You are on page 1of 42

Introductory Statistics Using SPSS 2nd

Edition Knapp Solutions Manual


Visit to Download in Full: https://testbankdeal.com/download/introductory-statistics-usi
ng-spss-2nd-edition-knapp-solutions-manual/
1

Chapter 7
Paired t Test and Wilcoxon Test
Solutions to All Exercises

Exercise Page Exercise Page


7.1A....................2 7.6A ................. 22
7.1B....................4 7.6B ................. 24
7.2A....................6 7.7A ................. 26
7.2B....................8 7.7B ................. 28
7.3A..................10 7.8A ................. 30
7.3B..................12 7.8B ................. 32
7.4A..................14 7.9A ................. 34
7.4B..................16 7.9B ................. 36
7.5A..................18 7.10A ............... 38
7.5B..................20 7.10B ............... 40

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
2

Exercise 7.1, Data Set A

(a)
H0: Acme Monster Spray has no effect on children’s monster anxiety.
H1: Acme Monster Spray reduces children’s monster anxiety.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
3

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 6.75 12 1.485 .429
posttest 4.08 12 1.443 .417

Paired Samples Test


Paired Differences
Std. 95% Confidence Interval of
Std. Error the Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest 2.667 1.923 .555 1.445 3.888 4.804 11 .001

Prior to using the Monster Spray, the children’s mean anxiety level was 6.75; after using
the spray, that level dropped to 4.08. This 2.67-point reduction in reported anxiety is
statistically significant since the p value of .001 is less than the specified  level of .05.
Based on these findings, we would reject H0, and not reject H1.

(d)
In order to help children rest easier at night, 12 children were asked to rate their
bedtime anxiety due to fear of nighttime monsters on a 1 to 10 scale (1 = not afraid at
all, 10 = very afraid), then Acme Monster Spray, an inert sweet-smelling mist was
administered by their parents, after which children were asked to rate their anxiety level.
Upon spraying, children reported a 2.67-point average drop in anxiety (6.75 before
spraying, down to 4.08 after spraying). Paired t test analysis revealed this change to be
statistically significant (p = .001) using a .05  level, suggesting that the (placebo) effect
of this spray may help children rest more comfortably.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
4

Exercise 7.1, Data Set B

(a)
H0: Acme Monster Spray has no effect on children’s monster anxiety.
H1: Acme Monster Spray reduces children’s monster anxiety.

(b) After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
5

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 5.70 10 1.947 .616
posttest 5.40 10 2.066 .653

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .300 .483 .153 -.046 .646 1.964 9 .081

Prior to using the Monster Spray, the children’s mean anxiety level was 5.70; after using
the spray, that level dropped to 5.40. This .30-point reduction in reported anxiety is not
statistically significant since the p value of .081 is greater than the specified  level of
.05. Based on these findings, we would not reject H0, and reject H1.

(d)
In order to help children rest easier at night, 10 children were asked to rate their
bedtime anxiety due to fear of nighttime monsters on a 1 to 10 scale (1 = not afraid at
all, 10 = very afraid), then Acme Monster Spray, an inert sweet-smelling mist was
administered by their parents, after which children were asked to rate their anxiety level.
Upon spraying, children reported a .30-point average drop in anxiety (5.70 before
spraying, down to 5.40 after spraying). Paired t test analysis revealed a p level of .081;
using a .05  level, this suggests that this difference in scores is not statistically
significant, hence, parents will need to find a different way to console their children at
nighttime.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
6

Exercise 7.2, Data Set A

(a)
H0: The heart health lecture will not change elevator/stair usage.
H1: The heart health lecture will change elevator/stair usage.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
7

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 2.56 16 1.209 .302
posttest 2.38 16 1.310 .328

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .188 2.007 .502 -.882 1.257 .374 15 .714

Prior to the heart health seminar, attendees reported that they took the stairs (as
opposed to the elevator) an average of 2.56 times per week; a week after the seminar,
that number (unexpectedly) dropped to 2.38, however, using an  level of .05, the
observed p value of .714 suggests that this difference is not statistically significant.

