Deck 6: Qualitative and Quantitative Sampling

Question 1

Which random sampling method is best when the population has several groups and a researcher wants to ensure that each group is in the sample? Give an example.

Answer 1

The random sampling method best suited for a situation where the population has several distinct groups and the researcher wants to ensure representation from each group is stratified random sampling.

Stratified random sampling involves dividing the population into different subgroups or strata that share similar characteristics. Each subgroup is then sampled as an independent sub-population, out of which individual elements are randomly selected. This method ensures that each subgroup is proportionally represented in the final sample, which can provide more accurate and generalizable results than simple random sampling, especially when there are significant differences between groups.

Example:
Imagine a researcher is studying the job satisfaction levels of employees in a large multinational corporation that has offices in multiple countries. The researcher wants to ensure that the sample reflects the geographical diversity of the company's workforce. To achieve this, the researcher could use stratified random sampling by first dividing the population into strata based on the location of the offices (e.g., North America, Europe, Asia, etc.). Within each of these geographical strata, the researcher would then randomly select a proportionate number of employees to participate in the study. This approach would guarantee that each region is adequately represented in the sample, allowing the researcher to compare job satisfaction across different locations and to draw conclusions that are relevant to the entire company.

Question 2

Explain how the logic of sampling is related to the logic of measurement related.

Answer 2

The logic of sampling is closely related to the logic of measurement in that both processes involve the collection and analysis of data in order to make inferences about a larger population.

Sampling involves selecting a subset of individuals or items from a larger population in order to make inferences about the population as a whole. This process is based on the idea that by studying a representative sample of the population, we can gain insights into the characteristics and behaviors of the entire population. Similarly, measurement involves assigning numerical values to observations or variables in order to quantify and describe their characteristics.

In both sampling and measurement, the goal is to gather data that accurately represents the larger population or phenomenon of interest. This requires careful planning and consideration of factors such as sample size, sampling methods, and measurement techniques to ensure that the data collected is valid and reliable.

Furthermore, the logic of sampling and measurement is also related in the sense that both processes involve making inferences and generalizations based on the data collected. Whether it is using statistical techniques to make inferences about a population based on a sample, or using measurement data to draw conclusions about the characteristics of a variable, both sampling and measurement involve using data to make broader statements about the world around us.

Overall, the logic of sampling and measurement are closely intertwined, as both processes involve the collection and analysis of data in order to gain insights into larger populations or phenomena. Both processes require careful planning and consideration of factors such as representativeness and reliability in order to ensure that the data collected is valid and can be used to make meaningful inferences.

Question 3

Hubert Humbort wants to draw a sample of people in Michigan who own an authentic, imported espresso coffee maker. There is no list of the people, and there is no reason to believe that they know each other. What type of sampling should Hubert use?

A) cluster sampling
B) snowball sampling
C) simple random sampling
D) purposive sampling
E) accidental sampling

Answer 3

purposive sampling

Question 4

Refer to the following paragraph to answer the questions below.
The Young Children's Charity would like to provide an accurate estimate of how much it costs to raise a child between the ages of 2 and 12 each year for its fundraising campaign. They contracted with you to find out how much households spent on raising a child in 2006 for clothing, toys/recreation, education, baby sitting, and a percentage of the expenses for family food, utilities (electricity and gas), transportation, and rent or house payments. They gave you a list 4 million residential telephone customers in the area they will operate the campaign. You sampled every 4,000th address on the list. The Charity will have a professional survey company contact each sampled household by telephone and ask whether or not there is a child between 2 and 12 years old living in the household. If there is, their interviewers will ask other questions and record the total amount spent raising a child in the past two months. They will multiply this amount by 6 to get an annual cost.

-How large is your sample?

A) 200
B) 500
C) 2,000
D) 2,500
E) 5,000

Question 5

Refer to the following paragraph to answer the questions below.
The Young Children's Charity would like to provide an accurate estimate of how much it costs to raise a child between the ages of 2 and 12 each year for its fundraising campaign. They contracted with you to find out how much households spent on raising a child in 2006 for clothing, toys/recreation, education, baby sitting, and a percentage of the expenses for family food, utilities (electricity and gas), transportation, and rent or house payments. They gave you a list 4 million residential telephone customers in the area they will operate the campaign. You sampled every 4,000th address on the list. The Charity will have a professional survey company contact each sampled household by telephone and ask whether or not there is a child between 2 and 12 years old living in the household. If there is, their interviewers will ask other questions and record the total amount spent raising a child in the past two months. They will multiply this amount by 6 to get an annual cost.

