Question ID 85939da5

Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Inference from

and Data Analysis sample statistics
and margin of error

ID: 85939da5

Does not talk

Texting Talks on cell on cell phone
behavior phone daily daily Total

Light 110 146 256

Medium 139 164 303

Heavy 166 74 240

Total 415 384 799

In a study of cell phone use, 799 randomly selected US teens were asked how
often they talked on a cell phone and about their texting behavior. The data are
summarized in the table above. Based on the data from the study, an estimate
of the percent of US teens who are heavy texters is 30% and the associated
margin of error is 3%. Which of the following is a correct statement based on
the given margin of error?

A. Approximately 3% of the teens in the study who are classified as heavy texters are not really heavy texters.

B. It is not possible that the percent of all US teens who are heavy texters is less than 27%.

C. The percent of all US teens who are heavy texters is 33%.

D. It is doubtful that the percent of all US teens who are heavy texters is 35%.

ID: 85939da5 Answer

Correct Answer: D
Rationale
Choice D is correct. The given margin of error of 3% indicates that the actual percent of all US teens who are heavy
texters is likely within 3% of the estimate of 30%, or between 27% and 33%. Therefore, it is unlikely, or doubtful, that the
percent of all US teens who are heavy texters would be 35%.

Choice A is incorrect. The margin of error doesn’t provide any information about the accuracy of reporting in the study.
Choice B is incorrect. Based on the estimate and given margin of error, it is unlikely that the percent of all US teens who
are heavy texters would be less than 27%, but it is possible. Choice C is incorrect. While the percent of all US teens who
are heavy texters is likely between 27% and 33%, any value within this interval is equally likely. We can’t be certain that
the value is exactly 33%.

Question Difficulty: Hard

Question ID c7e73ece
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Inference from

and Data Analysis sample statistics
and margin of error

ID: c7e73ece
Views on Nuclear Energy
Use
Response Frequency

Strongly
56
favor

Somewhat
214
favor

Somewhat
104
oppose

Strongly
37
oppose

A researcher interviewed 411 randomly selected US residents and asked about their views on the use of nuclear energy.
The table above summarizes the responses of the interviewees. If the population of the United States was 300 million
when the survey was given, based on the sample data for the 411 US residents, what is the best estimate, in millions, of
the difference between the number of US residents who somewhat favor or strongly favor the use of nuclear energy and
the number of those who somewhat oppose or strongly oppose it? (Round your answer to the nearest whole number.)

ID: c7e73ece Answer

Rationale
The correct answer is 94. Of those interviewed, “strongly favor” or “somewhat favor” the use of nuclear
energy, and interviewees “somewhat oppose” or “strongly oppose” the use of nuclear energy. The
difference between the sizes of the two surveyed groups is . The proportion of this difference among

the entire group of interviewees is . Because the sample of interviewees was selected at random from US
residents, it is reasonable to assume that the proportion of this difference is the same among all US residents as in the
sample. Therefore, the best estimate, in millions, of the difference between the number of US residents who somewhat
favor or strongly favor the use of nuclear energy and the number of those who somewhat oppose or strongly oppose it

is , which to the nearest million is 94.

Question Difficulty: Hard
Question ID 1ea09200
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Evaluating

and Data Analysis statistical claims:
Observational
studies and
experiments

ID: 1ea09200

A sample of 40 fourth-grade students was selected at random from a certain

school. The 40 students completed a survey about the morning
announcements, and 32 thought the announcements were helpful. Which of the
following is the largest population to which the results of the survey can be
applied?

A. The 40 students who were surveyed

B. All fourth-grade students at the school

C. All students at the school

D. All fourth-grade students in the county in which the school is located

ID: 1ea09200 Answer

Correct Answer: B

Rationale
Choice B is correct. Selecting a sample of a reasonable size at random to use for a survey allows the results from that
survey to be applied to the population from which the sample was selected, but not beyond this population. In this case,
the population from which the sample was selected is all fourth-grade students at a certain school. Therefore, the
results of the survey can be applied to all fourth-grade students at the school.

Choice A is incorrect. The results of the survey can be applied to the 40 students who were surveyed. However, this isn’t
the largest group to which the results of the survey can be applied. Choices C and D are incorrect. Since the sample was
selected at random from among the fourth-grade students at a certain school, the results of the survey can’t be applied
to other students at the school or to other fourth-grade students who weren’t represented in the survey results. Students
in other grades in the school or other fourth-grade students in the country may feel differently about announcements
than the fourth-grade students at the school.