(d)
In an effort to increase healthy behaviors, a heart health seminar was offered to
voluntary participants. Prior to the presentation, each attendee was asked to write down
the number of times that he or she used the stairs instead of the elevator at work over
the course of the last week; a week after the seminar, the same data were gathered
from those who attended. Prior to the seminar, attendees reported using the stairs an
average of 2.56 times per week; a week after the seminar, that figure unexpectedly
dropped to 2.38 times per week. Despite this unexpected decline in stair usage, the p
value of .714 suggests that the mean weekly change in stair usage is not statistically
significant (using a traditional .05  level). Based on these findings, we are considering
retooling this seminar.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
8

Exercise 7.2, Data Set B

(a)
H0: The heart health lecture will not change elevator/stair usage.
H1: The heart health lecture will change elevator/stair usage.

(b) After computing the difference between the pretest score and posttest score (diff
= posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
9

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 4.15 13 2.703 .750
posttest 7.00 13 2.160 .599

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -2.846 4.100 1.137 -5.324 -.369 -2.503 12 .028

Prior to the heart health seminar, attendees reported that they took the stairs (as
opposed to the elevator) an average of 4.15 times per week; a week after the seminar,
that number increased to 7.00; using an  level of .05, the observed p value of .028
suggests that this difference is statistically significant. Based on these findings, I would
reject H0 and not reject H1.

(d)
In an effort to increase healthy behaviors, a heart health seminar was offered to
voluntary participants. Prior to the presentation, each attendee was asked to write down
the number of times that he or she used the stairs instead of the elevator at work over
the course of the last week; a week after the seminar, the same data were gathered
from those who attended. Prior to the seminar, attendees reported using the stairs an
average of 4.15 times per week; a week after the seminar, that figure increased to 7.00
times per week (% = 68.67); this increase is statistically significant (p = .028) using the
traditional .05  level. These findings suggest that the seminar was useful. It would be
interesting to see if these figures are sustainable over time.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
10

Exercise 7.3, Data Set A

(a)
H0: Students do not significantly improve their spelling skills over the course of a month.
H1: Students improve their spelling skills over the course of a month.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
11

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 62.05 20 9.041 2.022
posttest 69.55 20 9.478 2.119

Paired Samples Test


Paired Differences
95% Confidence Interval
Std. Std. Error of the Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -7.500 13.097 2.929 -13.629 -1.371 -2.561 19 .019

At the start of the month, students were administered a 100-word spelling test to find out
how many words on the prescribed spelling list for that month students could already
spell correctly. Students spelled an average of 62.05 words correctly on this pretest. At
the end of the month, students were readministered this 100-word test, and spelled an
average of 69.55 words correctly. The 7.5 increase in score constitutes a statistically
significant improvement since the p level is .019, which is less than the specified  level
of .05. Based on these findings, we would reject H0, and not reject H1.

(d)
In order to assess if students substantially advance their spelling proficiency on a
monthly basis, students were given a 100-word spelling test at the beginning of the
month; after studying 25 words per week from the list, students took the same 100-word
test 30 days later. At the end of the month, students spelled an average of 69.55 of the
words correctly, compared to 62.05 at the beginning of the month. Using an  level of
.05, this 7.5-point improvement in score was found to be statistically significant (p =
.019).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
12

Exercise 7.3, Data Set B

(a)
H0: Students do not significantly improve their spelling skills over the course of a month.
H1: Students improve their spelling skills over the course of a month.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
13

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 86.14 22 4.201 .896
posttest 87.59 22 4.469 .953

Paired Samples Test


Paired Differences
95% Confidence Interval of
Std. Std. Error the Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -1.455 6.185 1.319 -4.197 1.288 -1.103 21 .283

At the start of the month, students were administered a 100-word spelling test to find out
how many words on the prescribed spelling list for that month students could already
spell correctly. Students spelled an average of 86.14 words correctly on this pretest. At
the end of the month, students were readministered this 100-word test, and spelled an
average of 87.59 words correctly. The 1.45-point increase in score is not a statistically
significant improvement since the p level is .283, which is greater than the specified 
level of .05. Based on these findings, we would not reject H0, and reject H1.

