Practice for Test 2

1. A random sample of 20 weights is taken from babies born at Northside Hospital. A mean of 6.87lb and a
standard deviation of 1.76lb were found for the sample. Estimate with a 95% confidence interval the mean
weight of all babies born on this hospital. What is the margin of error?
2. A recent study is designed to estimate the proportion of crimes related to firearms in an area with one of the
highest crime rates in the country. The study samples 300 randomly selected files and determines that a
firearm was reportedly used in 180 of them. Construct a 95% confidence interval for the true proportion
of crimes committed in the area in which some type of firearm is reported used. What is the margin of
error?
3. We want to create a 99% confidence interval for the proportion of students who typically spend more than
$20 on alcoholic beverages during a meal. If ultimately the confidence interval needs to have a margin of
error of 3%, how many students do we need survey?
4. A company wishes to study how long it takes to fill orders. A random sample from the company’s files show
that orders were filled in 12, 10, 17, 14, 13, 18, 11, and 9 days. Estimate with a 90% confidence interval the
mean time it takes for the company to fill orders.
5. We will be creating a 90% confidence interval for the mean. Our sample has 35 observations in it. What
are the degrees of freedom? What value of t* (the critical value) will be used in the calculation?
6. We will be creating a 90% confidence interval for the proportion. What value of z* will be used in the
calculation? If we switch to a 98% confidence interval, will the margin of error change? If so, how?
7. You are asked to investigate the mean vacation days employees take during a given year. You are able to
randomly sample a number of employee files and find the mean number of days in the sample is 12.2 days.
You report that a 98% confidence interval for the mean number of vacation days is from 10.1 to 14.3
days. What is the margin of error for this confidence interval?
8. The housing market recovered slowly from the economic crisis of 2008. Recently, in a large community,
realtors randomly samples 36 bids from potential buyers to estimate the average loss in home value. The
sample showed an average loss from the peak was $9560 with a standard deviation of $1500.
a. What assumptions must be checked? Check them.
b. How many degrees of freedom does this CI have?
c. Find the 95% confidence interval for the mean loss in value per home.
9. In the past, the mean running time for a certain type of flashlight battery has been 8.5 hours. The
manufacturer has introduced a change in the production method and want to perform a hypothesis test to
determine whether the mean running time has changed as a result. Which will be correct?
a. H0 µ = 8.5 and Ha µ ≠ 8.5
b. H0 p ≤ 8.5% and Ha p > 8.5%
c. H0 µ ≥ 8.5 and Ha µ < 8.5
d. H0 µ ≠ 0.085 and Ha µ = 0.085
e. H0 p = 8.5 and Ha p ≠ 8.5
10. The owner of a football team claims that the average attendance at games is over 74,900 and he is
therefore justified in moving the team to a city with a larger stadium. An independent investigator will
conduct a hypothesis test to determine whether his claim is accurate. Which will be correct?
a. H0 µ = 74,900 and Ha µ ≠ 74,900
b. H0 P ≤ 74,900 and Ha P > 74,900
c. H0 µ = 74,900 and Ha µ > 74,900
d. H0 P = 0.74900 and Ha P > 0.74900
e. H0 µ > 74,900 and Ha µ = 74,900
11. Suppose that a manufacturer is testing one of its machines to make sure that the machine is producing
more than 97% good parts. What are the null & alternate hypotheses?
12. A candy company claims that 20% of their candies are purple. Purple is my favorite color and I just don’t
believe their claim. My bags seem to always have less way fewer purple! In fact, the bags that I opened
actually has only 17 of the 102 candies were purple! What are the null & alternate hypotheses? What level
of significance will you use? What is the test statistic & p-value? What is the decision? What conclusion is
made?
13. In 1960, census results indicated that the age at which American men first married had a mean of 23.3
years. It is wildly suspected that young people are waiting longer to marry. We sample 40 men and find a
sample average of 24.2 years and sample standard deviation of 5.3 years. What are the null & alternate
hypotheses? What level of significance will you use? What is the test statistic & p-value? What is the
decision? What conclusion is made?
14. Explain what the p-value in the above question means.
15. In the 1980’s it was generally thought that congenital abnormalities affected 5% of the nation’s children.
Some people believe that there has been an increase in the incidence of abnormalities. A recent study of
384 children found 34 of them showed signs of an abnormality. Is there strong evidence that the incidence
rate has increased? What are the null & alternate hypotheses? What level of significance will you use?
What is the test statistic & p-value? What is the decision? What conclusion is made?
16. A study is designed to determine if the mean time college students spend doing homework each week
exceeds 12 hours (per week). The resulting p-value is 0.0504. What decision would you make? What
would be your conclusion? [Hint: think about what the null & alternate hypotheses would be.]
17. A p-value indicates:
a. The probability that the null hypothesis is true
b. The probability that the alternative hypothesis is true
c. The probability of the observed test statistic given that the null hypothesis is true
d. The probability of the observed test statistic given that the alternative hypothesis is true
18. Suppose that a manufacturer is testing one of its machines to make sure that the machine is producing
more than 97% good parts (H0: p = 0.97 vs Ha: p > 0.97). The test results in a p-value of 0.122. What is
the decision and conclusion?
19. A can of pumpkin pie mix is labeled with its weight. A small hypothesis test using 4 cans is conducted to
determine if the average exceeds 96 ounces. [We believe that the distribution of weights is normal and we
have a random independent sample.] The resulting p-value is 0.047. What is the decision and conclusion?
20. A Type I error occurs if you reject the null hypothesis but the null was false. TRUE / FALSE
21. When making a decision, you compare the test statistic to the p-value. TRUE / FALSE
22. If p is less than α, your decision is that H0 is true. TRUE / FALSE
23. Conclusions are written in terms of H0 – is it true or not true. TRUE / FALSE