-Which of the following samples will have the SMALLEST sampling error?
[NOTE: The standard deviation measures diversity with a larger number indicating greater heterogeneity and diversity.]

A) sample size = 1,000, Sample standard deviation = 55
B) sample size = 100, Sample standard deviation = 5.5
C) sample size = 10,000, Sample standard deviation = 55
D) sample size = 1,000, Sample standard deviation = 5.5
E) sample size = 10,000, Sample standard deviation = 1.0

Question 6

Refer to the following paragraph to answer the questions below.
The Young Children's Charity would like to provide an accurate estimate of how much it costs to raise a child between the ages of 2 and 12 each year for its fundraising campaign. They contracted with you to find out how much households spent on raising a child in 2006 for clothing, toys/recreation, education, baby sitting, and a percentage of the expenses for family food, utilities (electricity and gas), transportation, and rent or house payments. They gave you a list 4 million residential telephone customers in the area they will operate the campaign. You sampled every 4,000th address on the list. The Charity will have a professional survey company contact each sampled household by telephone and ask whether or not there is a child between 2 and 12 years old living in the household. If there is, their interviewers will ask other questions and record the total amount spent raising a child in the past two months. They will multiply this amount by 6 to get an annual cost.

-How large is his sample?

A) 500
B) 1,000
C) 1,500
D) 2,000
E) 20,000

Question 7

Refer to the following paragraph to answer the questions below.
The Young Children's Charity would like to provide an accurate estimate of how much it costs to raise a child between the ages of 2 and 12 each year for its fundraising campaign. They contracted with you to find out how much households spent on raising a child in 2006 for clothing, toys/recreation, education, baby sitting, and a percentage of the expenses for family food, utilities (electricity and gas), transportation, and rent or house payments. They gave you a list 4 million residential telephone customers in the area they will operate the campaign. You sampled every 4,000th address on the list. The Charity will have a professional survey company contact each sampled household by telephone and ask whether or not there is a child between 2 and 12 years old living in the household. If there is, their interviewers will ask other questions and record the total amount spent raising a child in the past two months. They will multiply this amount by 6 to get an annual cost.

-What is his sampling ratio?

A) .002 or 0.2%
B) .005 or 0.5%
C) .025 or 2.5%
D) .050 or 5%
E) none of the above

Question 8

Refer to the following paragraph to answer the questions below.
Dr. Martinez draws a systematic sample of 350 churches from all churches and religious institutions in the three Pacific coast states of the continental U.S. and British Columbia. His sampling frame has 35,000 institutions.

-What is the sampling interval?

A) .001
B) .01
C) .10
D) 10
E) 100

Question 9

Refer to the following paragraph to answer the questions below.
Dr. Martinez draws a systematic sample of 350 churches from all churches and religious institutions in the three Pacific coast states of the continental U.S. and British Columbia. His sampling frame has 35,000 institutions.

-What is the sampling ratio?

A) .001
B) .01
C) .10
D) 10
E) 100

Question 10

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-What is Drunkman's sampling frame?

A) the drinking behavior of students
B) students both in the directory and not in the directory
C) the 2002 student telephone directory
D) Wild State University
E) the class levels

Question 11

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-What is Drunkman's population?

A) all students living in Boozville, Texas
B) all students enrolled in Wild State University
C) students who drink alcohol
D) all residents of Boozville, Texas
E) students who have roommates

Question 12

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-About how many students are listed in the 2002 Wild State University student telephone directory?

A) 2,000
B) 10,000
C) 15,000
D) 20,000
E) 35,000

Question 13

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-What type of sampling did Drunkman use?

A) stratified
B) quota
C) cluster
D) systematic
E) simple random

Question 14

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-The registrar's office stated that 3,000 currently-enrolled student names were omitted from the directory, but everyone in the directory was still enrolled. With this revised estimate of the population and final number of students on which he has data, what is Drunkman's actual sampling ratio?

A) 1 percent
B) 5 percent
C) 8 percent
D) 10 percent
E) 15 percent

Question 15

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-Professor Drunkman's questionnaire asked each student to name every other student with whom they drank during the past two months. Dry Temperance, his assistant, sent a questionnaire to each person named by students in the first sample in which the same question was asked. Ms. Temperance next sent a questionnaire to everyone named as a drinking partner in the second sample and so on until she had sent out ten different waves of questionnaires. She then drew up a list of all students who drank with anyone else that returned a questionnaire, and used it as a sample for a study on drinking networks on campus. What kind of sampling was Ms. Temperance using?

