Deck 7: Data for Decisions

Question 1

Define a simple random sample.

Answer 1

A simple random sample is one in which every group of same size from the population has an equally likely chance of being chosen.

Question 2

A high school principal wishes to determine what proportion of the school's students likes the new school mascot. The principal decides to survey every 25th name from the school enrollment records (an alphabetical list of all students at the school). Is this a valid simple random sampling technique? Why or why not?

Answer 2

Taking every 25th name from an alphabetical listing of students is not a valid simple random sampling technique since no two students who follow one another on the list could ever be part of the sample. This contradicts the idea that in a simple random sample, every collection of individuals has an equal chance of being chosen.

Question 3

In order to determine what proportion of a town's residents approves of a new plan for trash collection, the town council placed an ad in the newspaper asking residents to phone in their opinions. Explain why the results of this poll may differ from the actual beliefs of all residents in the town.

Answer 3

The sampling technique described uses self-selection. Only those residents with extreme views on the new plan are likely to respond to the advertisement. A second problem with the poll technique described is that not all residents who will be affected by the new trash collection plan are likely to read the newspaper. The opinions of a large number of residents are therefore unlikely to be included in the results of the poll.

Question 4

A school principal is concerned with the increasing level of absenteeism in the school. A meeting of parents, teachers, and students is called at which the principal expresses her concern and describes an experimental program that will be instituted to try to curb absenteeism. After two months, absenteeism is down by 15%. Explain how confounding variables may have affected the results of the experiment.

Question 5

In order to determine if students on a college campus are in favor of a tuition hike to pay for expanded parking services, a member of the student senate surveys 25 people in a commuter parking lot. Why is this a poor sampling technique?

Question 6

One July, the city council of a small town decides to impose an experimental curfew on all residents younger than 18 years to cut down on loitering in the town square. After four months, the number of teens found in the square after dark has decreased by 80%, and the council declares the curfew a success. Explain why this conclusion may not be valid.

Question 7

We must use a random digits table to choose a sample of eight names from the roster of a club with 100 members. Why can we use two-digit numbers from the table to select our sample?

Question 8

What is the difference between an observational study and an experiment?

Question 9

Explain the difference between bias and variability in sampling results.

Question 10

A biologist draws a sample of 200 fish from a lake to test for mercury levels. She finds that 8% have levels above the limits set as healthy. Describe the population for this study.

Question 11

Convenience samples are said to be highly likely to produce bias in survey results. Explain why this is true.

Question 12

A farmer believes that exposing chickens to classical music will cause them to produce more eggs. Describe how the farmer may design a randomized comparative experiment to test this theory.

Question 13

A polling company working for a candidate for governor surveys a sample of 2500 registered voters in the state to determine if they are in agreement with the candidate's stand on gun control. Describe the population for this study.

Question 14

In a poll of 2500 residents of a state, it is found that 480 are in favor of naming the grasshopper the state insect. What is the sample proportion for this poll?

Question 15

Why do opinion polls usually report a margin of error with the results of a survey? What does the margin of error mean?

Question 16

We wish to know what proportion of students at a major university believes too much emphasis is placed on athletics at the school. Explain how we could choose a sample of 500 students to reduce the possibility of bias in the results.

Question 17

The marketing department of a large national corporation wishes to determine what proportion of the residents of a state may be interested in buying its new product. Describe how the corporation might use a multistage random sampling process to choose a sample of the state's residents to survey.

Question 18

A school superintendent believes a new approach to teaching children to read will produce better standardized test scores in the district. Describe how the superintendent may design a randomized comparative experiment to test this theory.

Question 19

Can we eliminate variability in results of sampling? Why or why not?

Question 20

A marketing department surveys 1500 shoppers and finds that 950 would visit a new store more often if it was open on Sunday evenings. What is the sample proportion in this survey?

Question 21

A new stadium is being planned for the local professional football team. One of the proposed funding methods involves the allocation of tax monies to build the stadium. To determine the public opinion of this plan, the first 100 people entering the stadium for a game are surveyed. Explain why the results of this survey might not be valid.

Question 22

A marketing company conducted a survey of college students to obtain data for an advertising campaign. They selected 1421 students randomly from campus directories of 132 colleges and universities. The students in the directories at the 132 colleges and universities represent:

A) the population.
B) the sample.

