In order to improve reliability:
A. Repeat your measurements several times
B. Take the average of several measurements
C. Use a better instrument
D. Both taking the average of several measurements and using a
better instrument E. Both repeating your measurements several times
and using a better instrument
12. The measured value considers:
A. True value
B. Bias
C. Random error
D. All of the choices are correct.
13. Measuring a healthy lifestyle. You want to measure the
“healthiness” of college students’ lifestyles. Give an example of a
clearly invalid way to measure healthiness. Then briefly describe a
measurement process that you think is valid.
14. Rates versus counts. Thirty students in my Stat 1350 class
last semester took Test 1 and 25 of them passed the test.
Fifty-five students in my Stat 1350 WEB class last semester took
Test 1 and 43 of them passed the test.
(a) More students in my WEB Stat 1350 class passed Test 1 than
in my traditional Stat 1350 class. Why does this NOT show that my
WEB students did better than my traditional students?
(b) What is the passing rate (percentage of students who
passed) for each of my Stat 1350 classes?
WEB: __________________________________ Traditional:
______________________________
15. Obesity. An article in the June 30, 2010, Columbus
Dispatch reported on the prevalence of obesity among adults in the
50 states. Based on information in the article, California has
approximately 6.7 million obese adults, and Texas has approximately
5.2 million. On the other hand, Mississippi has a little over
730,000 obese adults. Do these numbers make a convincing case that
California and Texas have a more substantial problem with obesity
than Mississippi?
16. Measuring intelligence. “Intelligence” means something
like “general problem-solving ability.” Explain why it is not valid
to measure intelligence by a test that asks questions such as
Who wrote “The Star-Spangled Banner”?
______________________________________________________ Who won the
last soccer World Cup?
__________________________________________________________
17. Testing job applicants. The law requires that tests given
to job applicants must be shown to be directly job related. The
Department of Labor believes that an employment test called the
General Aptitude Test Battery (GATB) is valid for a broad range of
jobs. As in the case of the SAT, blacks and Hispanics get lower
average scores on the GATB than do whites. Describe briefly what
must be done to establish that the GATB has predictive validity as
a measure of future performance on the job.
18. Validity, bias, reliability. This winter I went to a local
pharmacy to have my blood pressure measured using a sophisticated
electronic machine at the front of the store next to the checkout
counter. Will the measurement of my blood pressure be biased?
Reliable? Valid? Explain your answer.
19. More on bias and reliability. You cut 5 pieces of string
having these lengths in inches: 2.9 9.5 5.7 4.2 7.6
A subject measures each length by eye. Make up a set of
results from this activity that matches each of the descriptions
below. For simplicity, assume that bias means the same fixed error
every time rather than an “on the average” error in many
measurements.
(a) The subject has a bias of 0.5 inch too long and is
perfectly reliable. ___________ ___________ ___________ ___________
___________
(b) The subject has no bias but is not perfectly reliable, so
that the average difference in repeated measurements is 0.5
inch.
___________ ___________ ___________ ___________
___________
(c) A subject measures the first length (true length = 2.9
inches) four times by eye. His measurements are 3.0 2.9 3.1
3.0
What are the four random errors for his measurements?
___________ ___________ ___________ ___________
20. The best earphones. You are writing an article for a
consumer magazine based on a survey of the magazine’s readers that
asked about satisfaction with mid-priced earphones for the iPod and
iPhone. Of 1648 readers who reported owning the Apple in-ear
headphone with remote and mic, 347 gave it an outstanding rating.
Only 69 outstanding ratings were given by the 134 readers who owned
Klipsch Image S4i earphones with microphone. Describe an
appropriate variable, which can be computed from these counts, to
measure high satisfaction with a make of earphone. Compute the
values of this variable for the Apple and Klipsch earphones. Which
brand has the better high-satisfaction rating?
21. Measuring pulse rate. You want to measure your resting
pulse rate. You might count the number of beats in 5 seconds and
multiply by 12 to get beats per minute. Why is this method less
reliable than actually measuring the number of beats in a
minute?
22. Measuring crime. Twice each year, the National Crime
Victimization Survey asks a random sample of about 40,000
households whether they have been victims of crime and, if so, the
details. In all, nearly 135,000 people answer these questions per
year. If other people in a household are in the room while one
person is answering questions, the measurement of, for example,
rape and other sexual assaults could be seriously biased. Why?
Would the presence of other people lead to over-reporting or
underreporting of sexual assaults?
Lesson 3 – Making Sense of Numbers
23. A survey in a local newspaper stated that of the
individuals who frequent bookstores, 14% were male and 28% were
female. What is wrong with this picture?
