You are on page 1of 14

Name ______________________________

1
AP Stats – Chapter 10-14 Review Introduction to Inference Name ______________________________

Notes for a one-sample confidence interval on µ

  
x  z  
 n

Assumptions
The sample must be reasonably random.
The data must be from a normal distribution or a large sample (n > 30).
 must be known.
The sample must be less than 10% of the population.
This interval is exact when the population distribution is normal and approximately normal for other cases if
the sample size is large (n > 30).

Questions:

1. A survey of 81 theatres showed that the average length of a feature film was 98 minutes. Past studies
indicate that 12 minutes. Determine a 90% confidence interval for estimating the mean length of all
feature films. Interpret this interval in context of the problem.

2. A set of 50 animals is fed a new type of grain to examine weight gain. After four weeks it is found that
the average weight gain in 48 lbs with2.3 lbs. Construct a 98 % confidence interval for the average
weight gain for all animals being fed this grain. Interpret this interval in the context of this problem.

3. The administration of a university desires to estimate the average teaching experience of its 3000
faculty members. How large a sample must be drawn to be 90% confident that the sample mean does
not differ from the population mean by more than ½ year. Past experience indicates that  2.6 years.

Notes for hypothesis test for a mean

x
z

n

Assumptions
The sample must be reasonably random (SRS).
The data must be from a normal distribution or a large sample (n > 30).
must be known.

Questions

1. According to the U.S. Department of Agriculture, the mean yield of corn in the U.S. is about 120
bushels per acre with  = 10. A survey of 50 farmers this year gives a sample yield of 123.6 bushels per
acre. We want to know whether this is good enough evidence that the national average is not 120
bushels per acre. Let   .01 .

2
2. The average score of a sixth grader on the Binet Math Aptitude Test is 75 with 8.1. The mean of a
random sample of 50 students was 71. Does this indicate that the students in this school are less skilled in
their math abilities? Use a level of significance of 5%.

3. For many semesters, an instructor has recorded his students’ grades and has found that the mean
grade for all of these classes is 72 with 12. The current class of 37 students seems to be better than
average in their ability and he want to know if this class is superior to his previous classes. Does a current
class average 75.2 present sufficient evidence of this? Use a 5% level of significance.

Notes for a one-sample confidence interval on µ

 s 
x  t   with df = n – 1
 n

Assumptions
In theory, using the t-distribution requires that the data be from a normal distribution. In practice, the t-
distribution is robust to violations of this assumption provided that there are on outliers in the data.
The data must be reasonably random.
Make a graphical check of shape and outliers if n < 15.

Questions

1. A company has a new process for manufacturing large artificial sapphires. The production of each
gem is expensive, so the number available for examination is limited. In a trial run, 12 sapphires are
produced. The mean weight is 6.75 carats and the standard deviation is 0.33 carat. Find a 99%
confidence interval for the mean population weight for all sapphires produced by this new process.
What assumptions did you make about the distribution of sapphires in order to obtain this answer?

2. A study of the ability of individuals to walk a straight line reported the accompanying data on
cadence (strides per second) for a sample of 20 randomly selected healthy men. Based on the data,
construct a 99% confidence interval to estimate the mean cadence of all males.

.95 .85 .92 .95 .93 .86 1.00 .92 .85 .81
.78 .93 .93 1.05 .93 1.06 .96 .81 .96 1.06

3. Suppose an archeologist discovers 7 fossil skeletons from a previously unknown species of miniature
horse. Reconstructions of the skeletons allow her to measure the shoulder height (in centimeters) of the
horses. The heights are 45.3, 47.1, 44.2, 46.8, 46.5, 45.5, and 47.6. From this data, construct a 98%
confidence interval for the mean shoulder length of this species of miniature horse.

3
Notes for a one-sample t-test on µ

x
t with n  1 degrees of freedom
s
n

Assumptions:
In theory, using the t-distribution requires that the data be from a normal distribution. In practice, the t-
distribution is robust to violations of this assumption provided that there are on outliers in the data.
The data must be reasonably random.
Make a graphical check of shape and outliers if n < 15.