Question 23

Officials at a university conduct a survey to determine the percentage of students who favor the allocation of a percentage of tuition funds toward the construction of a new campus parking garage. One thousand people driving through the administration building parking lot are surveyed, and it is found that 75% of these people favor the garage. What is the population in this study? What is the sample?

Question 24

Officials at a university conduct a survey to determine the percentage of students who favor the allocation of a percentage of tuition funds toward the construction of a new campus parking garage. One thousand people driving through the administration building parking lot are surveyed, and it is found that 75% of these people favor the garage. Explain why this conclusion might not be valid.

Question 25

A marketing company conducted a survey of college students to obtain data for an advertising campaign. They selected 1421 students randomly from campus directories of 132 colleges and universities. The 1421 students represent:

A) the population.
B) the sample.

Question 26

You take a simple random sample of 500 from a large population. If the population proportion is p = 0.2, describe the sampling distribution in terms of shape, center and variability.

Question 27

You sample all 1000 people in the population of interest and find that is 0.2. Why should you not use the margin of error formula to find a 95% confidence interval for p?

Question 28

A poll asks 312 randomly selected students whether they prefer the new beaver mascot or the old moose mascot. It turns out that the sample proportion for those preferring the moose is 0.54. What is the margin of error for a 95% confidence interval?

Question 29

To estimate the proportion of voters in a town likely to favor a tax increase for road repair, a random sample of people chosen from the voter registration list is surveyed and the proportion who favors the increase is found to be 43%. The actual proportion in the town is 40%. This difference is most likely an example of sampling:

A) bias.
B) variability.

Question 30

Describe what is meant by the placebo effect in an experiment.

Question 31

A polling company conducted a survey of voters to obtain data for a political campaign. They selected 3500 voters randomly from the 168,000 names on the voter registration lists of the county and found that 1372 intended to vote for candidate Doe. The 168,000 names represent:

A) the population.
B) the sample.

Question 32

Explain the difference between a parameter and a statistic.

Question 33

How does a randomized comparative experiment control the effects of confounding variables?

Question 34

A survey of 127 patients in a particular hospital determined that 54% of them felt that they were receiving excellent care. Determine the 95% confidence interval for this survey.

Question 35

What is nonresponse?

Question 36

Medical experiments are frequently double-blind. Explain what this means and why medical experiments are designed this way.

Question 37

What does it mean to say that the results of an experiment are statistically significant?

Question 38

In order to determine the mean weight of bags of chips filled by its packing machines, a company inspects 50 bags per day and weighs them. In this example, the population is:

A) the 50 bags inspected each day.
B) all potato chips produced by the company.
C) all bags of chips produced by the company.
D) the weight of the 50 bags inspected.

Question 39

A polling company conducted a survey of voters to obtain data for a political campaign. They selected 3500 voters randomly from the 168,000 names on the voter registration lists of the county and found that 1372 intended to vote for candidate Doe. The 3500 voters represent:

A) the population.
B) the sample.

Question 40

A recent survey of 536 employees of U.S. corporations determined that 84% of them feel that the executives who run their companies are ethical. Determine the 95% confidence interval for this survey.

Question 41

A polling company conducted a survey of voters to obtain data for a political campaign. They selected 3500 voters randomly from the 16,800 names on the voter registration lists of the county and found that 1365 intended to vote for candidate Doe. The sample proportion is:

A) 0.081.
B) 0.208.
C) 0.39.
D) 0.1365.

Question 42

A polling company surveys 200 people outside a county courthouse concerning tighter restrictions on smoking in public buildings. Their results indicate that 34% of those surveyed favor tighter restrictions. The actual proportion of county residents who favor tighter restrictions is 65%. The difference is most likely due to:

A) variability in sampling.
B) bias due to the use of a convenience sample.
C) confounding variables in the survey.
D) the wording of the survey.

Question 43

Beginning with line 102 of the random digits table above, select three individuals from the list below to serve on a student affairs committee.

A) Cooper, Doe, Smith
B) Doe, Ewing, Grant
C) Doe, Smith, Smith
D) Crosby, Ewing, Grant

Question 44

Which of the following sampling method is unbiased?

