Deck 17: Turning Information Into Wisdom

Question 1

Use the following information for questions:
In a study of 900 European American children born in 1991 ("Maternal Employment and Child Cognitive Outcomes in the First Three Years of Life: The NICHD Study of Early Child Care," Brooks-Gunn, Han, and Waldfogel, Child Development 73[2002]: 1052-1072), the authors wrote "children whose mothers worked at all by the ninth month of their life had lower [cognitive outcome] scores at 36 months than did children whose mothers did not work by that time." These results were statistically significant.

-What is the response variable in this study?

A) The age of the child
B) The cognitive outcome score of the child at 36 months
C) The working history of the mother (did or did not work during the first nine months of the child's life)
D) None of the above

Answer 1

The cognitive outcome score of the child at 36 months

Question 2

Use the following information for questions:
In a study of 900 European American children born in 1991 ("Maternal Employment and Child Cognitive Outcomes in the First Three Years of Life: The NICHD Study of Early Child Care," Brooks-Gunn, Han, and Waldfogel, Child Development 73[2002]: 1052-1072), the authors wrote "children whose mothers worked at all by the ninth month of their life had lower [cognitive outcome] scores at 36 months than did children whose mothers did not work by that time." These results were statistically significant.

-What is the explanatory variable in this study?

A) The age of the child
B) The cognitive outcome score of the child at 36 months
C) The working history of the mother (did or did not work during the first nine months of the child's life)
D) None of the above

Answer 2

The working history of the mother (did or did not work during the first nine months of the child's life)

Question 3

Use the following information for questions:
In a study of 900 European American children born in 1991 ("Maternal Employment and Child Cognitive Outcomes in the First Three Years of Life: The NICHD Study of Early Child Care," Brooks-Gunn, Han, and Waldfogel, Child Development 73[2002]: 1052-1072), the authors wrote "children whose mothers worked at all by the ninth month of their life had lower [cognitive outcome] scores at 36 months than did children whose mothers did not work by that time." These results were statistically significant.

-Can a conclusion be made that mothers who worked during the first nine months after having a child caused the lower scores seen at 36 months?

A) A cause and effect relation might be concluded because of the large sample size and the long follow-up period (3 years).
B) A cause and effect relation might be concluded because the results were statistically significant.
C) A cause and effect relation cannot be concluded because the study was an observational study.
D) A cause and effect relation cannot be concluded because the study was a case-control study.

Answer 3

A cause and effect relation cannot be concluded because the study was an observational study.

Question 4

Use the following information for questions:
A study of 42 patients with carpal tunnel syndrome (symptoms of burning numbness in the wrist to difficulty making a fist) was done ("Yoga based intervention for carpal tunnel syndrome," Garfinkel et al, JAMA 1998; 280:1601-1603). Subjects were randomly assigned to a yoga group or a control (where they were offered a wrist splint to supplement their current treatment). Results showed that subjects in the yoga group had significant improvement in grip strength, while subjects in the control group did not.

-What is the response variable in the study?

A) The improvement in grip strength
B) The treatment group (either Yoga or control group)
C) The severity of carpal tunnel syndrome
D) The age of the patient.

Question 5

Use the following information for questions:
A study of 42 patients with carpal tunnel syndrome (symptoms of burning numbness in the wrist to difficulty making a fist) was done ("Yoga based intervention for carpal tunnel syndrome," Garfinkel et al, JAMA 1998; 280:1601-1603). Subjects were randomly assigned to a yoga group or a control (where they were offered a wrist splint to supplement their current treatment). Results showed that subjects in the yoga group had significant improvement in grip strength, while subjects in the control group did not.

-What is the explanatory variable in the study?

A) The improvement in grip strength
B) The treatment group (either Yoga or control group)
C) The severity of carpal tunnel syndrome
D) The age of the patient.

Question 6

Use the following information for questions:
A study of 42 patients with carpal tunnel syndrome (symptoms of burning numbness in the wrist to difficulty making a fist) was done ("Yoga based intervention for carpal tunnel syndrome," Garfinkel et al, JAMA 1998; 280:1601-1603). Subjects were randomly assigned to a yoga group or a control (where they were offered a wrist splint to supplement their current treatment). Results showed that subjects in the yoga group had significant improvement in grip strength, while subjects in the control group did not.

