Scribd
Upload
5K views

Topics: Confidence Intervals

The document discusses confidence intervals and surveys. It provides examples to test understanding of key concepts like population, parameter, sample size, margin of error, confidence level, and interpreting confidence intervals. It asks questions about these topics, including calculating required sample sizes to achieve a given margin of error at different confidence levels.

Uploaded by

Raja Ram Chaudhary
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
5K views

Topics: Confidence Intervals

The document discusses confidence intervals and surveys. It provides examples to test understanding of key concepts like population, parameter, sample size, margin of error, confidence level, and interpreting confidence intervals. It asks questions about these topics, including calculating required sample sizes to achieve a given margin of error at different confidence levels.

Uploaded by

Raja Ram Chaudhary
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Topics: Confidence Intervals

1. For each of the following statements, indicate whether it is True/False. If false,


explain why.

I. The sample size of the survey should at least be a fixed percentage of the
population size in order to produce representative results.
Ans) False.
The results depend on the size(n) of the sample. The sample size should have at least 30
observations.

II. The sampling frame is a list of every item that appears in a survey sample,
including those that did not respond to questions.
Ans) False.
The sampling frame is a list of all the items in the target population from which the
sample is selected.
III. Larger surveys convey a more accurate impression of the population than
smaller surveys.
Ans) True.
Large sample size will result in less standard deviation compared to small sample size.
Thus we can say larger sample is more accurate.

2. PC Magazine asked all of its readers to participate in a survey of their satisfaction


with different brands of electronics. In the 2004 survey, which was included in an
issue of the magazine that year, more than 9000 readers rated the products on a scale
from 1 to 10. The magazine reported that the average rating assigned by 225 readers
to a Kodak compact digital camera was 7.5. For this product, identify the following:

A. The population
B. The parameter of interest
C. The sampling frame
D. The sample size
E. The sampling design
F. Any potential sources of bias or other problems with the survey or sample
Ans) A: Readers of the magazine =9000
B: Rating of the camera(7.5)
C: Sampling frame: All readers of the issue where the survey was included.
D: 225
E: Voluntary response
F: It is possible that only those who were particularly pleased or only who are
displeased with the product participated in the survey which can makes the results
unreliable.

3. For each of the following statements, indicate whether it is True/False. If false,


explain why.
I. If the 95% confidence interval for the average purchase of customers at a
department store is $50 to $110, then $100 is a plausible value for the
population mean at this level of confidence.
Ans) True.
Confidence interval identifies the collection of values for the population parameter that
are consistent with the observed sample.

II. If the 95% confidence interval for the number of moviegoers who purchase
concessions is 30% to 45%, this means that fewer than half of all moviegoers
purchase concessions.
Ans) False.
We have evidence in that direction but we cannot confirm 100% based on this
data. We have to consider the values out of this range(i.e. more than 95% confidance
interval).

III. The 95% Confidence-Interval for μ only applies if the sample data are nearly
normally distributed.
Ans) False.
We should have a moderately large sample(usually at least larger than 30 for
many cases), the central limit theorem implies that the sampling distribution is normal
regardless of the data itself.

4. What are the chances that X >μ ?

A. ¼
B. ½
C. ¾
D. 1
Ans) B.
This is pure assumption. There is a 50% chance that the sample mean( X́ ) is
greater than the population mean(µ).

5. In January 2005, a company that monitors Internet traffic (WebSideStory) reported


that its sampling revealed that the Mozilla Firefox browser launched in 2004 had
grabbed a 4.6% share of the market.

I. If the sample were based on 2,000 users, could Microsoft conclude that
Mozilla has a less than 5% share of the market?
Ans) here,
X́ =0.046, n=2000, Z 95= 1.96, q= 0.954
95% confidence interval for the proportion of web users using Mozilla is
X́ q 0.046∗0.954

0.0551.
X́ ±
√
Z
n
= 0.046± 1.96
√ 2000
=0.046±0.00918 = 0.0368-
II. WebSideStory claims that its sample includes all the daily Internet users. If
that’s the case, then can Microsoft conclude that Mozilla has a less than 5%
share of the market?
Ans) In this case, we have data on the entire population and the sample value
accurately reflects the population number. Thus we can conclude that the share is less
than 5%.

6. A book publisher monitors the size of shipments of its textbooks to university


bookstores. For a sample of texts used at various schools, the 95% confidence
interval for the size of the shipment was 250 ± 45 books. Which, if any, of the
following interpretations of this interval are correct?

A. All shipments are between 205 and 295 books.


Ans) Incorrect.
The interval of (205,295) is for 95% confidence not for 100%.
B. 95% of shipments are between 205 and 295 books.
Ans) Incorrect.
The interval doesn’t describe individual shipments.
C. The procedure that produced this interval generates ranges that hold the
population mean for 95% of samples.
Ans) Correct.
95% of intervals created in this way contain the true population mean.
D. If we get another sample, then we can be 95% sure that the mean of this
second sample is between 205 and 295.
Ans) Incorrect.
The interval doesn’t describe the mean of another sample.
E. We can be 95% confident that the range 160 to 340 holds the population
mean.
Ans) Incorrect.
The interval doesn’t correspond to a 95% confidence level.

7. Which is shorter: a 95% z-interval or a 95% t-interval for μ if we know that σ =s?

A. The z-interval is shorter


B. The t-interval is shorter
C. Both are equal
D. We cannot say
Ans) A. the Z- interval is shorter.

Questions 8 and 9 are based on the following: To prepare a report on the economy,
analysts need to estimate the percentage of businesses that plan to hire additional
employees in the next 60 days.

8. How many randomly selected employers (minimum number) must we contact in


order to guarantee a margin of error of no more than 4% (at 95% confidence)?
A. 600
B. 400
C. 550
D. 1000
Ans) here, n=number of employers, Assume ^ P=0.5,q^ =0.5 Margin of Error=0.04
For 95% confidence interval, the critical value Z= 1.96
p^ q^
ME = Z *
√ n

0.5∗0.5
0.04 = 1.96 *
2
√
n
1.96 ∗0.5∗0.5 0.9604
n= = =600 =A
0.04 2 0.0016

9. Suppose we want the above margin of error to be based on a 98% confidence level.
What sample size (minimum) must we now use?

A. 1000
B. 757
C. 848
D. 543
Ans) Z= 2.576
0.5∗0.5
0.04 = 2.326 *
√ n

2.3262∗0.5∗0.5 1.3525
n= = = 845.35 = C
0.04 2 0.0016

You might also like