Chapter 1 - The Nature of Probability and Statistics

EXERCISE SET 1-1 14. This is a descriptive statistic.

1. Statistics is the science of conducting 15. This is a descriptive statistic since it is

studies to collect, organize, summarize, based on an online survey of 500 students.
analyze, and draw conclusions from data.
16. This is a descriptive statistic.
2. A variable is a characteristic or attribute
that can assume different values. 17. This is an inferential statistic since it is a
generalization made from data obtained from
3. In a census, the researchers collect data a sample.
from all subjects in the population.
18. Answers will vary.
4. A population consists of all subjects
under study while a sample is a subgroup of 19. Answers will vary.
the population.
EXERCISE SET 1-2
5. Descriptive statistics consists of the
collection, organization, summarization, and 1. Qualitative variables are variables that
presentation of data while inferential can be placed in distinct categories
statistics consists of generalizing from according to some characteristic or attribute
samples to populations, performing and cannot be ranked; while quantitative
estimations and hypothesis testing, variables are numerical in nature and can be
determining relationships among variables, ordered or counted.
and making predictions.
2. Discrete variables assume values that can
6. Probability is used in gambling, be counted while continuous variables are
insurance, and weather forecasting. variables obtained by being measured.
Theoretically, continuous variables can
7. Samples are used more than populations assume all values between any given two
both because populations are usually large values.
and because researchers are unable to use
every subject in the population. 3. Continuous variables need to be rounded
because of the limits of the measuring
8. A biased sample is a sample that is not device.
representative of the population.
4. There are four types of measurement
9. This is inferential because a levels used in statistics: nominal, ordinal,
generalization is being made about the interval, and ratio levels of measurement.
population.
5. Qualitative 6. Quantitative
10. This is inferential since a prediction is
being made about the future. 7. Quantitative 8. Quantitative

11. This is a descriptive statistic. 9. Qualitative 10. Quantitative

12. This is a descriptive statistic since it is 11. Qualitative 12. Continuous

based on the results of a survey of 2739
individuals. However, if an inference were 13. Continuous 14. Discrete
made to all pet owners, it would be an
inferential statistic. 15. Discrete 16. Continuous

13. This is an inferential statistic since a 17. 6.5-7.5 feet

generalization about the population is being
made. 18. 6.25-6.35 millimeters

1
 Chapter 1 - The Nature of Probability and Statistics

19. 142.5-143.5 miles 7. continued

pet owners who do not take their pets to
20. 3.245-3.255 pounds veterinarians. Answers will vary.

21. 25.75-25.85 yards 8. The population could be all adults in the

United States. A sample could be obtained
22. 18.5-19.5 quarts by checking the blood levels of calcium. Of
course, the researcher must be aware of the
23. Nominal 24. Interval privacy rights of the respondents. Answers
will vary.
25. Ratio 26. Ratio
9. The population could be all adults in the
27. Ordinal 28. Nominal United States who develop diabetes during
2012. The sample could be surveying
29. Ratio 30. Ratio patient records of these people to see if they
have been taking statins. Again, the privacy
EXERCISE SET 1-3 rights must be considered. Answers will
vary.
1. Data can be collected by using telephone
surveys, mail questionnaire surveys, 10. The population could consist of all daily
personal interview surveys, by taking a look temperatures in Boston since records were
at records, or by direct observation methods. kept, and a sample could be obtained by
looking at weather report statistics.
2. Sampling error is the difference between
a sample measure and a population measure. 11. Systematic 12. Stratified
Nonsampling error is the result of collecting
data incorrectly or selecting a biased sample. 13. Random 14. Cluster