-Can a conclusion that yoga caused the improvement in grip strength be made?

A) A cause and effect relation might be concluded because the study had a control group.
B) A cause and effect relation might be concluded because the study was a randomized experiment.
C) A cause and effect relation cannot be concluded because of the small sample size.
D) A cause and effect relation cannot be concluded because not all subjects in the control group accepted the splint.

Question 7

Use the following information for questions:
43 sedentary adults between the ages of 50 and 76 were randomly assigned to either an exercise program four times a week for 16 weeks or they continued to be sedentary (King, et al, JAMA, January 1, 1997,pp 32-37). The exercise group slept an average of 42 minutes a night longer than the control group.

-What is the response variable in the study?

A) Amount of food eaten (calories) per day
B) Following an exercise program four times a week for 16 weeks (or not)
C) Amount of sleep per night
D) Age

Question 8

Use the following information for questions:
43 sedentary adults between the ages of 50 and 76 were randomly assigned to either an exercise program four times a week for 16 weeks or they continued to be sedentary (King, et al, JAMA, January 1, 1997,pp 32-37). The exercise group slept an average of 42 minutes a night longer than the control group.

-What is the explanatory variable in the study?

A) Amount of food eaten (calories) per day
B) Following an exercise program four times a week for 16 weeks (or not)
C) Amount of sleep per night
D) Age

Question 9

Use the following information for questions:
43 sedentary adults between the ages of 50 and 76 were randomly assigned to either an exercise program four times a week for 16 weeks or they continued to be sedentary (King, et al, JAMA, January 1, 1997,pp 32-37). The exercise group slept an average of 42 minutes a night longer than the control group.

-Can a conclusion that the exercise program caused more sleep?

A) A cause and effect relation might be concluded because the study had a control group.
B) A cause and effect relation might be concluded because the study was a randomized experiment.
C) A cause and effect relation cannot be concluded because the study was an observational study.
D) A cause and effect relation cannot be concluded because diet was not controlled.

Question 10

Use the following information for questions:
Twenty volunteers were assigned at random to one of two groups (10 people in each group). One group would take Vitamin C for two weeks, while the second group would take a placebo. At the end of the study, the proportion of colds in the two groups will be compared.

-What is the response variable in the study?

A) Getting a cold (or not)
B) Taking Vitamin C (or a Placebo)
C) Getting 8 hours of sleep a night (or not)
D) Exposure to other people with colds (or not)

Question 11

Use the following information for questions:
Twenty volunteers were assigned at random to one of two groups (10 people in each group). One group would take Vitamin C for two weeks, while the second group would take a placebo. At the end of the study, the proportion of colds in the two groups will be compared.

-What is the explanatory variable in the study?

A) Getting a cold (or not)
B) Taking Vitamin C (or a Placebo)
C) Getting 8 hours of sleep a night (or not)
D) Exposure to other people with colds (or not)

Question 12

Use the following information for questions:
Twenty volunteers were assigned at random to one of two groups (10 people in each group). One group would take Vitamin C for two weeks, while the second group would take a placebo. At the end of the study, the proportion of colds in the two groups will be compared.

-Can a conclusion that taking Vitamin C causes a change in the risk of a cold?

A) A cause and effect relation might be concluded because the study had a control group.
B) A cause and effect relation might be concluded because the study was a randomized experiment.
C) A cause and effect relation cannot be concluded because the study was an observational study.
D) A cause and effect relation cannot be concluded because exposure to other people with colds was not controlled.

Question 13

Use the following information for questions:
A random sample of 500 students showed a statistically significant relationship between where the students usually sat in class (front, middle, or back) and grade point average (GPA).

-What is the response variable in the study?

A) Where a student typically sits in class (front, middle, or back)
B) Grade point average (GPA)
C) Class standing (freshman, sophomore, junior, or senior)
D) Gender (Male or Female)

Question 14

Use the following information for questions:
A random sample of 500 students showed a statistically significant relationship between where the students usually sat in class (front, middle, or back) and grade point average (GPA).