24. Has the percentage of young girls drinking milk changed over time? The following table is consistent with
the results from “Beverage Choices of Young Females: Changes and Impact on Nutrient Intakes” (Shanthy
A. Bowman, Journal of the American Dietetic Association, 102(9), pp. 1234-1239):

Nationwide Food Survey Years

1987-1988 1989-1991 1994-1996 Total
Drinks Fluid Milk Yes 354 502 366 1222
No 226 335 366 927
Total 580 837 732 2149
Find the following:
a. What percent of the young girls reported that they drink milk? ___________________
b. What percent of the young girls were in the 1989-1991 survey? ___________________
c. What percent of young girls reported that they drink milk and were in the 1989-1991 survey? ___________________
d. What percent of the young girls who reported that they drink milk were in the 1989-1991survey? ____________
e. What percent of the young girls in 1989-1991 reported they drink milk? ____________

25. The true proportion of students who enjoy statistics is called the ‘population statistic’. TRUE/FALSE
ANSWERS:

1. We will use t – “weights” are typically symmetric and this is a random sample (and
presumably independent since one newborn weight should be independent from another).
Stat  t-Stats  One Sample  With Summary  Confidence Interval for μ (sample
mean= 6.87; sample std dev= 1.76; sample size=20) to find (6.0463, 7.6937) A 95%
confidence interval for the true mean weight of all babies born in
this hospital is from 6.043 to 7.69 lbs. The margin of error = (upper
– lower) / 2 = ±0.8237 lb
2. We will use z – we should have at least 10 successes & 10 failures & the sample was selected randomly
Stat  Proportion Stats  One Sample  With Summary  Confidence Interval for p
(successes = 180, observations = 300):
(0.54456, 0.65544). A 95% confidence interval for the true
proportion of crimes committed in the area in which some type of
firearm is reported used is from 0.54456 to 0.65544. The margin of
error = (upper – lower) / 2 = ±0.05544 (or ±5.54%).

3. Since we don’t have any previous information about what proportion of students typically
spend more than $20 on alcoholic beverages during a meal, we will estimate our target p to
be 0.5
Stat  Proportion Stats  One Sample  Width/Sample Size  (Confidence Level = 0.99; target proportion =
0.5; width= .06): Sample size = 1844
4. We will use t – this is a random sample and a quick box plot shows that it is fairly symmetric. We have
no reason to think that the values are not independent.
Type the data into StatCrunch. Stat  t-Stats  One Sample  With Data  Click the
appropriate Column  Confidence Interval for μ to find (10.85, 15.15)). A 90%
confidence interval for the mean fill time of all packages is from 10.85
to 15.15 days
Do you remember how to calculate the mean & std dev of a sample of data? Stat  Summary Stats

5. df = n-1 = 35-1 = 34; t* would be 1.6909

Stat  Calculators  T, Between, enter df = 34 and 0.9 to the right of the equal sign