Question Difficulty: Hard

Question ID 308084c5
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Inference from

and Data Analysis sample statistics
and margin of error

ID: 308084c5

Sample Percent in favor Margin of error

A 52% 4.2%

B 48% 1.6%

The results of two random samples of votes for a proposition are shown above.
The samples were selected from the same population, and the margins of error
were calculated using the same method. Which of the following is the most
appropriate reason that the margin of error for sample A is greater than the
margin of error for sample B?

A. Sample A had a smaller number of votes that could not be recorded.

B. Sample A had a higher percent of favorable responses.

C. Sample A had a larger sample size.

D. Sample A had a smaller sample size.

ID: 308084c5 Answer

Correct Answer: D

Rationale
Choice D is correct. Sample size is an appropriate reason for the margin of error to change. In general, a smaller sample
size increases the margin of error because the sample may be less representative of the whole population.

Choice A is incorrect. The margin of error will depend on the size of the sample of recorded votes, not the number of
votes that could not be recorded. In any case, the smaller number of votes that could not be recorded for sample A
would tend to decrease, not increase, the comparative size of the margin of error. Choice B is incorrect. Since the
percent in favor for sample A is the same distance from 50% as the percent in favor for sample B, the percent of
favorable responses doesn’t affect the comparative size of the margin of error for the two samples. Choice C is
incorrect. If sample A had a larger margin of error than sample B, then sample A would tend to be less representative of
the population. Therefore, sample A is not likely to have a larger sample size.

Question Difficulty: Hard

Question ID 7d68096f
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Evaluating

and Data Analysis statistical claims:
Observational
studies and
experiments

ID: 7d68096f

A trivia tournament organizer wanted to study the relationship between the number of points a team scores in a trivia
round and the number of hours that a team practices each week. For the study, the organizer selected teams at
random from all trivia teams in a certain tournament. The table displays the information for the teams in the sample
that practiced for at least hours per week.

Number of points per round

Hours practiced
6 to 13 points 14 or more points Total

3 to 5 hours

More than 5 hours

Total
Which of the following is the largest population to which the results of the study can be generalized?

A. All trivia teams in the tournament that scored or more points in the round

B. The trivia teams in the sample

C. The trivia teams in the sample that practiced for at least hours per week

D. All trivia teams in the tournament

ID: 7d68096f Answer

Correct Answer: D
Rationale
Choice D is correct. It's given that the organizer selected teams at random from all trivia teams in the tournament. A
table is also given displaying the information for the teams in the sample that practiced for at least hours per
week. Selecting a sample of a reasonable size at random to use for a survey allows the results from that survey to be
applied to the population from which the sample was selected, but not beyond this population. Thus, only the sampling
method information is necessary to determine the largest population to which the results of the study can be
generalized. Since the organizer selected the sample at random from all trivia teams in the tournament, the largest
population to which the results of the study can be generalized is all trivia teams in the tournament.

Choice A is incorrect. The sample was selected at random from all trivia teams in the tournament, not just from the
teams that scored an average of or more points per round.

Choice B is incorrect. If a study uses a sample selected at random from a population, the results of the study can be
generalized to the population, not just the sample.

Choice C is incorrect. If a study uses a sample selected at random from a population, the results of the study can be
generalized to the population, not just a subset of the sample.

Question Difficulty: Hard

Question ID 7ce2830a
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Evaluating

and Data Analysis statistical claims:
Observational
studies and
experiments

ID: 7ce2830a

A psychologist designed and conducted a study to determine whether playing a

certain educational game increases middle school students’ accuracy in adding
fractions. For the study, the psychologist chose a random sample of 35
students from all of the students at one of the middle schools in a large city.
The psychologist found that students who played the game showed significant
improvement in accuracy when adding fractions. What is the largest group to
which the results of the study can be generalized?

A. The 35 students in the sample

B. All students at the school

C. All middle school students in the city

D. All students in the city

ID: 7ce2830a Answer

Correct Answer: B

Rationale
Choice B is correct. The largest group to which the results of a study can be generalized is the population from which
the random sample was chosen. In this case, the psychologist chose a random sample from all students at one
particular middle school. Therefore, the largest group to which the results can be generalized is all the students at the
school.

Choice A is incorrect because this isn’t the largest group the results can be generalized to. Choices C and D are
incorrect because these groups are larger than the population from which the random sample was chosen. Therefore,
the sample isn’t representative of these groups.