A) convenience sampling
B)
C)
D) voluntary response sampling
Simple random sampling
Forced response sampling

Question 45

Beginning with line 104 of the random digits table above, select three individuals to receive a prize in a contest.

A) Jones, Jones, Ubet
B) Hunsaker, Jones, Miller
C) Jones, Ubet, Wilson
D) Butts, Jones, Ubet

Question 46

To determine the proportion of students at a university who favor the construction of a parking garage, a student senate member surveys students as they leave the student union. This type of sample is a:

A) convenience sample.
B) simple random sample.
C) multistage random sample.
D) voluntary response sample.

Question 47

A poll of 60 students selected at random at State University found that 12 were in favor of higher parking fees to support extra police patrols of parking lots on campus. In this example, the sample proportion is:

A) 0.2.
B) 0.12.
C) 0.05.
D) 0.6.

Question 48

To determine the proportion of students at a university who favors the construction of a parking garage, a sample of people driving through the student center parking lot is surveyed and it is found that 45% favor the garage. The actual proportion of the student body who favors the garage is 40%. This difference is most likely an example of sampling:

A) bias.
B) variability.

Question 49

If the design of a statistical study favors systematically certain outcomes, the design has:

A) bias.
B) variability.

Question 50

A random sample of 600 voters in Centralville indicated that 48 of them believed their congressional representative was honest and trustworthy. The sample proportion is:

A) 0.08.
B) 0.125.
C) 0.48.
D) 0.8.

Question 51

Use the random digits table above, starting at line 105, to select four people from a list numbered 00 to 99.

A) 75, 02, 95, 10
B) 75, 51, 25, 02
C) 10, 57, 50, 29
D) 01, 00, 05, 07

Question 52

A marketing company conducts a survey of college students to obtain data for a marketing campaign. They randomly select five in-state colleges and then randomly choose 100 students from the registration lists of these colleges. This type of sample is a:

A) convenience sample.
B) simple random sample.
C) multistage random sample.
D) voluntary response sample.

Question 53

Use the random digits table above, starting at line 102, to choose five people from a list numbered 001 to 500 to receive a survey.

A) 491, 050, 075, 242, 403
B) 491, 007, 392, 440, 354
C) 49, 10, 50, 07, 55
D) 49, 61, 08, 75, 91

Question 54

In order to determine the proportion of voters in a small town who favor a candidate for mayor, the campaign staff takes out an ad in the paper asking voters to call in their preference for mayor. This type of sample is a:

A) convenience sample.
B) simple random sample.
C) multistage random sample.
D) voluntary response sample.

Question 55

You wish to survey the students at your college to determine their feelings about the quality of services in the student center. Which of the following sampling designs is best for avoiding bias?

A) Place an ad in the student newspaper asking all readers to mail in their opinions.
B) Obtain a list of student names from the registrar and select 250 names to contact.
C) Air an announcement on the campus radio station asking all listeners to phone in their opinions.
D) Survey every 10th student who enters the student center.

Question 56

Use the random digits table above, starting at line 104, to choose four people from a list numbered 001 to 500 to call for a poll.

A) 089, 008, 778, 873
B) 089, 008, 192, 445
C) 089, 008, 77, 88
D) 089, 192, 459, 256

Question 57

Use the random digits table above, starting at line 103, to choose five people from a list numbered 01 to 99 to receive a survey.

A) 01, 03, 30, 87, 05
B) 10, 36, 15, 89, 35
C) 61, 58, 93, 54, 86
D) 61, 58, 35, 48, 59

Question 58

A poll of 80 students selected at random at Midtown University found that 20 were in favor of a fee increase to support extra maintenance of gardens on campus. In the example, the sample proportion is:

A) 0.2.
B) 0.08.
C) 0.8.
D) 0.25.

Question 59

Which of the following sampling techniques is most likely to produce biased results?

A) multistage random sampling
B) sampling using a random digits table
C) voluntary response sampling
D) simple random sampling

Question 60

Random selection of subjects in a survey is used to eliminate variability in results.

Question 61

When an accurate and complete list of the population is not available, a sampling will suffer from:

A) nonresponse.
B)
C)
D) undercoverage.
Response.
Nonrandomness.

Question 62

A retrospective study may be used to show cause and effect.