-What is the explanatory variable in the study?

A) Where a student typically sits in class (front, middle, or back)
B) Grade point average (GPA)
C) Class standing (freshman, sophomore, junior, or senior)
D) Gender (Male or Female)

Question 15

Use the following information for questions:
A random sample of 500 students showed a statistically significant relationship between where the students usually sat in class (front, middle, or back) and grade point average (GPA).

-Can a conclusion that where a student sits in the classroom causes a change in GPA?

A) A cause and effect relation might be concluded because the results were statistically significant.
B) A cause and effect relation might be concluded because the study was a randomized experiment.
C) A cause and effect relation cannot be concluded because the study was an observational study.
D) A cause and effect relation cannot be concluded because seating preference was self-reported.

Question 16

A random sample of 5000 students were asked whether they prefer a 10 week quarter system or a 15 week semester system. Of the 5000 students asked, 500 students responded. The results of this survey ________

A) can be generalized to the entire student body because the sampling was random.
B) can be generalized to the entire student body because the margin of error was 4.5%.
C) cannot be generalized to the entire student body because the non-response rate was 90%.
D) cannot be generalized to the entire student body because the margin of error was 4.5%.

Question 17

Use the following information for questions:
A random sample of 100 students living in dormitories was selected to see whether students in dormitories were satisfied with the food being served in the cafeteria.

-The conservative margin of error for the sample of 100 students is

A) 1%.
B) 5%.
C) 10%.
D) none of the above.

Question 18

Use the following information for questions:
A random sample of 100 students living in dormitories was selected to see whether students in dormitories were satisfied with the food being served in the cafeteria.

-An additional group of 10 students, not selected in the sample, asked to be included so that their opinions could be heard. Adding these students to the sample

A) is a good idea because the margin of error would decrease.
B) is a good idea because students who have a strong opinion about the food should be represented in the sample.
C) is not a good idea because the margin of error would increase.
D) is not a good idea because the resulting sample may not be representative of the population of all students living in dormitories.

Question 19

Use the following information for questions:
A random sample of 600 students was asked whether student fees should be used to construct a new football stadium.

-The conservative margin of error for the sample of 600 students is

A) 0.2%.
B) 4%.
C) 5%.
D) none of the above.

Question 20

Use the following information for questions:
A random sample of 600 students was asked whether student fees should be used to construct a new football stadium.

-An additional group of 5 football players, not selected in the sample, asked to be included so that their opinions could be heard. Adding these students to the sample

A) is a good idea because the margin of error would decrease.
B) is a good idea because students who have a strong opinion about the issue should be represented in the sample.
C) is not a good idea because the margin of error would increase.
D) is not a good idea because the resulting sample may not be representative of all students.

Question 21

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-The conservative margin of error for the sample of 249 students is

A) 0.4%.
B) 6.3%.
C) 14.1%.
D) none of the above.

Question 22

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-What type of study was this?

A) A randomized study
B) A case-control study
C) An observational study
D) A matched-pairs designed study

Question 23

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-What was the response variable?

A) Thickness of carotid artery wall
B) Diet
C) Age
D) Physical activity level

Question 24

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-What was the explanatory variable?

A) Thickness of carotid artery wall
B) Diet
C) Age
D) Physical activity level

Question 25

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-What was a potential confounding variable?

A) Thickness of carotid artery wall
B) Diet
C) Age
D) Physical activity level

Question 26

Use the following information for questions:
In a report by CNN Health News, the headlines read "Study: Junk food raises teens' risk of heart disease." The headlines were based on a study of 249 high school students between the ages of 13 and 18. One researcher said, "more than 80% of those tested were eating a diet that could promote heart disease." Ultrasound was used to measure the thickness of the carotid artery, which, the researcher said, "can be an early indicator of atherosclerosis, an abnormal hardening of the artery walls."

-Is a conclusion that a junk food diet causes an increased risk of heart disease justified?

A) Yes, because the study was a randomized study.
B) Yes, because many students in the study were overweight.
C) No, because the study was a matched pairs designed study.
D) No, because the study was an observational study.

