NAME: ____________________________ Unit 3 Review

1) What are the 3 principles of experimental design?

Control, Randomization, Replication

2) What does it mean when two variables are confounded?

We cannot separate their effects on the response variable. We cannot determine what effect each variable has on the
response.

3) What is the difference between subjects and individuals (or experimental units)?
Subject = people experimental units = anything else

4) What is the placebo effect?

When an experimental unit reacts to a placebo. Can be + or – reaction. The experimental unit shows a response to the
placebo.

5) What is the difference between a placebo and a control?

Placebo is a dummy treatment that looks/tastes/smells/etc. the same as the real treatment.
Control is a group that we can compare our treatment too. Can be no treatment or an old treatment.

6) What is a lurking variable?

A variable that is not taken into account in our experiment, but that has an effect on the response variable.

7) We have 21 people that we need to assign to 3 different treatments (trt1, trt2, trt3). Use the section of the table of random
digits below and assign the 21 people to the 3 treatment groups. List the numbers that are selected under each of the 3
headings below. Clearly explain your procedure!!

Procedure:
 Assign each person a number #01 – 21.
 Read across the TRD below in sets of 2 digits
 Ignore repeats and #00, 22 – 99
 First 7 numbers that come up in our range are assigned to the Trt. 1, next 7 numbers that come up are assigned to Trt. 2,
and whomever is left is assigned to Trt. 3.

TABLE OF RANDOM DIGITS:

08768 11202 34859 09217 18194 45621 05078 66813 65461 50416 99742 08657

Sample:
Trt 1: 08, 12, 02, 17, 18, 19, 21

Trt. 2: 05, 07, 13, 15, 04, 16, 20

Trt. 3: 01, 03, 06, 09, 10, 11, 14

8) An investigator wants to study the effects of two different fertilizers on plant growth (call them A and B). There are 20 plots
available to test the fertilizers on. The investigator will measure the amount of growth by the plants after 3 months.
a. What is the explanatory variable? Fertilizer

b. What are the treatments? Fertilizer A and B

c. What is the response variable? Amount of growth after 3 months

d. What are the individuals/subjects? Individuals/experimental units = 20 plots of plants

e. Do you think you should use a placebo group? How about a control group?
No need for placebo since they are plants.
Yes, control group. We want to see how the plants grow naturally, and have something to compare to.

f. What are some lurking variables when it comes to plant growth?

Sunlight, water, etc.

g. Using one of the lurking variables you listed above, design a block design experiment:

9) An investigator wants to study the effectiveness of two surgical procedures to correct nearsightedness. Procedure A uses
cuts from a scalpel and procedure B uses a laser. The data to be collected are the degrees of improvement in vision after
the procedure is performed. There are 80 nearsighted people available for the experiment.
a. What is the explanatory variable? The procedure used

b. What are the treatments? Procedure A and Procedure B

c. What is the response variable? Degree of improvement in the vision

d. What are the individuals/subjects? Subjects = 80 registered voters

e. Do you think you should use a placebo group? How about a control group?
No placebo or control group necessary. Neither would work. A placebo/control is not possible- the patients
would know that they were getting a dummy treatment or no treatment.
 f. Design a randomized comparative experiment:

g. There are two treatments. Why is it NOT ok to do a matched pairs design for this experiment?
- If you were to do a matched pairs as “1st/2nd” there would be no way to “reset the patients to their
original vision after the first treatment so that you can then test the next procedure.
- If you were to do the matched pairs as “left/right” you would run the risk of one procedure not working
well on the patient while the other one does, and then the patient could suffer because they had one
eye that was good and the other is still nearsighted.

10) A researcher wants to see if more expensive mattresses really give a better night’s sleep than the discount mattress
brand. So they recruit 110 adult volunteers to participate in their study. They will have the adults sleep on the mattresses
for 10 nights and then rank their overall quality of sleep (due to the mattress only).
a. What is the explanatory variable? mattresses

b. What are the treatments? Expensive mattress &discount mattress

c. What is the response variable? Quality of sleep (ranking)

d. What are the individuals/subjects? Subjects = 110 adult volunteers

e. Design a matched pairs experiment:

11) 300 graduate students in psychology volunteer to be subjects in an experiment whose purpose is to look at the effect that
both dosage level and type of drug have on a performance task. There are 3 levels of drug to be tested: placebo, 200mg,
500mg, 750mg. There are two types of drugs to be tested as well: A and B. We will be measuring the students’ scores on
the performance task.
a. What are the treatments? (there are 8 of them)
Placebo & A Placebo & B
200mg & A 200mg & B
500mg & A 500mg & B
750mg & A 750mg & B

b. What are the explanatory variables? (there are 2 of them) Dosage and drug type
 c. What is the response variable? Students’ scores on a performance task

d. Who are the individuals/subjects? Subjects = 300 psychology grad student volunteers

e. Design a completely randomized experiment:

Placebo & A (37)

R
Placebo & B (38)
A
200mg & A (38) Compare
300 psych N
scores on
grad student 200mg & B (37)
D performance
volunteers
O 500mg & A (37) task

M 500mg & B (38)

750mg & A (38)

750mg & B (37)

OR

A (37)
R
Placebo (75)
R B (38)
A
A A (37) Compare
200mg (75) N scores on
300 psych N
B (38)
grad student performance
D D
volunteers A (37) task
O 500mg (75) O
B (38)
M
M
A (37)
750mg (75)
B (38)

f. What are some lurking/confounding variables?