3. Random numbers are used in sampling so 15. Cluster 16. Random

that every subject in the population has an
equal chance of being selected for a sample. EXERCISE SET 1-4
Random numbers can be generated by
computers or calculators; however, there are 1. In an observational study, the researcher
other ways of generating random numbers observes what it happening and tries to draw
such as using a random number table or conclusions based on the observations. In an
rolling dice. experimental study, the researcher
manipulates one of the variables and tries to
4. The four basic sampling methods are determine how this influences the variables.
random, systematic, stratified, and cluster
methods. 2. The three types of observational studies
are
5. The population could be all adults in the a. Cross-sectional studies in which the data
United States. A sample could have been are collected at one point in time
selected by a mail survey. Answers will b. Retrospective studies in which the data
vary. are collected from past records
c. Longitudinal studies in which the data are
6. The population could be all automobiles collected over a period of time
in the United States stolen during 2010. A
sample could be obtained by looking at 3. One advantage of an observational study
police records. Answers will vary. is that it can occur in a natural setting. In
addition, researchers can look at past
7. The population could be all pet owners in instances of statistics and draw conclusions
the United States. A sample could be from these situations. Another advantage is
obtained by looking at veterinarians' records. that the researcher can use variables, such as
However, these records would not include drugs, that he or she cannot manipulate.

2
 Chapter 1 - The Nature of Probability and Statistics

3. continued 11. Blinding is used to help eliminate the

One disadvantage is that since the variable placebo effect. Here the subjects are given a
cannot be manipulated, a definite cause-and- sugar pill that looks like the real medical
effect situation cannot be shown. Another pill. The subjects do not know which pill
disadvantage is that these studies can be they are getting. When double blinding
expensive and time-consuming. These occurs, neither the subjects nor the
studies can also be influenced by researchers are told who gets the real
confounding variables. Finally, in these treatment or the placebo.
studies, the researcher sometimes needs to
rely on data collected by others. 12. Randomized blocks are used to help
eliminate the Hawthorne effect and the
4. One advantage of an experimental study placebo effect. Also they help to eliminate
is that the researcher can decide how to confounding variables since randomly
select the subjects and assign them to the assigning subjects to groups tends to
specific groups. The researcher can also "balance out" the inconsistencies (such as
control the independent variable. One age, social class, etc.) that each of the
disadvantage of an experimental study is that subjects brings to the study.
it may not occur in a natural setting.
Another one is called the Hawthorne effect, 13. In a completely randomized design, the
and it is that the subject may change their subjects are assigned to the groups
natural behavior because they realize that randomly, whereas in a matched-pair design,
they are participating in a research study. subjects are matched on some variable.
Then one subject is randomly assigned to
5. In an experimental study, the researcher one group, and the other subject is assigned
has control of the assignment of subjects to to the other group. In both types of studies,
the groups whereas in a quasi-experimental the treatments can be randomly assigned to
study, the researcher uses intact groups. the groups.

6. An independent variable is the one being 14. When sample sizes are small,
manipulated by the researcher whereas the replication of the study is done with larger
dependent variable is the one being studied groups to help rule out inconsistencies
to identify the effects of the independent caused by a small group of subjects.
variable.
15. Experimental
7. In research studies, a treatment group
subject receives a specific treatment while 16. Observational
those in the control group do not receive a
treatment or are given a placebo. 17. Observational

8. The Hawthorne effect was the effect 18. Experimental

discovered when researchers realized that
the workers at the Hawthorne plant knew 19. Independent variable - minutes
that they were participating in a research exercising
study and this fact actually changed workers' Dependent variable - catching a cold
behavior.
20. Independent variable - hugs and hand
9. A confounding variable is one that can holding
influence the results of the research study Dependent variable - heart rate and blood
when no precautions were taken to eliminate pressure
it from the study.
21. Independent variable - time spent in
10. Sometimes subjects respond favorably meditation
when given a placebo rather than the real Dependent variable - rational decisions
treatment.
22. Independent variable - income

3
 Chapter 1 - The Nature of Probability and Statistics

22. continued 36. continued

Dependent variable - number of times whom the weight loss product worked
people visit a fast-food restaurant extremely well.