Question 27

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-The conservative margin of error for the sample of 3,617 adults is

A) 0.5%.
B) 1.0%.
C) 1.7%.
D) none of the above.

Question 28

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-What type of study was this?

A) A randomized study
B) A case-control study
C) An observational study
D) A matched-pairs designed study

Question 29

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-What was the response variable?

A) Living in an urban environment (or not)
B) Gender (male or female)
C) Death
D) General physical health

Question 30

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-What was the explanatory variable?

A) Living in an urban environment (or not)
B) Gender (male or female)
C) Death
D) General physical health

Question 31

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-What was a potential confounding variable?

A) Living in an urban environment (or not)
B) Gender (male or female)
C) Death
D) General physical health

Question 32

Use the following information for questions:
In a report by ABC News, the headlines read "City Living Increases Men's Death Risk" The headlines were based on a study of 3,617 adults who lived in the United States and were more than 25 years old. One researcher said, "Elevated levels of tumor deaths suggest the influence of physical, chemical and biological exposures in urban areas… Living in cities also involves potentially stressful levels of noise, sensory stimulation and overload, interpersonal relations and conflict, and vigilance against hazards ranging from crime to accidents."

-Is a conclusion that living in an urban environment causes an increased risk of death justified?

A) Yes, because the study was a randomized study.
B) Yes, because many of the men in the study were under stress.
C) No, because the study was a retrospective study.
D) No, because the study was an observational study.

Question 33

Use the following information for questions:
In a report by AP, the headline read "Even light weekend use of the party drug Ecstasy might harm intelligence, a study suggests". The headline was based on a study of 28 Ecstasy users in their mid-20s recruited off the dance floor of a nightclub. The researchers compared this group with two others of the same size, age range, and education level - one consisting of drug-free people and another of people who smoked about as much marijuana as the Ecstasy group. The researchers reported weeks after partying, those who used Ecstasy along with marijuana performed worse on intelligence tests than people who just smoked marijuana or took no drugs at all.

-What type of study was this?

A) A randomized study
B) A matched-pair case-control study
C) An observational study
D) A retrospective study

Question 34

Use the following information for questions:
In a report by AP, the headline read "Even light weekend use of the party drug Ecstasy might harm intelligence, a study suggests". The headline was based on a study of 28 Ecstasy users in their mid-20s recruited off the dance floor of a nightclub. The researchers compared this group with two others of the same size, age range, and education level - one consisting of drug-free people and another of people who smoked about as much marijuana as the Ecstasy group. The researchers reported weeks after partying, those who used Ecstasy along with marijuana performed worse on intelligence tests than people who just smoked marijuana or took no drugs at all.

-What was the response variable?

A) Drug use (Ecstasy, Marijuana only, or no drugs)
B) Score on Intelligence Tests
C) Educational Level
D) IQ (intelligence level) before taking drugs

Question 35

Use the following information for questions:
In a report by AP, the headline read "Even light weekend use of the party drug Ecstasy might harm intelligence, a study suggests". The headline was based on a study of 28 Ecstasy users in their mid-20s recruited off the dance floor of a nightclub. The researchers compared this group with two others of the same size, age range, and education level - one consisting of drug-free people and another of people who smoked about as much marijuana as the Ecstasy group. The researchers reported weeks after partying, those who used Ecstasy along with marijuana performed worse on intelligence tests than people who just smoked marijuana or took no drugs at all.

-What was the explanatory variable?

A) Drug use (Ecstasy, Marijuana only, or no drugs)
B) Score on Intelligence Tests
C) Educational Level
D) IQ (intelligence level) before taking drugs

Question 36

Use the following information for questions:
In a report by AP, the headline read "Even light weekend use of the party drug Ecstasy might harm intelligence, a study suggests". The headline was based on a study of 28 Ecstasy users in their mid-20s recruited off the dance floor of a nightclub. The researchers compared this group with two others of the same size, age range, and education level - one consisting of drug-free people and another of people who smoked about as much marijuana as the Ecstasy group. The researchers reported weeks after partying, those who used Ecstasy along with marijuana performed worse on intelligence tests than people who just smoked marijuana or took no drugs at all.