Questions

1. The manufacturer of a certain foreign car claims that it will average 36 mpg. To test the claim, a
sample of 10 cars is taken and driven under normal conditions. These cars average 31 mpg with s = 8.3
mpg. Should we reject this claim at   .05 ? When   .01 ?

2. A new formula for the propellant of missiles is being tested to determine if it is superior to the current
formula. From past experience, the mean distance traveled has been 340 miles. Twelve missiles with the
new propellant are fired into the Pacific Ocean and their distances measured. The data is given below.
Support your conclusion with appropriate statistical evidence.

375 376 304 345 368 329


372 347 342 391 400 325

Notes for a one-sample confidence interval on p

p  z*

p 1 p 
n

Assumptions
The sample must be reasonably random.
The population must be ten times the sample size.
The sample must be large enough so that 
n p  10 and n 1  p  10 . 
Questions

1. In a random sample of 100 people, 40 of them stated they were in favor of the school district bond
election. Find a 95% confidence interval for the true proportion of all district residents who favor the bond
election. Interpret this interval in the context of the problem.

2. A union member reported that 80 out of 125 workers interviewed supported some sort of work
stoppage to further their demands for better safety conditions. Find a 99% confidence interval estimate of
the true proportion of workers who support such action.

4
Notes for a hypothesis test for a proportion

p p
z
p 1  p 
n

Assumptions
The sample must be reasonably random.
The population must be ten times the sample size.
The sample must be large enough so that  
n p  10 and n 1  p  10 .
Questions

1. The director of advertising sales for a local radio station claims that 35% of the listeners will remember
the name of the product advertised at his station. A sample of 100 listeners is taken and 43 recall the
product name. Is the claim accurate at   .05 ?

2. A large university provides enough housing for 10 percent of its graduate students to live on campus.
The university’s housing office thinks that the percent of graduate students looking for housing on campus
may be more than 10 percent. The housing office decides to survey a random sample of graduate
students, and 62 of the 481 respondents say they are looking for housing on campus.

a) On the basis of the survey, would you recommend that the housing office consider increasing the
amount of housing on campus available to graduate students? Give the appropriate statistical
evidence to support your recommendation.

b) In addition to the 481 graduate students who responded to the survey, there were 19 who did not
respond. If those 19 had responded, is it possible that your recommendation would have changed?
Explain.

5
AP Stats – Chapter 10-14 Review Introduction to Inference #2 Name ______________________________

Notes for matched pairs t-procedures

x d  d
Test Statistic: t df  n  1
sd
n

sd
Confidence Interval: xd  t * df  n  1
n

To compare the responses to two treatments in a matched pairs design, apply the one-sample t
procedures to the observed differences.

Assumptions:
The sample of paired differences must be reasonably random.
The paired differences will be reasonably normally distributed. This condition is robust if there are no
outliers in the differences.

Questions:

1. In a study of memory recall, eight students from a large psychology class were selected at random
and given 19 minutes to memorize a list of 20 nonsense words. Each was asked to list as many words as he
or she
could remember both 1 hour and 24 hours later, as shown in the accompanying table.

Subject 1 2 3 4 5 6 7 8
1 hr later 14 12 18 7 11 9 16 15
24 hrs later 10 4 14 6 9 6 12 12

Is there evidence to suggest that the mean number of words recalled after 1 hour exceeds the mean
recalled after 24 hours? Use a level .01 test.

2. Do girls think they don’t need to take as many science classes as boys? The article “Intentions of
Young Students to Enroll in Science Classes in the Future: An Examination of Gender Differences” (Science
Education (1999): 55-76) gives information from a survey of 224 children in grades 4, 5, and 6. The 224 girls
participating in the survey each indicated the number of science courses they intended to take in the
future, and they indicated the number of science courses they thought boys their age should take in the
future. For each girl, the authors calculated the difference between the number of science classes she
intends to take and the number she thinks boys should take.

a) Explain why this data is paired.


b) The mean difference was -.83 and the standard deviation was 1.51. Construct and interpret a 95%
confidence interval for the mean difference.