(d)
In order to find out if students substantially advance their spelling proficiency on a
monthly basis, students were given a 100-word spelling test at the beginning of the
month; after studying 25 words per week from the list, students took the same 100-word
test 30 days later. At the end of the month, students spelled an average of 87.59 of the
words correctly, compared to 86.14 at the beginning of the month. Using an  level of
.05, this 1.45-point improvement in score is not statistically significant (p = 283).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
14

Exercise 7.4, Data Set A

(a)
H0: Short-term therapy does not reduce depression.
H1: Short-term therapy reduces depression.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff despite the
slightly skewed distribution, thus satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
15

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 53.83 18 8.893 2.096
posttest 52.44 18 10.782 2.541

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest 1.389 4.217 .994 -.708 3.486 1.397 17 .180

Depression levels were measured before and after administering short-term therapy
using the Acme Depression Inventory; the mean pretest score was 53.83, whereas the
mean posttest score dropped to 52.44. However, the p value is .180, which is greater
than the .05  level; hence, this reduction in depression is not considered to be
significant. Based on this, I would not reject H0 and reject H1.

(d)
In order to determine the effectiveness of a short-term therapeutic implementation, we
administered the Acme Depression Inventory (ADI), which renders a score from 0 to 75
(0 = low depression, 75 = high depression) before and after treatment for a group of 18
participants. The mean level of the depression dropped from 53.83 to 52.44, however
this 1.39-point change is not considered to be statistically significant (p = .180;  = .05).
Based on these findings, we will be reassessing our therapeutic methods in terms of
treatment approach and duration of care.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
16

Exercise 7.4, Data Set B

(a)
H0: Short-term therapy does not reduce depression.
H1: Short-term therapy reduces depression.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
17

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 69.80 15 2.007 .518
posttest 67.20 15 3.968 1.024

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest 2.600 3.247 .838 .802 4.398 3.101 14 .008

Depression levels were measured before and after administering short-term therapy
using the Acme Depression Inventory; the mean pretest score was 69.80, whereas the
mean posttest score dropped to 67.20. The p value is .008, which is less than the .05 
level; hence, this reduction in depression is considered to be significant. Based on this, I
would reject H0 and not reject H1.

(d)
In order to determine the effectiveness of a short-term therapeutic implementation, we
administered the Acme Depression Inventory (ADI), which renders a score from 0 to 75
(0 = low depression, 75 = high depression) before and after treatment for a group of 15
participants. The mean level of the depression dropped from 69.80 to 67.20; this 2.6-
point change (% = -3.72) is considered to be statistically significant (p = .008;  = .05).
These findings suggest that this short-term therapy model facilitates an effective
reduction in depressive symptoms among these individuals.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
18

Exercise 7.5, Data Set A

(a)
H0: A 15-minute individual session with a bowling coach has no effect on bowling
scores.
H1: A 15-minute individual session with a bowling coach enhances bowling scores.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
19

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 151.60 15 15.459 3.991
posttest 157.73 15 13.936 3.598

Paired Samples Test


Paired Differences
95% Confidence Interval
Std. Std. Error of the Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -6.133 6.512 1.681 -9.740 -2.527 -3.648 14 .003

Fifteen students bowled an initial game and scored a average of 151.60, then a coach
provided 15 minutes of instructions based on observations made during the student’s
first game. Student’s average score in the second game was 157.73. This 6.13-point
increase produced a p level of .003; since this is less than the  level of .05, this is
considered to be a statistically significant difference. Based on these findings, we would
reject H0, and not reject H1.

