Experimental Design

Directions:
Select the BEST response alternative for each of the questions below.

1. When two or more independent variables are simultaneously manipulated such that the results cannot be unambiguously attributed to any independent variable, ________ has occurred.

A) a confound

B) an experiment

C) a correlation

D) a quasi-experiment

2. Researchers concerned with good internal validity must ensure that only the ________ variable could have caused the results obtained.

A) dependent

B) independent

C) confounding

D) nusiance

3. In order for a researcher to be able to claim that the independent variable in her experiment caused the results, she must:

A) prove that the experiment has high external validity.

B) eliminate competing, alternative explanations.

C) eliminate experimental control and randomization.

D) keep the dependent variable constant.

4. To increase the likelihood of good internal validity in an experiment with random assignment of participants to conditions, researchers should:

A) remove confounding variables.

B) use a pretest.

C) use experimental control.

D) Both (A) and (C).

5. To determine the effect of types of music listened to while studying, Cleo uses two equivalent groups, one exposed to Trip-Hop music, the other exposed to Big Beat music. After studying history while listening to their assigned music styles, the participants complete a brief multiple-choice exam about the material studied. The type of experimental design employed in this example was a:

A) pretest-posttest design.

B) pretest-only design.

C) four-group posttest design.

D) posttestнн-only design.

6. What happens if the participant groups in an experiment are not equivalent?

A) The results are likely to be affected by ethnocentrism.

B) The experiment will high external validity, but low internal validity.

C) Replicating the experiment will be nearly impossible.

D) It will not be possible to determine that the independent variable caused the results.

7. William completed a study testing the effects of wall color on learning advanced physics. He tested students on the first day of class in two different physics classrooms (one pink and the other orange). William then tested the students half way through the semester, an again on the final day of classes. He found that the classes were about the same up through the second test, but differed significantly on the final test (pink outperformed orange). William also observed that 71 percent of the students in the orange classroom had dropped the class before taking the last exam. In this example, what is a likely alternative explanation for the results?

A) Maturation.

B) Attrition.

C) Statistical regression.

D) Diffusion of treatment.

8. An advantage of using a pretest-posttest design compared with using a posttest-only design is that it:

A) can establish whether the groups are equivalent.

B) helps participants figure out what the study is about.

C) requires less time and effort on the part of the researcher.

D) effectively eliminates demand characteristics and researcher bias.

9. When an experiment has a high mortality rate it means that participants:

A) have cheated on the response measures.

B) engaged in intragroup conflict.

C) dropped out of the study.

D) prevented other participants from responding honestly.

10. When using a Solomon four-group design, if there is no influence of the pretest, researchers will observe that:

A) posttest scores will be different for all groups.

B) pretest scores will be different for all groups.

C) posttest scores will be the same for all groups.

D) posttest scores will not be affected by the treatment.

11. In a ________ design, every participant is exposed to, and provides data for, all of the experimental conditions.

A) matched sets

B) independent groups

C) within-subjects

D) between-subjects

12. Sam wants to conduct an experiment to find out if putting apple seeds in people's ears will help treat their tinnitus. If she uses an independent groups design, she would manipulate the independent variable by:

A) randomly requiring half of the partitipants with tinnitus to put apple seeds in their ears and the other half to participate without apple seeds in their ears.

B) having all participants first respond with apple seeds in their ears, and then later respond without apple seeds in their ears.

C) having all participants first respond without apple seeds in their ears, and then later respond with apple seeds in their ears.

D) allowing participants to choose whether to have apple seeds placed in their ears or not.

13. Ted wants to conduct an experiment to find out if putting apple seeds in people's ears will help treat their tinnitus. If Ted uses a between-subjects design, He would manipulate the independent variable by:

A) randomly requiring half of the partitipants with tinnitus to put apple seeds in their ears and the other half to participate without apple seeds in their ears.

B) having all participants first respond with apple seeds in their ears, and then later respond without apple seeds in their ears.

C) having all participants first respond without apple seeds in their ears, and then later respond with apple seeds in their ears.

D) allowing participants to choose whether to have apple seeds placed in their ears or not.

14. One issue that almost always concerns researchers when they employ a repeated measures design is that the order of presentation of conditions will affect performance in addition to the independent variables. This is known as:

A) carryover effects.

B) incomplete counterbalancing.

C) indifference effects.

D) systematic counterbalancing.

15. When researchers observe an improvement in performance due to repeated exposure to a task, this is called a ________ effect.

A) reactive

B) fatigue

C) nonreactive

D) practice

16. When researchers observe a decline in performance due to repeated exposure to a task, this is called a ________ effect.

A) reactive

B) fatigue

C) nonreactive

D) practice

17. To control for order effects in a repeated measures design, researchers should make use of:

A) counterbalancing techniques.

B) independent groups designs.

C) reversal designs.

D) short time intervals between conditions.