6
Notes for Two-sample Confidence Interval and t-test on 1  2

x  x      
x  x t* s12 s22 1 2 1 2
1 2  t df  n1  1, or n2  1 (which ever is smaller)
n1 n2 s12 s22

n1 n2
Assumptions:

Samples are independent.


Samples must be reasonably random.
Both populations have to be normally distributed and the means and standard deviations are unknown.

Questions

1. The local chapter of NOW claims that female college graduates earn less than male college
graduates as determined by starting salaries. A survey of 45 men and 40 women indicated that the
average starting male salary was $26,800 and that females started with an average salary of $26,000. For
males, s = $600 and for females, s = $900. Do these figures support the claim that women earn less than
men? Use a 1% level of significance.

2. Based on information from Consumer Reports (October 1993), a random sample of 22 thirty-gram
servings of Duncan Hines Chocolate Chip Cookies had a mean of 132 calories with s = 27. A random
sample of 53 Pepperidge Farm Chocolate Chip Cookies (30-gram servings) had a mean of 124 calories
with s = 33. Does this information indicate that there is a difference in the average number of calories for
these two cookie brands? Use a 5% level of significance.

3. Are girls less inclined to enroll in science courses than boys? In the previously referenced study,
randomly selected fourth, fifth, and sixth graders were asked how many science courses they intended to
take. The following data was obtained:

n Mean Std. Deviation


Males 203 3.42 1.49
Females 234 2.42 1.35

Calculate a 95% confidence interval for the difference between males and females in mean number of
science courses planned. Interpret this interval. Based on your interval, how would you answer the
question posed at the beginning of the problem?

7
Notes for Two-sample Confidence Interval and z-test on p1  p2


p1 1  p1   p 1  p 
 p  p  z*
1 2
n1
2

n2
2

z
 p  p  p  p 
1 2 1 2
p  pooled p1 and p2
p 1  p    
1 1
n n 
1 2

Assumptions

The two samples must be independent and reasonably random.


The sample size must be large enough so that n1 p1 , n1 1  p1  , n2 p2 , n2 1  p2  are all five or more.
The population must be ten times the sample size.

Questions

1. Researchers at the National Cancer Institute released the results of a study that examined the effect
of weed-killing herbicide on house pets. Dogs, some of whom were from home where the herbicide was
used on a regular basis, were examined for the presence of malignant lymphoma. These are the results:

Group Sample Size Number with Lymphoma


Exposed 827 473
Unexposed 130 19

We wish to give a 95% confidence interval for the difference between the proportion of unexposed dogs
and the proportion of exposed dogs that developed lymphoma. Interpret this interval in the context of
the problem.

2. An insurance company wants to offer discounts on its life insurance policies to nonsmokers. As part of
its analysis, the company randomly selects 200 men and women who are 50 years old and asks them if
they smoke and if they have ever suffered from heart disease. The results indicate that 20 out of 80
smokers and 15 out of 120 nonsmokers suffer from heart disease. Can the company conclude that
smokers have a higher incidence of heart disease than non-smokers?

3. Two competing drugs are available for treating a specific ailment. There are no apparent side effects
from drug #1, whereas there are (headache and nausea) from drug #2. A group of researchers have
decided, however, that they are willing to recommend drug #2 if the proportion of cures are higher than
that of drug #1. The researchers use the drug experimentally on two groups of people who are suffering
from the ailment. By the end of the experiment, 52 out of 80 treated with drug #1 are cured, while 63 out
of 90 given drug #2 were cured. At the 1% level of significance, should the second drug be
recommended?

8
AP Stats – Chapter 10 -14 Review Introduction to Inference #3 Name ______________________________

Notes for Chi-Square Goodness of Fit

Ho : the data fits the hypothesized distribution.


Ha : the true data has a different distribution than the one hypothesized.

(observed count - expected count)2


2   df  n  1
expected count

Assumptions
The sample should be reasonably random.
All expected counts must be greater than or equal to 1.
No more than 20% of the expected counts are less than 5.

Questions

1. College students have regularly insisted on freedom of choice when registering for courses. This
semester there are seven sections of Statistics. They are scheduled to meet at various times with a variety
of instructors. The table below shows the number of students who selected each section. Do the data
indicate that the students had a preference for certain sections?