(d)
To assess the effectiveness of short-term coaching, 15 students bowled one game with
a coach observing unobtrusively. Next, the coach provided 15 minutes of individual
coaching, after which, each student bowled a second game. On the average, scores
were 6.13 points higher in the second game (157.73, up from 151.60 in their first game).
This finding is considered to be statistically significant (p = .003,  = .05). Based on
these findings, we will continue to use this coaching method for bowling, and we will be
evaluating the utility of providing this style of coaching for other sports.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
20

Exercise 7.5, Data Set B

(a)
H0: A 15-minute individual session with a bowling coach has no effect on bowling
scores.
H1: A 15-minute individual session with a bowling coach enhances bowling scores.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
21

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 72.50 8 3.742 1.323
posttest 74.25 8 3.919 1.386

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -1.750 4.773 1.688 -5.741 2.241 -1.037 7 .334

Eight students bowled an initial game and scored a average of 72.50, then a coach
provided 15 minutes of instructions based on observations made during the student’s
first game. Student’s average score in the second game was 74.25. This 1.75-point
increase produced a p level of .334; since this is greater than the  level of .05, this is
not considered to be a statistically significant difference. Based on these findings, we
would not reject H0, and reject H1.

(d) To assess the effectiveness of short-term coaching, eight students bowled one
game with a coach observing unobtrusively. Next, the coach provided 15 minutes of
individual coaching, after which, the student bowled a second game. On the average,
students’ scores were 1.75 points higher in their second game (74.25, up from 72.50 in
their first game). This finding is not considered to be statistically significant (p = .334, 
= .05). Based on these findings, we are considering evaluating if this form of coaching
may provide more promising results for different individual sports (e.g., golf, archery,
weight training).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
22

Exercise 7.6, Data Set A

(a)
H0: Using a laptop computer does not enhance grades.
H1: Using a laptop computer enhances grades.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
23

(c)
The paired t test revealed the following:

Paired Samples Statistics


Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 67.50 20 18.936 4.234
posttest 76.15 20 16.024 3.583

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
-8.650 24.528 5.485 -20.129 2.829 -1.577 19 .131
Pair 1 pretest - posttest

Grades (0%-100%) were gathered on 20 students for the semesters before and after
having been issued a laptop computer. Mean grade scores increased from 67.50 to
76.15; however, since the p value of .131 is greater than the .05  level, this modest
increase is not considered to be significantly different. Based on this, I would not reject
H0 and reject H1.

(d)
We issued laptop computers to 20 students to determine if having such a device would
impact academic performance. We compared their mean overall grades (0%-100%)
from the semester before to the semester after the laptops were issued; overall, mean
grade scores increased from 67.50 to 76.15, however using an  level of .05, this 8.65-
point increase was determined not to be statistically significant (p = .131).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
24

Exercise 7.6, Data Set B

(a)
H0: Using a laptop computer does not enhance grades.
H1: Using a laptop computer enhances grades.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
25

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 75.50 20 11.062 2.474
posttest 82.70 20 7.512 1.680

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -7.200 13.117 2.933 -13.339 -1.061 -2.455 19 .024

Grades (0%-100%) were gathered on 20 students for the semesters before and after
having been issued a laptop computer. Mean grade scores increased from 75.5 to 82.7,
with a p value of .024; since this is less than the .05  level, this increase is considered
to be significant. Based on this, I would reject H0 and not reject H1.