Question 63

Nonresponse occurs when:

A) all participants surveyed answer "no" to a question.
B) a sample for a study is not chosen at random.
C) an individual selected for a sample cannot be contacted or refuses to participate.
D) a study does not produce statistically significant results.

Question 64

In an experiment, an observed effect is called statistically significant if:

A) the experiment will help a large number of people.
B) the study was double-blind.
C) the experiment was well designed.
D) the effect is too large to attribute to chance.

Question 65

A group of 200 students is identified. Half took Latin in high school and half did not. The students are compared to see if the students who took Latin received higher SAT verbal scores. This type of study is a(n):

A) comparative experiment.
B) experiment with compounding variables.
C) prospective study.
D) simple random sample.

Question 66

A group of 100 students is randomly chosen and divided into two groups. One group is taught typing using a set of new materials and the other using traditional methods. After instruction, typing speeds are compared to determine if the new materials improve learning. This type of study is a(n):

A) comparative experiment.
B) observational study.
C) prospective study.
D) retrospective study.

Question 67

A physical education researcher wishes to determine if walking every day affects the health of middle-aged men. The researcher randomly identifies 400 participants in two groups. Members of one group belong to a health club and walk on its track and members of the other do not. After two months, the researcher decides the group who walks daily is healthier and concludes that walking positively affects men's health. This conclusion is of questionable validity because:

A) of the placebo effect.
B) of confounding variables in the study.
C) the study is not double-blind.
D) the researcher is not a doctor.

Question 68

The type of statistical study that can show cause and effect is:

A) an experimental study.
B) an observational study.
C) a prospective study.
D) a retrospective study.

Question 69

Consider the following situation: Doctors identify 500 women, half of whom had mothers with breast cancer and half of whom did not. The medical records of the women are followed up for 20 years to determine if heredity plays a role in breast cancer. This describes a prospective study.

Question 70

Random selection of subjects for surveys is used to avoid:

A) placebo effects.
B) variability.
C) double-blindness.
D) bias.

Question 71

A prospective study may be used to show cause and effect.

Question 72

A study gathered data on 1000 randomly selected students and showed that students who took Latin in high school had much higher scores on a test of verbal English skills than those who did not take Latin. The study cannot conclude that taking Latin improves verbal English skills because:

A) the study was not an experiment.
B) the study was not double-blind.
C) of the placebo effect.
D) the verbal English test was faulty.

Question 73

A 10-year study of low-birth-weight babies is performed to determine if birth weight affects IQ and performance in elementary school. Children are identified in hospitals at birth and their performance is tracked until they are 10 years old. This type of study is a(n):

A) comparative experiment.
B) experiment with compounding variables.
C) observational study.
D) retrospective study.

Question 74

When variables have effects on the outcome which cannot be distinguished from each other, those variables are said to be:

A) correlated.
B)
C)
D) confounded.
Well designed.
Dependent.

Question 75

If an experiment is designed in a way that both of the experimental subjects and the person who interact with them don't know which treatment each subject received, then the experiment is:

A) randomized.
B) well controlled.
C) double-blind.
D) not controllable.

Question 76

Consider the following situation: A group of 300 students is randomly selected at a local high school and required to fill out yearly questionnaires on family income. Students' performances on standardized tests are then followed throughout their high school years to determine if socioeconomic status affects SAT scores. This describes an experiment.

Question 77

In a medical study that is double-blind, participants do not know whether they are taking the experimental drug. This is to avoid:

A) selection bias.
B) the placebo effect.
C) having too small a sample.
D) statistical significance.

Question 78

A survey is sent to 100 employees at a community hospital asking if they support a law requiring motorcycle riders to wear helmets. The results indicate 88% support the law. If the actual proportion of the community's residents who support the law is 72%, the difference is most likely a result of:

A) sampling bias.
B) sampling variability.
C) an insufficient sample size.
D) a poorly worded questionnaire.

Question 79

A dummy medication (such as a salt tablet) will often help a patient who trusts the doctor who administers the medicine. This is called:

A) confidentiality.
B) double-blindness.
C) confounding variables.
D) the placebo effect.

Question 80

Consider the following situation: Doctors question 5000 women who have had breast cancer to determine if there is a genetic factor that increases the likelihood of cancer. Each woman is asked about her family background and incidences of breast or other types of cancer among her relatives. This describes a prospective study.