Sections 1 2 3 4 5 6 7
# of Students 18 12 25 23 8 19 14

2. The Mendelian theory of inheritance claims that the frequencies of round and yellow, wrinkled and
yellow, round and green, and wrinkled and green peas will occur in the ratio of 9:3:3:1 when two specific
varieties of peas are crossed. In testing this theory, Mendel observed the frequencies of 315, 101, 108, and
32 respectively. Do these sample data provide sufficient evidence to reject this theory at a .05 level of
significance?

Notes for Chi-Square Homogeneity of Populations

Ho : the true category proportions are the same for all populations (homogeneity).
Ha : the true category proportions are not the same for all populations.

(observed cell count - expected cell count)2


2  
expected cell count

df   number of rows - 1 number of columns - 1

Assumptions and Conditions


The data must be categorical and bivariate.
The sample must be reasonably random.
All expected counts must be greater than or equal to 1.
No more than 20% of the expected counts are less than 5.

9
Questions

1. Fear of darkness is a common emotion. The following data were obtained by asking 200 people in
each age group whether they had serious fears of darkness.

Age group Elementary Jr. High Sr. High College Adult

# with fear 83 72 49 36 114


# without
fear

At   .01 , do we have sufficient evidence to reject the hypothesis that “the same proportion of each
age group has serious fears of darkness”?

2. A nationwide market research study was undertaken to determine the preference of various age
groups of males for different sports. A random sample of 1000 men was selected, and each individual was
asked to indicate their favorite sport. The results are as follows:

Age Group Baseball Football Basketball Hockey


Under 20 26 47 41 36
20 – 29 38 84 80 48
30 – 39 72 68 38 22
40 – 49 96 48 30 26
50 and over 133 44 18 5

a) At the .01 level of significance, is there evidence that the preferences for age groups are different?
b) Compute the p-value and interpret its meaning in context of the  -value.

10
Notes for Chi-Square Test of Independence

Ho : the two variables are independent.


Ha : the two variables are not independent.

The assumptions and conditions are the same as the test for homogeneity.

Questions

1. In Dallas, residents can watch national news on ABC, NBC, CBS and Fox. A researcher wishes to know
whether there is any relationship between political philosophy and preferred news program. A random
sample of 300 regular viewers is selected and each is asked their political preference and news network
viewed. The results are given below. Are preferred network and political philosophy independent?

ABC CBS NBC FOX Row Total


Liberal 20 20 25 15
Moderate 45 35 50 20
Conservative 15 40 10 5
Column Total

2. Dr. Denton Fender is doing a study to see if the type of driving a person prefers to do and the gender
of the person is related. A random sample of 100 drivers gave the data below. Test whether driving
preference and gender are independent. Use   .01 .

Multi-Lane with Two-Lane with


Gender Freeway Row Total
Lights Traffic Lights
Female 18 16 11
Male 22 19 14
Column Total

11
Name _____________________________ Period _________________

Chapter 10-14 Test Review

Directions: Read the following problems. Decide which test should be used to answer the problem. Write
the Hypotheses. Then list and verify the assumptions necessary for inference. Do not actually
perform the test of significance or do a complete write-up.

1. Elite Foods, a supermarket, is losing customers and suspects it’s because another neighborhood
market has lower prices. To determine whether this is the case, researchers from Elite Foods take a
random sample of ten items from their store and compare their prices to prices for the same items
at the other store. Is there evidence that Elite has higher prices on the same items as the
competition?

Item 1 2 3 4 5 6 7 8 9 10
Elite Foods 1.65 2.19 1.99 3.49 .99 1.59 2.89 4.50 1.19 1.99
Other store 1.49 2.00 2.09 2.99 .99 1.79 2.39 4.25 .99 1.79

2. The Associated Press found that 730 of 1000 randomly selected adults preferred to watch movies at
home rather than at a movie theater. Is there convincing evidence that the majority of adult
Americans prefer watching movies at home?

3. It is difficult to accurately determine a person’s body fat percentage without immersing him or her in
water. Researchers hoping to find ways to make a good estimate immersed 20 male subjects, and
then measured their weights. (Bock, Velleman, DeVeaux, p. 160)

Weight
175 181 200 159 196 192 205 173 187 188 188 240 175 168 246 160 215 159 146 219
(lbs.)