(d)
We issued laptop computers to 20 students to determine if having such a device would
impact academic performance. We compared their mean overall grades (0%-100%)
from the semester before to the semester after the laptops were issued; overall, mean
grade scores increased 7.2 points, from 75.5 to 82.7. Using an  level of .05, this 7.2-
point increase (% = 9.5) was determined to be a statistically significant improvement (p
= .024).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
26

Exercise 7.7, Data Set A

(a)
H0: Acme allergy medicine has no effect on sneezing.
H1: Acme allergy medicine reduces sneezing.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
27

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 15.39 18 3.274 .772
posttest 12.61 18 3.165 .746

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest 2.778 3.318 .782 1.128 4.428 3.552 17 .002

Eighteen people who were experiencing allergy symptoms were recruited and asked to
count their total number of sneezes for a day. The next day, they took the Acme allergy
medicine as directed, and counted their total number of sneezes for that (second) day.
This resulted in an average of 2.77 fewer sneezes (an average of 15.39 on the first day,
and 12.61 on the second day). Using a .05  level, this difference is considered to be
statistically significant. As such, we would reject H0, and not reject H1.

(d)
To determine if Acme allergy medicine helps reduce sneezing among those
experiencing allergy symptoms, 18 participants were recruited and instructed to count
their total number of sneezes for 1 day. The next morning, each took the Acme allergy
medicine as directed, and counted their total sneezes for that day. On average,
participants sneezed a total of 15.39 times the day before taking the medication, and
12.61 times after taking the medication; an average of 2.77 fewer sneezes. This
difference is considered to be statistically significant (p = .002,  = .05), suggesting that
the medicine provided some allergy symptom relief.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
28

Exercise 7.7, Data Set B

(a)
H0: Acme allergy medicine has no effect on sneezing.
H1: Acme allergy medicine reduces sneezing.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
29

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 4.10 21 1.640 .358
posttest 3.95 21 1.830 .399

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .143 2.798 .611 -1.131 1.416 .234 20 .817

Twenty-one people who were experiencing allergy symptoms were recruited and asked
to count their total number of sneezes for a day. The next day, they took the Acme
allergy medicine as directed, and counted their total number of sneezes for that
(second) day. This resulted in an average of .15 fewer sneezes (an average of 4.10 on
the first day, and 3.95 on the second day). This rendered a p of .817; since the p value
is greater than the designated .05  level, this difference is not considered to be
statistically significant. As such, we would not reject H0, and reject H1.

(d)
To determine if Acme allergy medicine helps reduce sneezing among those
experiencing allergy symptoms, 21 participants were recruited and instructed to count
their total number of sneezes for 1 day. The next morning, they took the Acme allergy
medicine as directed, and counted their total sneezes for that day. On average,
participants sneezed a total of 4.10 times the day before taking the medication, and
3.95 times after taking the medication, amounting to an average of .15 fewer sneezes.
This reduction is not considered to be statistically significant (p = .817,  = .05),
suggesting that this medicine was not particularly helpful for these individuals.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
30

Exercise 7.8, Data Set A

(a)
H0: Tutsle root tea does not lower fever.
H1: Tutsle root tea lowers fever.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
31

(1c) The paired t test revealed the following:

Paired Samples Statistics


Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 100.893 100 .7727 .0773
posttest 100.782 100 .6911 .0691

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .1110 1.1094 .1109 -.1091 .3311 1.001 99 .319

Body temperatures were gathered on 100 patients who had fever before and after
giving each a cup of Tutsle root tea. The mean pre-tea temperature was 100.893; after
the tea, the mean temperature dropped .111 to 100.782, however, since the p value of
.319 is greater than the .05  level, we would say that the Tutsle root tea did not make a
statistically significant impact in terms of reducing temperature in these patients. Based
on this, I would not reject H0 and reject H1.

