Graded Assignment MTH510A: AP Statistics Unit 4 4.3 Test

Graded Assignment MTH510A: AP Statistics | Unit 4 | 4.3 Test

Graded Assignment

Test: Planning a Study

Directions

· Mark your answers to the multiple-choice questions on the answer sheet at the end of the multiple-choice section. Use a black or blue pen.

· Remember to complete the submission information on every page you turn in.

Section I (Multiple-Choice): 12 questions, 48 points, 20 minutes

Score

(4 points)

1. A placebo is:

A. a control group.

B. the true cause of a response.

C. an important part of a double-blind experiment.

D. a non-treatment treatment given to a control group.

E. a pure random sample.

Score

(4 points)

2. An important reason for using randomization in assigning subjects to treatment groups in an experiment is:

A. to produce the placebo effect.

B. to eliminate bias among groups.

C. to equalize blocks in a block design.

D. to create a double-blind experiment.

E. to be sure treatment groups are normally distributed.

Score

(4 points)

3. Data were collected in twenty major urban areas on the percent of women in the labor force. Data were collected in 1968 and again in 1972. Gains or losses in this percent were the basis for discussing whether or not there had been an increase in percent of women in the labor force over the period. What kind of design was this?

I. A matched-pairs design

II. A randomized block design

III. An observational study

A. I only

B. II only

C. III only

D. I and II only

E. I and III only

Score

(4 points)

4. Your school wants a simple random sample of 50 students from the junior class at the school. Which of the following will produce the simple random sample?

A. Give each student in the junior class a number and use a random number generator to select 50 students.

B. Choose two United States history classes (assuming that all juniors take U.S. History).

C. Choose 50 students at random from those at lunch in the school cafeteria.

D. Select the first 50 juniors to arrive at school on one given day.

E. Randomly select 5 students from each of the junior English classes (assume all juniors take junior English).

(4 points)

5. A stratified random sample for a survey corresponds to which of the following experimental

Score

designs?

A. A double-blind experiment

B. An experiment with a placebo

C. A block design

D. An observational study

E. A matched-pairs design

(4 points)

6. Which of the following are important in the design of experiments?

Score

I. Control of confounding and lurking variables

II. Randomization in assigning subjects to different treatments

III. Replication of the experiment using sufficient number of subjects

A. I and II only

B. I only

C. I and III only

D. II and III only

E. I, II, and III

(4 points)

7. The control group in an experiment should:

Score

A. receive the opposite of the treatment given to the treatment group.

B. have the same experiences as the treatment group, except for receiving the treatment variable.

C. be carefully matched to the treatment group on a large number of characteristics.

D. receive none of the same experiences as the treatment group.

E. receive the treatment at random.

(4 points)

8. A good reason for a market researcher to use a stratified random sample rather than a simple

Score

random sample is:

A. convenient data collection.

B. to eliminate bias.

C. to collect data systematically.

D. to collect data at lower cost.

E. to reduce unwanted variability.

(4 points)

9. You've developed a protein drink that you think cures arthritis. To test it, you find 200 volunteers (100 men and 100 women) and test them to measure the amount of discomfort they feel from their arthritis. Then you give the treatment to the subjects, test them all at regular intervals, and measure their improvement. You find that most subjects show marked improvement. The people performing the tests are not told that they're studying an arthritis medication. You excitedly report your findings in an infomercial that is seen by

millions of people, and sales skyrocket. But were the test results truly useful for determining the effectiveness of the treatment? Why or why not?

Score

A. Yes, it's a valid before-and-after design.

B. Yes, it's a blind study.

C. No, there was no control.

D. No, there was no replication.

E. Can't be determined without knowing how much improvement was measured.

(4 points)

10. In a famous study from the late 1920s, the Western Electric Company wanted to study the effect of lighting on productivity in a factory. The factory workers

were aware that the study was in progress. The researchers discovered that worker productivity increased with each change of lighting, whether the lighting increased or decreased. What's the most likely reason this phenomenon occurred?