Body
Fat 6 21 15 6 22 31 32 21 25 30 10 20 22 9 38 10 27 12 10 28
(%)

Is there sufficient evidence that weight can be used to predict body fat?

4. In November 1997, the Gallup Organization released the results of a poll on family values. One of the
questions asked was: “For you personally, do you think it is necessary or not necessary to
have a child at some point in your life in order to feel fulfilled?” Results from adults in
different countries appear below. Is there evidence that the proportion of adults who
would give each answer is different for each country?

U.S. India Mexico Canada Germany Total


Yes 460 930 610 590 490 3080
No 510 60 380 370 450 1770
Undecided 30 10 10 40 60 150
Total 1000 1000 1000 1000 1000 5000

12
5. Fumonisins are environmental toxins produced by a type of mold; they are found in corn and in
products made from raw corn. Based on a study of corn meal, the Center for Food and Safety and
Applied Nutrition provided recommendations on allowable fumonisin levels in human food and in animal
feed. The study compared corn meal made from partially degermed corn (corn that has had the germ
partially removed) with corn meal made from corn that has not been degermed. Specimens of corn
meal were analyzed, and the total fumonisin level (in ppm) was determined for each specimen.
Summary statistics for total fumonisin level from the U.S. Food and Drug Administration’s web site
are given:
n x s
Partially degermed 50 .59 1.01
Not degermed 50 1.21 1.71
Is there evidence to suggest that the mean fumonisin levels of the two types of corn meal are
different?

6. The Advanced Placement Statistics examination was first administered in May 1997. Students’
papers are graded on a scale of 1 to 5, with 5 being the highest score. Over 7600 students took the
exam the first year, with the following distribution of scores.

Score: 5 4 3 2 1
Percent 15.3 22.0 24.8 19.8 18.1
A sample of students who took the exam had the following distribution of grades:
Score: 5 4 3 2 1
Number of students: 167 158 101 79 30
Is the distribution of scores for this sample significantly different from the distribution of scores
for all students who took the inaugural exam?
7. A Gallup Poll taken in May 2000 asked the question: “In general, do you feel that the laws covering
the sale of firearms should be made: more strict, less strict, or kept as they are now?” Of the n =
493 men who responded, 52% said “more strict,” while of the n = 538 women who responded, 72% said
“more strict.” Assuming these respondents constitute random samples of U.S. men and women, is
there sufficient evidence to conclude that a higher proportion of women than men in the population
think these laws should be made more strict?

8. The Statistical Abstract of the United States (109th edition) reported that the average cost per day
of owning an automobile in the United States is $7.62 with standard deviation of $1.77. This includes
the cost of the car, general maintenance, gasoline, and insurance. A random sample of 54 students
who own cars found the average cost per day to be $6.78. Is the college student’s average daily
expense less than the national average?

9. The times of first sprinkler activation (in seconds) for a series of tests of fire-prevention sprinkler
systems that use aqueous film-forming foam is as follows. The system has been designed so that the
true average activation time is supposed to be at most 25 seconds. Does the data indicate the design
specifications have not been met?

27 41 22 27 23 35 33 24 28 22 24 30
13
10. “Buddies” is a volunteer social service that works with disadvantaged children. Adults, 18 years &
older, volunteer from 1 to 6 hours per week to take a disadvantaged child to the zoo, to a museum, to
a movie, or to some other activity. The program recruits adults from three main groups: college
students, non-students living in the inner city, and non-students living in the suburbs. A random
sample of adult volunteers gave the following information. Is there an association between the
number of hours volunteered and the type of volunteer?

Hours Volunteered

1-2 3-4 5-6


College student 115 93 47
Inner-city resident 88 150 56
Suburban resident 95 133 60

*Other items to consider as you prepare for the Chapter 10-14 Test*
1. Confidence Intervals: What they are for, what “confidence” means, how they are constructed.
2. The Empirical Rule.
3. Desired Sample Size for either means or proportions.

14

You might also like