10.1 Power Point Part 2

Confidence Intervals:
The Basics
Section 10.1 continued
Let’s See It Again
http://bcs.whfreeman.com/tps3e/defa
ult.asp?s=&n=&i=&v=&o=&ns=0&
uid=0&rau=0
Steps in Creating a
CI
1. Name the Procedure
2. Check the Conditions
3. Do the Math
4. Interpretation in Context
1. Name the Procedure
A Confidence Interval for a
Population Mean (σ Known)
2. Check the Conditions
• The data are from an SRS from
the population of interest.
• The sampling distribution of x is
approximately Normal
• Individual observations are
independent (when sampling w/o
replacement N ≥ 10n)
It Is a “SIN” to Not
Check the Conditions
• S – SRS
• I – Independent Observations
(N ≥ 10n)
• N – Sampling Distribution is ~N
3. Do the Math
• Find the critical value, z*, based
on the confidence level, C
• Calculate the standard deviation

of the sampling distribution,n
  
• Create the CI: x  z *  
 n
Finding z*
Some Common Critical
Values
• For a 95% CI, use z* = 1.96
• For a 90% CI, use z* = 1.64
• For a 99% CI, use z* = 2.58
– If not stated, assume C is 95%.

4. Interpretation in Context
• Use the phrasing: “Based on this
sample, I am __% confident that
the true mean of all
__________________________
is between ____ and ____.”
• Be sure that your conclusion is in

context!!!
Not To Be Confused With…
• Do not confuse the interpretation of the
confidence interval you create with the
interpretation of the confidence level, C.
• __% confidence means that if we were
to take many, many samples from the
population and create confidence
intervals from each, we would expect __
% of those intervals to include the value
of the true population parameter.
Correct Language
• If you construct one hundred 95% confidence
intervals, you expect the population proportion,
p, will be in 95 of them.
• This statement is not as obvious at it may seem
at first.
• It is correct to say that we expect the true value
of p to be in 95 out of every 100 of the 95%
confidence intervals that we construct.
• It is not correct, however, to say that after we
have found a confidence interval, that there is a
95% chance that p is in the confidence interval.
Let’s Try:
You take an SRS of 50
SAT scores from first-
year students from a
university with standard
deviation 150 and get a
sample mean of 1110.
Find and interpret a 95%
confidence interval for
the unknown mean.
Try, Try Again
A manufacturer of high-resolution video terminals
must control the tension on the mesh of fine
wires that lies behind the surface of the viewing
screen. Too much tension will tear the mesh and
too little will allow wrinkles. The tension is
measured by an electrical device with output
reading in millivolts (mV). Some variation is
inherent in the production process. Careful study
has shown that when the process is operating
properly, the standard deviation of the tension
readings is σ = 43 mV.
Here are the tension readings from an SRS of
20 screens from a single day’s production:
269. 297. 269. 283. 304.
5 0 6 3 8
280. 233. 257. 317. 327.
4 5 4 5 4
264. 307. 310. 343. 328.
7 7 0 3 1
Construct and interpret a 90% confident
342. for
interval 338.the340.
mean 374. 336.
tension of all screens
6 8 on 1this day.
μ produced 6 1
Margin of Error
• The margin of error in a confidence
  
interval is z *   .
 n
• We would like the margin of error to be

small and the confidence level to be high.
High confidence says that our method
almost always gives correct results. A
small margin of error says that we have
pinned down the parameter very precisely.
Shrinkage!
• How can we decrease the margin of
error?
• Use a smaller confidence level, C.
• Use a smaller σ.
• Increase the sample size, n.

Plan Smart!
A wise user of statistics never plans
data collection without planning
the inference at the same time. You
can have both high confidence and
a small margin of error if you take
enough observations.
You want to estimate the mean
resting pulse rate for all
athletes, and you know that for
many populations the
distribution of resting pulse
rates is roughly normal with σ
≈ 10 bpm. How many athletes
would you need in an SRS if
you want a 95% confidence
interval to have width 1 bpm?
Cautions
• Always round up when finding n.
• n, not N, determines the MoE (as long as N ≥
10n.)
• The data must be an SRS from the population.
(Formulas cannot rescue badly produced data.)
• Different methods are needed for different
designs.
• The shape of the population matters and outliers
can distort the results.
• The MoE in a CI only covers random sampling
errors.

10.1 Power Point Part 2

Uploaded by

Copyright:

Available Formats

10.1 Power Point Part 2

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

10.1 Power Point Part 2

Uploaded by

Copyright:

Available Formats

Confidence Intervals:

• For a 90% CI, use z* = 1.64

• For a 99% CI, use z* = 2.58

– If not stated, assume C is 95%.

• Be sure that your conclusion is in

• We would like the margin of error to be

• Increase the sample size, n.

You might also like