-What was a potential confounding variable?

A) Drug use (Ecstasy, Marijuana only, or no drugs)
B) Score on Intelligence Tests
C) Educational Level
D) IQ (intelligence level) before taking drugs

Question 37

Use the following information for questions:
In a report by AP, the headline read "Even light weekend use of the party drug Ecstasy might harm intelligence, a study suggests". The headline was based on a study of 28 Ecstasy users in their mid-20s recruited off the dance floor of a nightclub. The researchers compared this group with two others of the same size, age range, and education level - one consisting of drug-free people and another of people who smoked about as much marijuana as the Ecstasy group. The researchers reported weeks after partying, those who used Ecstasy along with marijuana performed worse on intelligence tests than people who just smoked marijuana or took no drugs at all.

-Is a conclusion that taking the drug Ecstasy causes a lower score on intelligence tests warranted?

A) Yes, because the study was a randomized study.
B) Yes, because there was a control group.
C) No, because the study was a retrospective study.
D) No, because the study was an observational study.

Question 38

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-The conservative margin of error for the sample of 847 women is

A) 0.1%.
B) 3.4%.
C) 4.0%.
D) none of the above.

Question 39

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-What type of study was this?

A) A randomized study
B) A case-control study
C) An observational study
D) A retrospective study

Question 40

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-What was the response variable?

A) Breast cancer mortality
B) Emotional expression and emotional support
C) Race (black or white)
D) Stage of breast cancer at diagnosis

Question 41

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-What was the explanatory variable?

A) Breast cancer mortality
B) Emotional expression and emotional support
C) Race (black or white)
D) Stage of breast cancer at diagnosis

Question 42

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-What was a potential confounding variable?

A) Breast cancer mortality
B) Emotional expression and emotional support
C) Race (black or white)
D) Stage of breast cancer at diagnosis

Question 43

Use the following information for questions:
In a report by Reuters, the headline read "Emotional Support Helps Breast Cancer Survival" The headline was based on a study of 847 women (442 black women and 405 white women) with breast cancer and how they fared over a 9-year period. The authors wrote "Women in our study who reported low levels of both emotional expression and emotional support experienced 2 to 4 times greater breast cancer mortality than women reporting high levels of both".

-Is a conclusion that more emotional support causes a reduction in breast cancer mortality justified?

A) Yes, because the study was a randomized study.
B) Yes, because there were women who had a lot of emotional support as well as women who did not.
C) No, because the study was a retrospective study.
D) No, because the study was an observational study.

Question 44

A significance test based on a small sample may not produce a statistically significant result even if the true value differs substantially from the null value. This type of result is known as

A) the significance level of the test.
B) the power of the study.
C) a Type 1 error.
D) a Type 2 error.

Question 45

Sometimes researchers are surprised to find "no effect" in a study when in fact there really is an effect in the population. This type of result is known as

A) the significance level of the test.
B) the power of the study.
C) a Type 1 error.
D) a Type 2 error.

Question 46

Use the following information for questions:
An observational study found a statistically significant relationship between regular consumption of tomato products (yes, no) and development of prostate cancer (yes, no), with lower risk for those consuming tomato products.

-Which of the following is not a possible explanation for this finding?

A) Something in tomato products causes lower risk of prostate cancer.
B) There is a confounding variable that causes lower risk of prostate cancer, such as eating vegetables in general, that is also related to eating tomato products.
C) A large number of food products were measured to test for a relationship, and tomato products happened to show a relationship just by chance.
D) A large sample size was used, so even if there were no relationship, one would almost certainly be detected.

Question 47

Use the following information for questions:
An observational study found a statistically significant relationship between regular consumption of tomato products (yes, no) and development of prostate cancer (yes, no), with lower risk for those consuming tomato products.

-Which of the following is a valid conclusion from this finding?

A) Something in tomato products causes lower risk of prostate cancer.
B) Based on this study, the relative risk of prostate cancer, for those who do not consume tomato products regularly compared with those who do, is greater than one.
C) If a new observational study were to be done using the same sample size and measuring the same variables, it would find the same relationship.
D) Prostate cancer can be prevented by eating the right diet.