23. Age, income, socioeconomic status. 37. There is no mention of how this
Answers will vary. conclusion was obtained.

24. Age, intelligence, family ties. Answers 38. "74% more calories" than what? No
will vary. comparison group is stated.

25. Wealth, age, education. Answers will 39. Since the word may is used, there is no
vary. guarantee that the product will help fight
cancer.
26. Climate, diet, medicine. Answers will
vary. 40. What is meant by "24 hours of acid
control"? Also can provide implies that it
27. How is a perfect body defined might or might not provide control.
statistically?
41. No. There are many other factors that
28. What is meant by the "whole truth"? contribute to criminal behavior.

29. How can 24 hours of pain relief be 42. Possible answer: It could be the amount
measured? of caffeine in the coffee or tea. It could have
been the brewing method.
30. How can reading a book increase a
person's IQ? 43. Answers will vary.

31. How much weight, if any, will be lost? 44. Answers will vary.

32. The stueps in conducting a medical 45. Answers will vary.

study are as follows:
46. Answers will vary.
Step 1. Formulate the purpose of the study.
Step 2. Identify the variables for the study. REVIEW EXERCISES - CHAPTER 1
Step 3. Define the population.
Step 4. Decide what sampling method to use 1. Inferential 2. Descriptive
to collect the data.
Step 5. Collect the data. 3. Descriptive 4. Descriptive
Step 6. Summarize the data and perform any
statistical calculations needed. 5. Inferential 6. Inferential
Step 7. Interpret the results.
7. Descriptive 8. Inferential
33. Only 20 people were used in the study.
9. Ratio 10. Ordinal
34. The only time claims can be proved is
when the entire population is used. 11. Interval 12. Ratio

35. It is meaningless since there is no 13. Ratio 14. Ratio

definition of "the road less traveled." Also,
there is no way to know that for every 100 15. Ordinal 16. Ratio
women, 91 would say that they have taken
"the road less traveled." 17. Ratio 18. Nominal

36. Since the results are not typical, the 19. Qualitative 20. Quantitative
advertisers selected only a few people for

4
 Chapter 1 - The Nature of Probability and Statistics

21. Quantitative 22. Qualitative 56. How many men would admit that they
wash dishes?
23. Quantitative 24. Quantitative
57. It depends on how the Internet is used.
25. Quantitative 26. Qualitative
58. Since most people swim in shallow
27. Discrete 28. Continuous water, there would be more shark attacks in
shallow water.
29. Discrete 30. Continuous
CHAPTER QUIZ
31. Continuous 32. Discrete 1. True
2. True
33. Continuous 34. Continuous 3. False, nonsampling error is the result of
collecting data incorrectly or selecting a
35. 35.5-36.5 36. 105.35-104.45 biased sample.
4. False, it is stratified sampling.
37. 72.55-72.65 38. 5.265-5.275 5. True
6. True
39. Cluster 40. Systematic 7. False, it is 5.5-6.5 inches.
8. c 9. b
41. Random 42. Systematic 10. d 11. a
12. c 13. a
43. Stratified 44. Volunteer or other 14. Descriptive, inferential
15. Gambling, insurance. Answers can
45. Experimental vary.
16. Population
46. Observational 17. Sample

47. Observational 18.

a. Saves time
48. Experimental b. Saves money
c. Use when population is infinite
49. Independent variable - type of coffee
Dependent variable - milligrams of caffeine 19.
per ounce a. Random c. Cluster
b. Systematic d. Stratified
50. Independent variable - miles walked
Dependent variable - triglyceride level 20. Quasi-experimental

51. Independent variable - thyme 21. Random

Dependent variable - antioxidants
22.
52. Independent variable - laughing a. Descriptive d. Inferential
Dependent variable - pain level b. Inferential e. Inferential
c. Descriptive
53. A telephone survey won't contact all the
types of people who shop online. 23.
a. Nominal d. Interval
54. It is impossible to count all the deer in a b. Ratio e. Ratio
county. c. Ordinal
55. People tend not to report their true
income.

5
 Chapter 1 - The Nature of Probability and Statistics

24.
a. Continuous d. Continuous
b. Discrete e. Discrete
c. Continuous

25.
a. 31.5  32.5 minutes d. 18.5  19.5 pounds
b. 0.475  0.485 millimeter e. 12.05  12.15 quarts
c. 6.15  6.25 inches