6. z* would be 1.6449. The margin of error will increase if the level of confidence
increases Stat  Calculators  Normal, Between, enter 0.9 to the right of the equal sign

7. margin of error = (upper-lower)/2 = 2.1

8.
a. Assumptions:
i. Random & Independent ? Yes, this is confirmed since we are told it is a random sample and we
could be swayed to believe these are also independent.
ii. Normal? The distribution must be Normal – if it is not “perfect”, a sample of size 36 is large
enough to overcome small departures from Normality
b. df = 35 (calculated as n-1)
c. Stat  t-Stats  One Sample  With Summary  Enter the Mean, S, and n  Confidence Interval for μ
The 95% confidence interval for the average loss is from $9052.473 to $10067.527
 .

9. a. H0 µ = 8.5 and Ha µ ≠ 8.5

10. c. H0 µ = 74,900 and Ha µ > 74,900
11. H0: p = 0.97 and Ha: p > 0.97
12. Stat  Proportion stats  One sample  With
summary
H0 p = 0.20
Ha p < 0.20
α=0.05
z = -0.84
p-value = 0.2000
ask is p-value < α …. (is 0.2000 < 0.05?) NO, so
Decision: Do NOT Reject H0
Conclusion: There is not evidence that the
proportion of purple candies is less than 0.02

13. Stat  t-stats  One sample  With summary

H0 µ = 23.3
Ha µ > 23.3
α=0.05
t = 1.074
p-value = 0.1447
ask is p-value < α …. (is 0.1447<0.05?) NO, so
Decision: Do NOT Reject H0
Conclusion: There is not evidence that the
average age men of first marrying is more than
23.3 years

14. A p-value is the probability of observing a test statistic as far or farther away from the null value given the null
value is true. In context of the problem above, there is a 14.47% chance of getting a sample with a mean of 24.2
or older understanding there is natural variability in sampling. [I said 24.2 or older because our null is only
interested in “older”.]

15. Stat  t-stats  One sample  With summary

H0 p = 0.05
Ha p > 0.05
α=0.05
z = 3.47
p-value = 0.0003
ask is p-value < α …. (is 0.0003<0.05?) yes, so
Decision: Reject H0
Conclusion: There is evidence that the proportion of children with a congenital abnormality is more than 0.05.

16. Take note that the alternate hypothesis would be µ>12. No level of significance was specified, so you can
assume α=0.05. For the decision, ask if the p-value is less than α? Answer, no – so the decision is Do Not
Reject H0. Conclusion: There is not significant evidence that the mean time college students spend doing
homework each week exceeds 12 hours (per week).

17. c. the probability of the observed test statistic given the null hypothesis is true

18. (H0: p = 0.97 vs Ha: p > 0.97). The hypothesis test results in a p-value of 0.122.
Decision (ask is p-value < α …. Is 0.122 < 0.05? … NO) Do NOT Reject H0
Conclusion: There is NOT significant evidence that the proportion exceeds 0.97 (meaning --- there is no evidence the
machine is producing more than 97% good parts).

19. (H0: µ = 96 vs Ha: µ > 96). The resulting p-value is 0.047.

Decision: (ask is p-value < α … is 0.047 < 0.05 … yes) Reject H0
Conclusion: There is significant evidence that the average weight of the pumpkin cans is more than 96 ounces.

20. FALSE – a Type I error occurs if you reject the null hypothesis, but the null was true.
21. FALSE -- When making a decision, you compare the p-value to the level of significance.
22. FALSE -- If p is less than α, your decision is to reject H0 (meaning you think H0 is not true)
23. FALSE -- Conclusions are written in terms of Ha – is there evidence that the alternate is proven or not proven

24. Find the following:

a. What percent of the young girls reported that they drink milk? 1222 / 2149 = 56.9%
b. What percent of the young girls were in the 1989-1991 survey? 837 / 2149 = 38.9%
c. What percent of young girls reported that they drink milk and were in the 1989-1991 survey? 502 / 2149 =
23.4%
d. What percent of the young girls who reported that they drink milk were in the 1989-1991 survey? 502 / 1222 =
41.1%
e. What percent of the young girls in 1989-1991 reported they drink milk? 502/ 837 = 60.0%

25. FALSE -- it is the population parameter (remember that statistics describe samples, parameters describe
populations).