Question 48

Use the following information for questions:
An observational study found a statistically significant relationship between regular consumption of tomato products (yes, no) and development of prostate cancer (yes, no), with lower risk for those consuming tomato products.

-Setting aside the feasibility of doing so, suppose the results had been based on a randomized experiment (instead of an observational study), in which men were randomly assigned to consume or abstain from tomato products. Which of the following problems associated with observational studies would no longer be a problem?

A) The statistically significant reduced risk of prostate cancer could be caused by confounding variables instead of by consumption of tomato products.
B) The people who volunteered for the study may not be a representative sample.
C) The people in the study may lie about what they actually ate.
D) The participants were not blind to what they were eating.

Question 49

A small college wants to know what percentage of its current 2,500 students wants a switch to a quarter system. Every student in the college is polled with a question attached to the course registration form. What's the margin of error for the poll?

A) 5%
B) 2%
C) 0%
D) It depends on the observed sample proportion.

Question 50

A newspaper article reported that a large study measured the daily habits of a sample of mid-level managers. The response variable of interest was whether or not the person had a serious illness during the five years of the study. The investigators looked at 30 habits such as smoking, exercise, church attendance, and so on, and for each habit they tested the null hypothesis that the proportions with serious illness were the same for those with and without the habit. The newspaper reported, "Only one habit was significantly related to the risk of serious illness. The researchers were surprised to find that the managers who drank tea were significantly more likely to have a serious illness than managers who did not drink tea." The most likely explanation for this finding is that

A) drinking tea increases the chance of serious illness.
B) there is a confounding variable that both increases the chances of drinking tea and increases the chances of serious illness.
C) people who are prone to illness are likely to prefer tea to other beverages.
D) there were 30 habits examined, so at least one was likely to be "significant" just by chance, and drinking tea happened to be it.

Question 51

The best way to determine whether a statistically significant difference in two means is of practical importance is to

A) find a 95% confidence interval and notice the magnitude of the difference.
B) repeat the study with the same sample size and see if the difference is statistically significant again.
C) see if the p-value is extremely small.
D) see if the p-value is extremely large.

Question 52

A researcher compares two different treatments for depression. The result is not statistically significant so the researcher says that the two treatments are equally effective. Which one of the following statements is correct?

A) The researcher should have instead concluded that there is not sufficient evidence to conclude that the two treatments are different.
B) The researcher should have instead concluded that there is not sufficient evidence to conclude that the two treatments are equally effective.
C) A Type I error may have been made.
D) A Type I error was definitely made.

Question 53

CNN reported the results of a survey of 902 adult Americans taken to assess whether Americans think there should be stronger gun control laws. The margin of error was reported to be 3%. Which choice best describes how to interpret the 3% margin of error?

A) The sample and population percentages differ by less than 3%.
B) The sample and population percentages differ by more than 3%.
C) There is a 95% chance that the sample and population percentages differ by more than 3%
D) There is a 95% chance that the sample and population percentages differ by less than 3%.

Question 54

A large company examines the annual salaries for all of the men and women performing a certain job and finds that the means and standard deviations are $32,120 and $3,240, respectively, for the men and $34,093 and $3521, respectively, for the women. The best way to determine if there is a difference in mean salaries for the population of men and women performing this job in this company is

A) to compute a 95% confidence interval for the difference.
B) to subtract the two sample means.
C) to test the hypothesis that the population means are the same versus that they are different.
D) to test the hypothesis that the population means are the same versus that the mean for men is higher.

Question 55

The margin of error for the results of a survey takes into account

A) error due to biased wording of questions.
B) error due to the order in which questions are presented.
C) variability due to random sampling.
D) all of the above.

Question 56

One problem with hypothesis testing is that a real effect may not be detected. This problem is most likely to occur when

A) the effect is small and the sample size is small.
B) the effect is large and the sample size is small.
C) the effect is small and the sample size is large.
D) the effect is large and the sample size is large.