(d)
To determine if Tutsle root tea had any viable medicinal value when it comes to
reducing fever, we recruited 100 patients who had fever. We recorded their body
temperature before and 1 hour after having a cup of this tea. Mean temperature went
from 100.893 down to 100.782, however this .111 decrease is not considered to be
statistically significant (p = .319;  = .05).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
32

Exercise 7.8, Data Set B

(a)
H0: Tutsle root tea does not lower fever.
H1: Tutsle root tea lowers fever.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
33

(c)
The paired t test revealed the following:

Paired Samples Statistics


Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 100.817 150 .8229 .0672
posttest 100.593 150 .7443 .0608

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .2240 1.0644 .0869 .0523 .3957 2.577 149 .011

Body temperatures were gathered on 100 patients who had fever before and after
giving each a cup of Tutsle root tea. The mean pre-tea temperature was 100.817; after
the tea, the mean temperature dropped .224 to 100.593. Since the p value of .011 is
less than the .05  level, we would say that the Tutsle root tea made a statistically
significant impact in terms of reducing temperature in these patients. Based on this, I
would reject H0 and not reject H1.

(d)
To determine if Tutsle root tea had any viable medicinal value when it comes to
reducing fever, we recruited 100 patients who had fever. We recorded their body
temperature before and 1 hour after having a cup of this tea. Mean temperature was
reduced .224, from 100.817 down to 100.593. This finding is considered to be
statistically significant (p = .011  = .05).

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
34

Exercise 7.9, Data Set A

(a)
H0: Chocolate has no effect on attitude.
H1: Chocolate enhances attitude.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
35

(c)
The paired t test revealed the following:
Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 81.91 75 5.707 .659
posttest 80.85 75 5.897 .681

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest 1.053 9.078 1.048 -1.035 3.142 1.005 74 .318

Seventy-five people were recruited to determine if chocolate helps enhance attitude. We


began by administering the Acme Attitude Survey (AAS) (0 = very bad attitude, 100 =
very good attitude) resulting in an average score of 81.91. After eating a piece of
chocolate fudge cake, participants answered the AAS again; on average, the results
dropped by .96 points to 80.85. This change in the AAS score produced a p level of
.318, which using a .05  level is not statistically significant. This suggests that the
chocolate cake did not make a statistically significant impact on attitude, hence, we
would not reject H0, and reject H1.

(d)
To better comprehend the effects that chocolate might have on attitude, 75 participants
were recruited; we began by administering the Acme Attitude Survey (AAS) which
renders a score ranging from 0 to 100 (0 = very bad attitude, 100 = very good attitude).
Next, each participant was served a generous slice of chocolate fudge cake. After the
cake, participants were asked to complete a second AAS. The pretest revealed a mean
attitude score of 81.91; contrary to our expectations, the post-chocolate AAS score
dropped an average of .96 points, to 80.85, however, this rendered a p value of .318.
Using a .05  level, this drop in the AAS is not considered to be statistically significant.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
36

Exercise 7.9, Data Set B

(a)
H0: Chocolate has no effect on attitude.
H1: Chocolate enhances attitude.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
37

(c) The paired t test revealed the following:


Paired Samples Statistics
Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 81.17 75 5.869 .678
posttest 83.27 75 6.267 .724

Paired Samples Test


Paired Differences
95% Confidence
Std. Interval of the
Std. Error Difference
Mean Deviation Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -2.093 8.056 .930 -3.947 -.240 -2.250 74 .027

Seventy-five people were recruited to determine if chocolate helps enhance attitude. We


began by administering the Acme Attitude Survey (AAS) (0 = very bad attitude, 100 =
very good attitude) resulting in an average score of 81.17. After eating a piece of
chocolate fudge cake, participants answered the AAS again; on average, the results
rose by 2.10 points to 83.27. This change in the AAS score produced a p level of .027,
which, using a .05  level, is considered to be statistically significant. This suggests that
the chocolate cake made a statistically significant impact on attitude; hence, we would
reject H0, and not reject H1.

