Deck 5: Randomized Experimental and Quasi-Experimental Designs

Question 1

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-Dudley Dooright is doing a study on Canadian Mounties. Dudley will do an experimental Polly Purebread training with one group of Mounties followed by an assessment at the end.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Answer 1

one group posttest only design

Question 2

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-Wiley E. Coyote wants to assess a new Acne Product on roadrunners. He has two groups of roadrunners. One group receives the new Acne Product, and the other group gets the old Acne Product. Wiley measures Roadrunner escape speed before and after getting the Acne Products.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Answer 2

pretest-posttest nonequivalent control group design

Question 3

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-Fred Flintstone wants to do a study on Bedrock vs. Stonetown with one evaluation at the end. Bedrock will get Rubble training and Stonetown will get no training.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Answer 3

posttest-only design with nonequivalent groups

Question 4

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-Scooby-Doo wants to assess the relationship between the presence or absence of ghosts on fear. He randomly assigns a group of college students to either the ghost group or the no- ghost group. The ghost group will visit a spooky, haunted house, and the no-ghost group will visit a spooky house without ghosts. Both will be assessed on fear prior to and after the experiment.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Question 5

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-A therapist is using a new type of hypnosis on clients who are afraid of heights. She measures clients' fear of heights at one week intervals for one month prior to treatment. The clients undergo hypnosis treatment. The therapist measures the clients' fear of heights at one week intervals for one month following the hypnosis therapy.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Question 6

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-A university is instituting a new cultural diversity program. Annual admission rates of minority students are examined for the three years prior to the program's onset. Annual admission rates are tracked for the first three years of the program.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Question 7

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-A group of bald patients are randomly assigned to either a treatment group or a control group. The treatment group receives an anti-baldness medication for one month, and the control group receives a placebo sugar pill for one month. Patients' new hair growth is measured at the end of the month.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Question 8

match the example with its specific design name. You may use the design name more than once. Also diagram the design.

-There are two high schools in a community that is interested in a new anti-drug program. One high school receives the new anti-drug program, and the other high school decides to wait a year before beginning the new program. Both high schools collect information on students' drug use at the beginning and end of the school year.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
E) single-group time-series design with temporary treatment
F) single-group time-series design with continuous treatment
G) posttest only control group design
H) pretest-posttest control group design

Question 9

Mick Jagger wants to do a hearing study on aging rock stars. He has asked old rock stars to assign themselves to one of two groups: a group that rocks for 7 days straight and a group that doesn't rock at all for 7 days. He will do a pretest and posttest on their hearing. This study is:
Explain your answer.

A) one group posttest only design
B) one group pretest-posttest design
C) posttest-only design with nonequivalent groups
D) pretest-posttest nonequivalent control group design
Self-assignment raises doubts about the equivalence of these two groups

Question 10

You have been contacted by a businesswoman who wants you to evaluate her company's leadership training program. She wants to know whether there are pretest influences on leadership assessment. She wants a strong design. What design will you suggest for her?
Diagram your design

A) Solomon four-group design
B) Posttest-only control or comparison group design
C) Pretest-posttest design
D) Posttest-only design with nonequivalent groups

Question 11

You have been contacted by a businesswoman who wants you to evaluate her company's leadership training program. She wants to know whether there are pretest influences on leadership assessment. She wants a strong design. What design will you suggest for her? Solomon four-group design
Diagram your design

Question 12

A researcher is interested in the effects of ethics training on corporate executive decision-making. She contacts three similar companies. These three companies receive either an ethics training course, a traditional leadership course, or no course (control). Which training they receive is delivered randomly. The participating executives are assessed before and after training. This design is:
Explain your answer. There is random assignment of treatments to similar intact groups. Assume there are no significant pretest differences between the group

A) Poor experimental design
B) Poor quasi-experimental design
C) Moderate quasi-experimental design
D) Strong quasi-experimental design

Question 13

A researcher is interested in the effects of ethics training on corporate executive decision-making. She contacts three similar companies. These three companies receive either an ethics training course, a traditional leadership course, or no course (control). Which training they receive is delivered randomly. The participating executives are assessed before and after training. This design is: Strong quasi-experimental design
Explain your answer. There is random assignment of treatments to similar intact groups. Assume there are no significant pretest differences between the group

Question 14

Possible problems with the one group pretest-posttest design include:

A) Maturation effects
B) Carryover effects
C) Ecological validity
D) a and b

Question 15

Problems with the weak pretest-posttest quasi-experimental design include:

A) Random assignment of groups to treatments/conditions
B) Participants assign themselves to groups
C) The researcher does not have control over the IV
D) b and c

Question 16

The strength of a pretest-posttest quasi-experimental design is influenced by:

A) Similarity of participant characteristics
B) Random sampling of the accessible population
C) Random assignment of treatments to intact groups
D) a and c

Question 17

Time series designs:

A) Are useful when there is no control group
B) Should only be used when groups can be randomly assigned to treatments
C) Require at least one measure before and after an intervention
D) a and b

Question 18

There is increased confidence that random assignment of participants to groups yields equivalent groups when:

A) There is a small sample size
B) Groups are fairly homogeneous
C) There is a control group
D) All of the above

Question 19

A pretest-posttest control group design should be used when:

A) There is a small sample size
B) There is a large sample size
C) The groups are homogeneous
D) a and c

Question 20

Some poor quasi-experimental designs have only one level so no comparisons can be made.

Question 21

Poor quasi-experimental designs have an attribute IV with only one level.

Question 22

The one-group posttest-only design can provide pilot information for future research.

Question 23

Internal validity is threatened with the one-group pretest-posttest design.

Question 24

Adding more DVs will increase the value of the one-group pretest-posttest design.

Question 25

Nonequivalence of groups does not affect the strength of a quasi-experimental design.

Question 26

In a strong quasi-experimental design, the researcher has control over the IV.

Question 27

With a time-series design, several pretests help establish a baseline.

Question 28

Random assignment of participants to groups always makes the groups equivalent.

Question 29

The pretest-posttest control group design is always a quasi-experimental approach.

Question 30

The Solomon four-group design is a quasi-experimental approach.

Question 31

The Solomon four-group design allows the researcher to test the effects of the pretest.

Question 32

When matching is used, the characteristic that is matched must not be related to the DV.

Question 33

Matching is used when the IV only has one level.

Question 34

When matching is used, participants are matched in pairs on a characteristic that might influence the DV and then are randomly assigned to groups.

Question 35

What are the pros and cons of using the four types of quasi-experimental designs with major limitations?

Question 36

How does random assignment affect bias?

Question 37

When conducting a pretest-posttest nonequivalent comparison group design is it possible to know that the groups are equivalent?

Question 38

What differentiates randomized experiments from quasi-experiments?