Skill at the performance task (intelligence), other medications, sleep, etc.

12) You have developed a weight loss treatment that involves exercise and a diet pills. You have 400 overweight volunteers.
You are looking at 3 types of exercise programs (low, medium, high) and 2 types of diet pills (placebo and 100mg). You
will measure the percent of weight lost by the subjects.
a. What are the treatments? (there are 6 of them)
Low & placebo Low & 100mg
Medium & placebo medium & 100 mg
High & placebo high & 100 mg

b. What are the explanatory variables? (there are 2 of them) exercise program & diet pill
 c. What is the response variable? The percent of weight lost by the subjects

d. Who are the individuals/subjects? Subjects = 400 overweight volunteers

e. What are some lurking/confounding variables?

Genetics, whether they stick to the diet, food they consume, gender, additional exercise they may do, lifestyle,
etc.

f. Using one of the lurking variables that you identified in part (e), design a block design experiment:

Low & placebo * = Compare

R % of weight
Low & 100 mg lost
A
N Med & placebo
Male D
O
Med & 100 mg *
M High & placebo
400 High & 100 mg
overweight
volunteers
*
Low & placebo
R
Low & 100 mg
A
N Med & placebo
Female D
O
Med & 100 mg *
M High & placebo

High & 100 mg

13) A study of reactions to different types of advertising is to be carried out. Two different versions of ads for the same product
are to be tested. The organizer wants to see which version people prefer. There are 130 people available for the
experiment, of which 50 are men and 80 are women.
a. What are the treatments? (there are 2 of them) Advertisement A and Advertisement B

b. What is the explanatory variable? Advertisements

c. What is the response variable? Which Ad the people prefer

d. Who are the individuals/subjects? Subjects = 130 people

e. Design a matched pairs design experiment:

R
A
N * = Compare

130 people
D
O
Trt. 1 = Ad A, then Ad B (65)

Trt. 2 = Ad B, then Ad A (65)

* Ad
preference
M
 *
*

f. Gender is a lurking variable. Re-do your matched pairs design, but block for gender as well.

*
R Trt. 1 = Ad A, then Ad B (25)
A
50 men N
Trt. 2 = Ad B, then Ad A (25) *
*
D
O
130 people M
*
*
R Trt. 1 = Ad A, then Ad B (40)
80 women A
N
Trt. 2 = Ad B, then Ad A (40) *
*
D
O
M
14) It is known that in a specific city the chance that a person has a red hair is only 1 in 7. A researcher wants to conduct a
study to see on average how many people in Philadelphia have red hair in a sample of 5 people. Write instructions for a
simulation and conduct 10 trials. Clearly label each trial and state your conclusion for the average number of people with
red hair in Philadelphia.
58280 17867 07990 85055 55279 83390 37598 93350 05666 55402 87042

55080 76185 19947 79551 77594 87381 99430 44251 30896 72183 3985

94385 55160 50680 68443 95437 74302 06204 71004 76768 16066 94109

90685 92058 81744 99133 36354 34292 90092 21703 64616 03431 47610

31968 61593 36259 70600 53491 95542 78269 12087 32204 81177 30333
 15) The World Series ends when a team wins 4 games. Suppose that sports analysts consider the Phillies to have a 65%
chance of winning any individual game over their opponent. Estimate the likelihood that the underdog team (not the
Phillies) wins the series. Write instructions for a simulation and conduct 10 trials. Clearly label each trial and state your
conclusion.
31968 61593 36259 70600 53491 95542 78269 12087 32204 81177 30333

83630 06026 89308 94179 54907 58280 17867 07990 85055 55279 83390

37598 93350 05666 55402 87042 55080 76185 19947 79551 77594 87381

94109 90685 92058 81744 99133 36354 34292 90092 21703 64616 03431

47610 99430 44251 30896 72183 39856 94385 55160 50680 68443 95437
 16) Complete the following book problems: p. 288-291 #7, 9, 11, 13, 26, 35, 36
(for #35 & 36, list possible sources of bias with the sampling method listed)

MULTIPLE CHOICE:
ANSWERS:

1- D 11- B 19- B 22- D 24- D

2- B 17- C 20- A 52- C
3- C 18- D 21- B 97- E