Score

A. Lack of randomization

B. Absence of a control group

C. The subjects knew they were being studied

D. Sampling variability overpowered other kinds of variability

E. Voluntary response bias and undercoverage

(4 points)

11. In 1936 the magazine Literary Digest, incorrectly predicted that Alf Landon would defeat

Score

Franklin Roosevelt in the presidential election. They based their prediction on 2.4 million responses out of 10 million sample ballots they sent to their own subscribers, individuals in telephone directories, and lists of automobile owners. What type of bias might have caused this error?

I. Nonresponse bias

II. Voluntary response bias

III. Undercoverage

A. I only

B. II only

C. I and II only

D. I, II, and III

E. II and III

(4 points)

12. You've developed a new anxiety reduction treatment. Your study design is to separate men and women into two groups, randomly divide each group in half, and then give one half the new medication and one half a placebo. The basic design of this study is:

Score

A. comparative randomized, blocked by gender.

B. completely randomized.

C. randomized, blocked by gender and type of medication.

D. randomized and stratified.

E. None of the above

This is the end of the multiple-choice question section of the Quiz. Now continue with the free-response portion.

Copying or distributing without K12’s written consent is prohibited.

Graded Assignment MTH510A: AP Statistics | Unit 4 | 4.3 Test

User Name: Instructor: Date:

(print clearly)

Answer Sheet

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Your Score / ___ of 48

Copying or distributing without K12’s written consent is prohibited.

Graded Assignment MTH510A: AP Statistics | Unit 4 | 4.3 Test

User Name: Instructor: Date:

(print clearly)

Directions

· Neatly write your responses in the spaces provided. Use a blue or black pen. Don't write in the margins.

· Remember to complete the submission information on every page you turn in.

Section II (Free Response): 3 questions, 52 points, 30 minutes

(15 points)

Score

1. A survey of physicians in 1979 found that some doctors give a placebo to patients who complain of pain for which the physician can find no cause. If the patients' pain improves, these doctors conclude that it had no physical basis. The medical researchers who conducted the survey claimed that these doctors do not understand the placebo effect. Why?

Copying or distributing without K12’s written consent is prohibited.

Graded Assignment MTH510A: AP Statistics | Unit 4 | 4.3 Test

User Name: Instructor: Date:

(print clearly)

(15 points)

2. Advertisements claim that taking an SAT preparation course can significantly improve performance on the SAT. Assume there are three SAT preparation services available for students to enroll in, and assume you have 200 students available to use in your experiment. Explain the design of an experiment to examine the claim that SAT preparation courses improve SAT scores and, if so, which of the preparation services would seem to be the best.

Score

Copying or distributing without K12’s written consent is prohibited.

Graded Assignment MTH510A: AP Statistics | Unit 4 | 4.3 Test

User Name: Instructor: Date:

(print clearly)

(22 points)

Score

3. You know that a certain trait occurs in 30% of the population. You want to know how many individuals you have to draw from the population, on average, to get three subjects with the trait.

A. Describe how you would use a table of random numbers to do a simulation of this situation. (11 points)

B. Use the table of random numbers below to carry out five repetitions in the simulation method you described in (A) above. Mark the table so that it's clear to the reader how you did the simulation. Begin with the first element of the first row. If it's necessary to continue on past the first row, start at the beginning of each row. Based on your five simulations, what's your best guess as to the number of trials required to achieve three successes? (11 points)

26411 / 94292 / 06340 / 97762 / 37033 / 85968 / 94165 / 46514
80011 / 09937 / 57195 / 33906 / 94831 / 10056 / 42211 / 65491
92813 / 87503 / 63494 / 71379 / 76550 / 45984 / 05481 / 50830
Score

User Name: Instructor: Date:

(print clearly)

Formulas

You may refer to these during the Quiz.

Formulae for Basic Descriptive Statistics