(d)
To better comprehend the effects that chocolate might have on attitude, 75 participants
were recruited. We began by administering the Acme Attitude Survey (AAS) which
renders a score ranging from 0 to 100 (0 = very bad attitude, 100 = very good attitude).
After collecting the AAS forms, each participant was served a generous slice of
chocolate fudge cake. After the cake, participants were asked to complete a second
AAS. The pretest revealed a mean attitude score of 81.17, and after dining on the
chocolate cake, the mean AAS score went up an average of 2.10 points, to 83.27.
These figures rendered a p value of .027. Using a .05  level, this increase in the AAS
score suggests that chocolate may help enhance attitude.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
38

Exercise 7.10, Data Set A

(a)
H0: The speech will not enhance voter’s likelihood to vote for the candidate.
H1: The speech will enhance voter’s likelihood to vote for the candidate.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
39

(1c) The paired t test revealed the following:

Paired Samples Statistics


Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 3.68 53 1.578 .217
posttest 4.47 53 1.908 .262

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest -.792 2.522 .346 -1.488 -.097 -2.288 52 .026

Prior to listening to a candidate’s speech, attendees were asked to rate their voting
intentions using a 7-point scale (1 = absolutely won’t vote for the candidate, 7 =
absolutely will vote for the candidate). After the speech, the survey is readministered to
determine the impact (if any) of the speech. The pretest score was 3.68, and the
posttest score was 4.47, with a p value of .026. Using the traditional .05  level, this
suggests that the speech facilitated a statistically significant (positive) impact. Based on
this, I would reject H0 and not reject H1.

(d)
To gauge the impact of a political speech, attendees were asked to rate their voting
intentions prior to the speech (1 = absolutely won’t vote for the candidate, 7 = absolutely
will vote for the candidate); at the conclusion of the speech, attendees were asked to
respond (again) to the same question. The scores increased by 21.47%, from 3.68 to
4.47. The p value of .026 is less than the  level of .05 suggesting that this speech
made a statistically significant impact. Per these findings, we will continue to use this
message in the campaign efforts.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
40

Exercise 7.10, Data Set B

(a)
H0: The speech will not enhance voter’s likelihood to vote for the candidate.
H1: The speech will enhance voter’s likelihood to vote for the candidate.

(b)
After computing the difference between the pretest score and posttest score (diff =
posttest - pretest), a histogram with normal curve was plotted for this difference (diff).
The graph below presents a symmetrical (bell-shaped) normal curve for diff, thus
satisfying this criterion.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.
41

(c)
The paired t test revealed the following:

Paired Samples Statistics


Mean N Std. Deviation Std. Error Mean
Pair 1 pretest 3.96 53 1.467 .202
posttest 3.94 53 1.549 .213

Paired Samples Test


Paired Differences
95% Confidence Interval of the
Difference
Mean Std. Deviation Std. Error Mean Lower Upper t df Sig. (2-tailed)
Pair 1 pretest - posttest .019 2.325 .319 -.622 .660 .059 52 .953

Prior to listening to a candidate’s speech, attendees were asked to rate their voting
intentions using a 7-point scale (1 = absolutely won’t vote for the candidate, 7 =
absolutely will vote for the candidate). After the speech, the survey is readministered to
determine the impact (if any) of the speech. The pretest score was 3.96, and the
posttest score was 3.94, with a p value of .953. Using the traditional .05  level, this
suggests that although the speech had a negative impact, it was not statistically
significant. Based on this, I would not reject H0 and reject H1.

(d)
To gauge the impact of a political speech, attendees were asked to rate their voting
intentions prior to the speech (1 = absolutely won’t vote for the candidate, 7 = absolutely
will vote for the candidate); at the conclusion of the speech, attendees were asked to
respond (again) to the same question. The scores (unexpectedly) dropped by .02 points
(% = -.51) from 3.96 to 3.94. The p value of .953 is greater than the  level of .05,
suggesting that while this speech appeared to have lowered voter’s opinions of the
candidate, this reduction is not considered to be statistically significant. Per these
findings, it is clear that the speech needs to be retooled to impart a more (positively)
effective message.

Knapp, Introductory Statistics Using SPSS, Second Edition. © 2017, SAGE Publications.

You might also like