A) accidental sampling
B) snowball sampling
C) quota sampling
D) cluster sampling
E) stratified sampling

Question 16

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-Drunkman wants to mail out questionnaires, but he wants to be certain that the sample contains an equal number of freshman, sophomore, junior, and senior students. He asks you to draw a stratified sample. What is the best way to draw the sample?

A) Choose every 10th, instead of 15^th, name when selecting students.
B) Use the registrar's list of students instead of the telephone directory, and select every 15th student name.
C) Get a list of each class (e.g., freshman, sophomore) from the registrar and use a random number table or program to select 500 names from each class.
D) Open up the phone book to a random page and begin calling up students listed. Ask them what class they are in, then send questionnaires to the first 500 who say "freshman," the first 500 saying "sophomore," etc.
E) First randomly select 10 dormitories, then randomly select every other floor of the 10 selected dormitories, then mail questionnaires to students on the students living in rooms on the selected floors.

Question 17

Refer to the following paragraph to answer the questions below.
Professor Alco Drunkman studied class level and drinking behavior at Wild State University in Boozville, Texas. Every student is required to live in one of six twenty-story-tall dormitories, and has a single room. All dormitory room phone numbers began with the prefix 747 or 757 (e.g., phone numbers are 747-0001 to 757-9999). He used the student telephone directory, and after a random start selected every 15th student. He then mailed a questionnaire to the 1,000 students selected and had two follow-up postcard reminders. A total of 900 students returned questionnaires. Later, he asked the university registrar whether any students were omitted from the directory.

-Drunkman decided to conduct his study using a telephone interview instead using a mailed questionnaire. Which is the most accurate way to sample?

A) a simple random sample from the directory of the town
B) a systematic sample of the student telephone directory
C) random-digit dialing using the 747 or 757 prefix
D) a quota sample with an equal number of students from each dormitory
E) a stratified sample from 100-, 200-, 300-, and 400-numbered courses

Question 18

John Johnson used nonrandom sampling in his study of successful scientists. He knew most got excellent grades in high school and college, excelled in mathematics, and had a strong interest in science from a young age. For his study, he was interested in sampling those who were successful as adult scientists, but did poorly in high school, did average or worse in math classes, and showed little interest in science until they reached college or later in life. What type of sampling was he using?

A) snowball
B) theoretical
C) sequential
D) deviant case
E) quota

Question 19

Refer to the following paragraph to answer the questions below.
Market researcher Shoko Yamamoto wants to test whether men like a new flavor of yogurt, which tastes like cigar smoke and beer, more than women. She has assistants go to four large grocery stores and distribute samples on each Saturday in April. In addition to distributing yogurt in white cups to anyone, the assistants give 50 samples in blue cups to adult males shopping alone and 50 samples in pink cups to adult females shopping alone. Afterwards, the assistants check nearby shelves and in a nearby garbage can to locate the used blue and pink cups. They measure whether more waste (i.e. uneaten yogurt) was left in blue or pink cups.

-What type of sampling was used in the study?

A) random
B) quota
C) stratified
D) snowball
E) cluster

Question 20

If you use this, you do not give an exact estimate of the population parameter. Instead the range is from a little below and a little above your best estimate of the population parameter based on statistics about the sampling error from your random sample and how certain you want to be of an answer.

A) confidence interval
B) deviant case sampling
C) sampling distribution
D) sampling ratio
E) probability proportionate to size (PPS)

Question 21

Talk about:
-cluster sampling

Question 22

Talk about:
-confidence intervals

Question 23

Talk about:
-haphazard sampling

Question 24

Talk about:
-hidden populations

Question 25

Talk about:
-inferential statistics

Question 26

Talk about:
-nonrandom sample

Question 27

Talk about:
-parameter

Question 28

Talk about:
-population

Question 29

Talk about:
-purposive sampling

Question 30

Talk about:
-quota sampling

Question 31

Talk about:
-random sample

Question 32

Talk about:
-sample

Question 33

Talk about:
-sampling distribution

Question 34

Talk about:
-sampling element

Question 35

Talk about:
-sampling error

Question 36

Talk about:
-sampling frame

Question 37

Talk about:
-sampling interval

Question 38

Talk about:
-sampling ratio

Question 39

Talk about:
-sequential sampling

Question 40

Talk about:
-snowball sampling

Question 41

Talk about:
-sociogram

Question 42

Talk about:
-statistic

Question 43

Talk about:
-stratified sampling

Question 44

Talk about:
-systematic sampling

Question 45

Talk about:
-target population