A. Only 42% of the people were surveyed.
B. The percentages only add up to 42%. It should be 100%. C.
There were twice as many females surveyed as males. D. 58% do not
go to bookstores.
E. Nothing is wrong with this picture.
24. A newspaper reported "Approximately 17% of all crime takes
place in the months of August and September." What is misleading
about this statistic?
A. August and September are low points for crime
statistically. B. It does not discuss the other months.
C. August and September make up 1/6 of the year which is 17%.
D. None of the choices are correct.
25. Mark has two coupons—one for 10% off and one for $5 off.
The store is allowing him to use both. He says to the cashier to
apply the coupons in any order she wants because in the end it's
the same amount off. Is he correct? (Hint: Try doing this for an
item that is $100.)
A. Yes, it doesn't matter the way the coupons are
applied.
B. No, you should apply the 10% off coupon first then apply
the $5 off coupon. C. No, you should apply the $5 off coupon then
apply the 10% off coupon.
D. You cannot determine from the information given.
26. When finding the percent change, your denominator should
be:
A. Amount of change B. Starting value
C. The smaller value D. The larger value.
27. If an amount increases from 10 to 40 then the percent
increase is:
A. 300% B. 400% C. 40% D. 10%
28. A newspaper reports "The percent decrease in the amount of
wolves is 150%." What does this mean?
A. The amount of wolves has decreased in half.
B. There is 1/3 the amount of wolves that there has been
previously. C. This is not possible. "Percent decrease" can't be
more than 100%. D. None of the choices are correct.
29. What percentage of 30 is 40?
A. 133% B. 75% C. 13.3% D. 7.5%
30. The percent increase from 40 to 70 is:
A. 125% B. 75% C. 175% D. 25%
31. In determining if the numbers make sense you should:
A. Look at the context of the numbers and determine if there
is missing information. B. Look for numbers that don't agree as
they should.
C. Compare numbers and look for numbers that are surprisingly
large or small.
D. All of the choices are correct.
E. None of the choices are correct.
32. Deer in the suburbs. Westchester County is a suburban area
covering 433 square miles immediately north of New York City. A
garden magazine claimed that the county is home to 800,000 deer. Do
a calculation that shows this claim to be implausible.
33. Trash at sea? A report on the problem of vacation cruise
ships polluting the sea by dumping garbage overboard said:
On a seven-day cruise, a medium-size ship (about 1,000
passengers and 1.000 crew members) might accumulate 222,000 coffee
cups, 72,000 soda cans, 40,000 beer cans and bottles, and 11,000
wine bottles.
Are these numbers plausible? Do some arithmetic to back up
your conclusion. Suppose, for example, that the crew is as large as
the passenger list. How many cups of coffee must each person drink
every day?
34. Airport delays. An article in a midwestern newspaper about
flight delays at major airports said: According to a Gannett News
Service study of U.S. airlines’ performance during the past five
months,
Chicago’s O’Hare Field scheduled 114,370 flights. Nearly 10
percent, 1,136, were canceled.
Check the newspaper’s arithmetic. What percent of scheduled
flights from O’Hare were actually canceled?
35. Battered women? A letter to the editor of the New York
Times complained about a Times editorial that said “an American
woman is beaten by her husband or boyfriend every 15 seconds.” The
writer of the letter claimed that “at that rate, 21 million women
would be beaten by their husbands or boyfriends every year. That is
simply not the case.” He cited the National Crime Victimization
Survey, which estimated 56,000 cases of violence against women by
their husbands and 198,000 by boyfriends or former boyfriends. The
survey showed 2.2 million assaults against women in all, most by
strangers or someone the woman knew who was not her past or present
husband or boyfriend.
(a) First do the arithmetic. Every 15 seconds is 4 per minute.
At that rate, how many beatings would take place in an hour? In a
day? In a year? Is the letter writer’s arithmetic correct?
(b) Is the letter writer correct to claim that the Times
overstated the number of cases of domestic violence against
women?
36. Stocks go down. On September 29, 2008, the Dow Jones
Industrial Average dropped 778 points from its opening level of
11,143. This was the biggest one-day decline ever. By what
percentage did the Dow drop that day? On October 28, 1929, the Dow
Jones Industrial Average dropped 38 points from its opening level
of 299. By what percentage did the Dow drop that day? This was the
second-biggest one-day percentage drop ever.
37. Too good to be true? The late English psychologist Cyril
Burt was known for his studies of the IQ scores of identical twins
who were raised apart. The high correlation between the IQs of
separated twins in Burt’s studies pointed to heredity as a major
factor in IQ. (“Correlation” measures how closely two variables are
connected. We will meet correlation in Chapter 14.) Burt wrote
several accounts of his work, adding more pairs of twins over time.
Here are his reported correlations as he published them:
What is suspicious here?