Question Difficulty: Hard

Question ID aa43b41f
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Evaluating

and Data Analysis statistical claims:
Observational
studies and
experiments

ID: aa43b41f

Near the end of a US cable news show, the host invited viewers to respond to a
poll on the show’s website that asked, “Do you support the new federal policy
discussed during the show?” At the end of the show, the host reported that 28%
responded “Yes,” and 70% responded “No.” Which of the following best explains
why the results are unlikely to represent the sentiments of the population of the
United States?

A. The percentages do not add up to 100%, so any possible conclusions from the poll are invalid.

B. Those who responded to the poll were not a random sample of the population of the United States.

C. There were not 50% “Yes” responses and 50% “No” responses.

D. The show did not allow viewers enough time to respond to the poll.

ID: aa43b41f Answer

Correct Answer: B

Rationale
Choice B is correct. In order for the poll results from a sample of a population to represent the entire population, the
sample must be representative of the population. A sample that is randomly selected from a population is more likely
than a sample of the type described to represent the population. In this case, the people who responded were people
with access to cable television and websites, which aren’t accessible to the entire population. Moreover, the people who
responded also chose to watch the show and respond to the poll. The people who made these choices aren’t
representative of the entire population of the United States because they were not a random sample of the population of
the United States.

Choices A, C, and D are incorrect because they present reasons unrelated to whether the sample is representative of the
population of the United States.

Question Difficulty: Hard

Question ID 4a422e3e
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Evaluating

and Data Analysis statistical claims:
Observational
studies and
experiments

ID: 4a422e3e

To determine the mean number of children per household in a community,

Tabitha surveyed 20 families at a playground. For the 20 families surveyed, the
mean number of children per household was 2.4. Which of the following
statements must be true?

A. The mean number of children per household in the community is 2.4.

A determination about the mean number of children per household in the community should not be made because
B. the sample size is too small.

The sampling method is flawed and may produce a biased estimate of the mean number of children per household
C. in the community.

The sampling method is not flawed and is likely to produce an unbiased estimate of the mean number of children
D. per household in the community.

ID: 4a422e3e Answer

Correct Answer: C

Rationale
Choice C is correct. In order to use a sample mean to estimate the mean for a population, the sample must be
representative of the population (for example, a simple random sample). In this case, Tabitha surveyed 20 families in a
playground. Families in the playground are more likely to have children than other households in the community.
Therefore, the sample isn’t representative of the population. Hence, the sampling method is flawed and may produce a
biased estimate.

Choices A and D are incorrect because they incorrectly assume the sampling method is unbiased. Choice B is incorrect
because a sample of size 20 could be large enough to make an estimate if the sample had been representative of all the
families in the community.

Question Difficulty: Hard

Question ID 9ba3e283
Assessment Test Domain Skill Difficulty

SAT Math Problem-Solving Inference from

and Data Analysis sample statistics
and margin of error

ID: 9ba3e283

In State X, Mr. Camp’s eighth-grade class consisting of 26 students was

surveyed and 34.6 percent of the students reported that they had at least two
siblings. The average eighth‑grade class size in the state is 26. If the students in
Mr. Camp’s class are representative of students in the state’s eighth-grade
classes and there are 1,800 eighth-grade classes in the state, which of the
following best estimates the number of eighth‑grade students in the state who
have fewer than two siblings?

A. 16,200

B. 23,400

C. 30,600

D. 46,800

ID: 9ba3e283 Answer

Correct Answer: C

Rationale
Choice C is correct. It is given that 34.6% of 26 students in Mr. Camp’s class reported that they had at least two siblings.
Since 34.6% of 26 is 8.996, there must have been 9 students in the class who reported having at least two siblings and
17 students who reported that they had fewer than two siblings. It is also given that the average eighth-grade class size
in the state is 26 and that Mr. Camp’s class is representative of all eighth-grade classes in the state. This means that in
each eighth-grade class in the state there are about 17 students who have fewer than two siblings. Therefore, the best
estimate of the number of eighth-grade students in the state who have fewer than two siblings is 17 × (number of
eighth-grade classes in the state), or .

Choice A is incorrect because 16,200 is the best estimate for the number of eighth-grade students in the state who have
at least, not fewer than, two siblings. Choice B is incorrect because 23,400 is half of the estimated total number of
eighth-grade students in the state; however, since the students in Mr. Camp’s class are representative of students in the
eighth-grade classes in the state and more than half of the students in Mr. Camp’s class have fewer than two siblings,
more than half of the students in each eighth-grade class in the state have fewer than two siblings, too. Choice D is
incorrect because 46,800 is the estimated total number of eighth-grade students in the state.

Question Difficulty: Hard