Lesson+8.2+Answer+Key+ +Intro+Stats+ +Stats+Medic
Lesson+8.2+Answer+Key+ +Intro+Stats+ +Stats+Medic
QUICK REVIEW:
Hate +
-
of error
Did your interval capture the true mean (the green line)?
yes
4. Use the slider to adjust the confidence level up to 99.5% What happened to the
intervals? Why?
They got wider We need
.
a wider interval in order to
be more confident it will capture the true mean .
variability .
n P , margin of error In
Lesson 8.2 – What Affects the Margin of Error
QuickNotes Interpret confidence level
"
LT# 1 to select random
If we were
many samples and construct a
LT#3
LT# 2 Margin of error
•
accounts for sampling variability
confidence level P , M E P
. .
•
does not account for non response,
T M E. b under
coverage , a response bias
n .
, .
Lesson App 8.2 Did you keep your New Year’s resolution?
At the end of 2019, Morning Consult surveyed a random sample of U.S. adults about New
Year’s Resolutions. Among the 655 people in the sample who made a resolution, 402 said
they had kept their resolution. The 95% confidence interval for the pro- portion of all
resolution-making U.S. adults who kept their resolution is 0.576 to 0.651.
If the Morning Consult selected many random samples of U.S. adults who made resolutions in 2019
and constructed a 95% confidence interval for each sample, about 95% of these intervals would
capture the true proportion of resolution–making U.S. adults who kept their resolution in 2019.
2. Name two actions Morning Consult could take to reduce the margin of error. What
drawbacks do these actions have?
Morning Consult could decrease the confidence level. The drawback is that we can’t be as confident
that our interval will capture the true propoportion. Morning Consult could also increase the sample
size. The drawback is that larger samples cost more time and money to obtain.
3. Describe one potential source of bias in this survey. Does the stated margin of error
account for this possible bias? Explain your answer.
The results of this survey may suffer from response bias. Respondents may be embarrassed to admit
that they did not keep their resolution and may be tempted to report that they kept their resolution
when, in reality, they did not. Response bias is not accounted for by the margin of error because the
margin of error accounts only for sampling variability.