Question 57

One problem with hypothesis testing is that a result may be statistically significant but be of little practical significance. This problem is most likely to occur when

A) the effect is small and the sample size is small.
B) the effect is large and the sample size is small.
C) the effect is small and the sample size is large.
D) the effect is large and the sample size is large.

Question 58

A newspaper reported that researchers were surprised when a new study found no relationship between a certain food and a certain disease, because two previous studies had found a relationship. Which of the following is not likely to explain what happened?

A) The new study had lower power than the other studies, and thus failed to detect a real relationship.
B) The two prior studies each made a Type 2 error.
C) The two prior studies each made a Type 1 error.
D) The new study made a Type 2 error.

Question 59

Suppose you are faced with a decision about whether to take a new job or remain with the one you have. In an analogy to hypothesis testing, the alternative hypothesis is that you will be better off if you change jobs. Your friends are encouraging you to take the new job. In statistical terms, they think:

A) You should not reject the null hypothesis because a type 1 error would be more serious for you than a type 2 error.
B) You should not reject the null hypothesis because a type 2 error would be more serious for you than a type 1 error.
C) You should reject the null hypothesis because a type 1 error would be more serious for you than a type 2 error.
D) You should reject the null hypothesis because a type 2 error would be more serious for you than a type 1 error.

Question 60

Which of the following would not allow conclusions to be made about a population?

A) Using a random sample.
B) Using a random sample, but not randomly assigning treatments.
C) Using a volunteer sample, but randomly assigning treatments.
D) Measuring the whole population.

Question 61

A student doing a history project finds the ages of all of the governors of California at the time they were first elected and the ages of all the governors of Massachusetts at the time they were first elected, and then computes the mean for each state. The best way to determine if there is a difference in the mean age for the past governors of the two states is to

A) compute a 95% confidence interval for the difference.
B) compute a confidence interval for each mean and see if they overlap.
C) test the hypothesis that the means are the same versus that they are different.
D) subtract the two computed means and see whether the difference is 0.

Question 62

In a study for which the sample does not represent the population for the question of interest

A) the results cannot be extended to the population and a cause and effect conclusion cannot be made.
B) the results cannot be extended to the population, but a cause and effect conclusion can be made if the treatments were randomly assigned.
C) the results can be extended to the population if they are statistically significant.
D) the results can be extended to the population if the sample size is large enough.

Question 63

In a study for which the sample represents the population of interest but the treatments are not randomly assigned

A) the results can be extended to the population but a cause and effect conclusion cannot be made.
B) the results can be extended to the population and a cause and effect conclusion can be made if the results are statistically significant.
C) the results cannot be extended to the population unless they are statistically significant.
D) the results cannot be extended to the population unless the sample size is large.

Question 64

An observational study found a statistically significant relationship between an explanatory variable and a response variable that surprised the researchers. If you wanted to explore possible explanations, which of the following pieces of information would not be useful?

A) How many different relationships were tested in the study and how many were statistically significant.
B) How large the sample size was for the study.
C) Whether it was a case-control study or not.
D) A confidence interval for the relative risk.

Question 65

An observational study found that people who listened to classical music regularly had statistically significantly lower average blood pressure than those who didn't listen to classical music. Which one of the following headlines is justified?

A) Classical Music Lowers Blood Pressure.
B) Listening to Music Lowers Blood Pressure.
C) Link Found Between Listening to Classical Music and Lower Blood Pressure.
D) Blood Pressure Level Affects Taste in Music.

Question 66

Explain why a cause and effect relationship can be inferred from a randomized experiment, but not from an observational study.

Question 67

An ABC news poll over the Internet resulted in 17,251 responses. Explain why the results of this poll should not be generalized to the American population.

Question 68

You are asked to serve on a jury for a trial to decide whether a defendant is guilty of murder. You will decide either that the defendant is guilty or not guilty. Explain what your Type 1 and Type 2 errors are.

Question 69

A Gallup Poll in 1997 asked "What do you think is the ideal number of children for a family to have?" In a random sample of 1000 Americans, aged 18 or older, 50% responded that having 0 to 2 children was ideal. What is the conservative margin of error for this sample?

Question 70

Write down one of the ten guiding principles summarized in Chapter 17. Give an example of the principle.