does fall in the critical region, we reject the

There is sufficient evidence to support the claim
A is more effecive than Drug B in lowering blood
pressure.
 Example: Testing the Difference Between Two Population Means (Independent Samples – Un
A company manufactures two types of paint: Type A and Type B. The company wants to compare the
average drying times of the two types of paint under normal conditions. They randomly selected two
groups of painters to test each type of paint. The drying times (in minutes) for each group are recorded
as follows:
Type A Type B
25 35
27 32
30 30
24 38
28 36
26 42
32 45
29 37
27 39
31 44
You want to test whether there is a significant difference in the mean drying times between Type A and
Type B at the 0.05 significance level.

H_0: mu1=mu2
H_A: mu1 \neq mu2

t-Test: Two-Sample Assuming Unequal Variances

Type A Type B Conclusion:

Mean 27.9 37.8
Variance 6.766666667 23.95556
Because the test statistic does fall in the c
Observations 10 10
null hypothesis. There is sufficient evide
Hypothesized Mean Difference 0 that there is a significant difference in
df 14 between Type A and Type B at the 0
t Stat -5.64818459
P(T<=t) one-tail 3.00546E-05
Type A, with a mean drying time of 27
t Critical one-tail 1.761310136 compared to Type B, which dries for an
P(T<=t) two-tail 6.01091E-05
t Critical two-tail 2.144786688
endent Samples – Unequal Variances)
nts to compare the
mly selected two
roup are recorded

ween Type A and

atistic does fall in the critical region, we reject the

There is sufficient evidence to support the claim
significant difference in the mean drying times
e A and Type B at the 0.05 significance level.

mean drying time of 27.9 minutes, dries faster

e B, which dries for an average of 37.8 minutes.
 Sales Sales
Before After
Store
Campaign Campaign
(Units) (Units)
1 150 180
2 200 240
3 180 210
4 160 170
5 220 250
6 190 230
7 170 210
8 210 230
9 240 270
10 180 220
Store A sales: 58, 64, 67, 62, 59, 63
Store B sales: 70, 75, 72, 68, 77, 73

58 64 67 62 59 63
70 75 72 68 77 73