1.

Question 1
Data from the year 2012 National Health and Nutrition Examination Survey (NHANES) survey includes
laboratory measurements on a random sample of more than four thousand 18-65 year old persons in
the United States. The following boxplots display the distribution of blood lead levels (mg/dL) for this
sample, separately for males and females. The accompanying tables display summary statistics of the
sample values.

Suppose you were able to take a random sample of 100 females from the same population of females as
the sample of 2,128. How would the sample standard deviation estimate for these 100 sample blood
lead level measurements (s100) compare, in value, to the sample standard deviation of the original
2,128 blood lead level measurements (1.16 microgram/dL)?

1 point

s100 <1.16

s100 > 1.16

s100 = 1.16/sqrt(100)

s100 should be similar in value to 1.16, but there is no way to predict exactly how the two estimates
will compare numerically

2.
Question 2
Data from the year 2012 National Health and Nutrition Examination Survey (NHANES) survey includes
laboratory measurements on a random sample of more than four-thousand 18-65 year old persons in
the United States. The following boxplots display the distribution of blood lead levels(mg/dL) for this
sample, separately for males and females. The accompanying tables display summary statistics of the
sample values.

Which of the following best describes the shape of the lead level distributions in both sex groups?

1 point

left (negatively) skewed

right (positively skewed)

Approximately normal
uniform

3.
Question 3
Data from the year 2012 National Health and Nutrition Examination Survey (NHANES) survey includes
laboratory measurements on a random sample of more than four-thousand 18-65 year old persons in
the United States. The following boxplots display the distribution of blood lead levels (mg/dL) for this
sample, separately for males and females. The accompanying tables display summary statistics of the
sample values.

What is the mean difference in blood lead levels for females compared to males?

1 point

0.7 micrograms/dL

-0.7 micrograms/dL

0.45 micrograms/dL

-0.45 micrograms/dL

4.
Question 4
Data from the year 2012 National Health and Nutrition Examination Survey (NHANES) survey includes
laboratory measurements on a random sample of more than four-thousand 18 - 65 year old persons in
the United States. The following boxplots display the distribution of blood lead levels (mg/dL) for this
sample, separately for males and females. The accompanying tables display summary statistics of the
sample values.

Suppose, based on these results, researchers decide to make the cutoff for high blood lead levels at 3.10
micrograms/dL. A binary variable is created such that a value of 1 indicates that an individual’s blood
lead level is greater than 3.10 micrograms/dL, and a value of 0 indicates that an individual’s HDL blood
lead level is less than or equal to 3.10 micrograms/dL. What percentage of the females would have a
value of 1 for this binary indicator? (you may assume there are no repeated values in these data)

1 point

2.5%
16%

5%

This cannot be answered without having access to a standard normal table

5.
Question 5
The relative risk of having high blood lead levels (greater than 3.10 micrograms/dL) for males compared
to females is 3.0. What percentage of males have high blood lead levels?

1 point

-0.7%

15%

10%

6.
Question 6
An article from the American Journal of Public Health reports the results from a randomized study
designed to evaluate the efficacy of an intervention targeted to Hispanic/Latino men who identify as
gay, bi-sexual or other men who have sex with men (MSM).

A representative sample of 254 such men was randomized to be in either the intervention group (n=152)
or the control group (n=152). The primary outcome under study getting tested for HIV within the six-
months following group assignment (randomization) among those who had been sexually active in this
same six-month follow-up period.

At six months of follow-up, 141 subjects in the intervention group reported having had sex (with men
and/or women) since randomization. Of these 141 men, 114 had been tested for HIV since being
randomized.

At six months of follow-up, 147 subjects in the control group reported having had sex (with men and/or
women) since randomization. Of these 147 men, 40 had been tested for HIV since being randomized.

What is the (approximate) estimated

1 point
Enter answer here= 3.0
7.
Question 7
A 2016 article in JAMA reports the results of a study of treatment outcomes for children with mild
gastroenteritis who were given oral rehydration therapy. Enrolled children were randomized to
received either rehydration with diluted apple juice (DAJ), or an electrolyte maintenance solution (EMS).
As per the study authors:

“The primary outcome was a composite of treatment failure defined by any of the following occurring
within 7 days of enrollment: intravenous rehydration, hospitalization, subsequent unscheduled
physician encounter, protracted symptoms, crossover, and 3%or more weight loss or significant
dehydration at in-person follow-up. Secondary outcomes included intravenous rehydration,
hospitalization, and frequency of diarrhea and vomiting.”

Of the 323 children randomized to DAJ, 54 experienced treatment failure. (17 %). Of he 324 children
randomized to EMS, 81 experienced treatment failure. (25 %)

Estimate the risk difference (difference in proportions) of treatment failure for children in the DAJ group
compared to children in the EMS group. (DAJ-EMS). Please report as a decimal, not a percentage.

1 point
Enter answer here= -0.08
8.
Question 8
A 2016 article in JAMA reports the results of a study of treatment outcomes for children with mild
gastroenteritis who were given oral rehydration therapy. Enrolled children were randomized to
received either rehydration with diluted apple juice (DAJ), or an electrolyte maintenance solution (EMS).
As per the study authors:

“The primary outcome was a composite of treatment failure defined by any of the following occurring
within 7 days of enrollment: intravenous rehydration, hospitalization, subsequent unscheduled
physician encounter, protracted symptoms, crossover, and 3%or more weight loss or significant
dehydration at in-person follow-up. Secondary outcomes included intravenous rehydration,
hospitalization, and frequency of diarrhea and vomiting.”

Of the 323 children randomized to DAJ, 54 experienced treatment failure. (17 %). Of he 324 children
randomized to EMS, 81 experienced treatment failure. (25 %)

Estimate the relative risk (risk ratio) of treatment failure for children receiving DAJ compared to children
receiving EMS.

1 point
Enter answer here= 0.68
9.
Question 9
An article from the American Journal of Public Health reports the results from a randomized study
designed to evaluate the efficacy of an intervention targeted to Hispanic/Latino men who identify as
gay, bi-sexual or other men who have sex with men (MSM).
A representative sample of 254 such men was randomized to be in either the intervention group (n=152)
or the control group (n=152). The primary outcome under study getting tested for HIV within the six-
months following group assignment (randomization) among those who had been sexually active in this
same six-month follow-up period.

At six months of follow-up, 141 subjects in the intervention group reported having had sex (with men
and/or women) since randomization. Of these 141 men, 114 had been tested for HIV since being
randomized.

At six months of follow-up, 147 subjects in the control group reported having had sex (with men and/or
women) since randomization. Of these 147 men, 40 had been tested for HIV since being randomized.

The difference in proportions of men being tested in the intervention group compared to the control
group is 0.54 (54%). Suppose this intervention were used in a community with 1,000 Hispanic/Latino
men who identify as gay, bi-sexual or other men who have sex with men (MSM). What would be the
expected effect on HIV testing outcomes?

1 point

There would be an estimated 540 more men getting tested for HIV (in the six months following the
intervention) than if the intervention was not given.

10.
Question 10
Consider studies designed to compare the occurrence of a binary outcome between two populations:
population A and population B. In general, which of the following statements best describes the
relationship between the relative risk estimate ( RR_hat ) and the odds ratio estimate (OR_hat ), both
based on the same two samples from populations A and B?

1 point

RR_hat and OR_hat may differ in value, but will show the same direction of association.

RR_hat and OR_hat will always be equal in value.

OR_hat = 1/RR_hat

If RR_hat is greater than 1, then OR_hat will be less than 1.

Honor Code Agreement