E) All of the above.

18. A technique to control for order effects that does not require using all possible orders is called a:

A) pretest-only design.

B) partial randomization design.

C) complete counterbalancing design.

D) balanced Latin square design.

19. A researcher is most likely to use a matched pairs design when:

A) the sample size used is very large.

B) both the matching and dependent variables are unrelated.

C) only a small number of participants is available for the study.

D) using naturally paired volunteers (e.g., romantic couples, twins, business partners, etc.).

End of Quiz!

Your score out of 100%: %

The correct answers are marked by a "C" in the box before each question. The incorrect questions are marked by an "X".

1.	When two or more independent variables are simultaneously manipulated such that the results cannot be unambiguously attributed to any independent variable, ________ has occurred.
	A)	a confound
	B)	an experiment
	C)	a correlation
	D)	a quasi-experiment
2.	Researchers concerned with good internal validity must ensure that only the ________ variable could have caused the results obtained.
	A)	dependent
	B)	independent
	C)	confounding
	D)	nusiance
3.	In order for a researcher to be able to claim that the independent variable in her experiment caused the results, she must:
	A)	prove that the experiment has high external validity.
	B)	eliminate competing, alternative explanations.
	C)	eliminate experimental control and randomization.
	D)	keep the dependent variable constant.
4.	To increase the likelihood of good internal validity in an experiment with random assignment of participants to conditions, researchers should:
	A)	remove confounding variables.
	B)	use a pretest.
	C)	use experimental control.
	D)	Both (A) and (C).
5.	To determine the effect of types of music listened to while studying, Cleo uses two equivalent groups, one exposed to Trip-Hop music, the other exposed to Big Beat music. After studying history while listening to their assigned music styles, the participants complete a brief multiple-choice exam about the material studied. The type of experimental design employed in this example was a:
	A)	pretest-posttest design.
	B)	pretest-only design.
	C)	four-group posttest design.
	D)	posttestнн-only design.
6.	What happens if the participant groups in an experiment are not equivalent?
	A)	The results are likely to be affected by ethnocentrism.
	B)	The experiment will high external validity, but low internal validity.
	C)	Replicating the experiment will be nearly impossible.
	D)	It will not be possible to determine that the independent variable caused the results.
7.	William completed a study testing the effects of wall color on learning advanced physics. He tested students on the first day of class in two different physics classrooms (one pink and the other orange). William then tested the students half way through the semester, an again on the final day of classes. He found that the classes were about the same up through the second test, but differed significantly on the final test (pink outperformed orange). William also observed that 71 percent of the students in the orange classroom had dropped the class before taking the last exam. In this example, what is a likely alternative explanation for the results?
	A)	Maturation.
	B)	Attrition.
	C)	Statistical regression.
	D)	Diffusion of treatment.
8.	An advantage of using a pretest-posttest design compared with using a posttest-only design is that it:
	A)	can establish whether the groups are equivalent.
	B)	helps participants figure out what the study is about.
	C)	requires less time and effort on the part of the researcher.
	D)	effectively eliminates demand characteristics and researcher bias.
9.	When an experiment has a high mortality rate it means that participants:
	A)	have cheated on the response measures.
	B)	engaged in intragroup conflict.
	C)	dropped out of the study.
	D)	prevented other participants from responding honestly.
10.	When using a Solomon four-group design, if there is no influence of the pretest, researchers will observe that:
	A)	posttest scores will be different for all groups.
	B)	pretest scores will be different for all groups.
	C)	posttest scores will be the same for all groups.
	D)	posttest scores will not be affected by the treatment.
11.	In a ________ design, every participant is exposed to, and provides data for, all of the experimental conditions.
	A)	matched sets
	B)	independent groups
	C)	within-subjects
	D)	between-subjects
12.	Sam wants to conduct an experiment to find out if putting apple seeds in people's ears will help treat their tinnitus. If she uses an independent groups design, she would manipulate the independent variable by:
	A)	randomly requiring half of the partitipants with tinnitus to put apple seeds in their ears and the other half to participate without apple seeds in their ears.
	B)	having all participants first respond with apple seeds in their ears, and then later respond without apple seeds in their ears.
	C)	having all participants first respond without apple seeds in their ears, and then later respond with apple seeds in their ears.
	D)	allowing participants to choose whether to have apple seeds placed in their ears or not.
13.	Ted wants to conduct an experiment to find out if putting apple seeds in people's ears will help treat their tinnitus. If Ted uses a between-subjects design, He would manipulate the independent variable by:
	A)	randomly requiring half of the partitipants with tinnitus to put apple seeds in their ears and the other half to participate without apple seeds in their ears.
	B)	having all participants first respond with apple seeds in their ears, and then later respond without apple seeds in their ears.
	C)	having all participants first respond without apple seeds in their ears, and then later respond with apple seeds in their ears.
	D)	allowing participants to choose whether to have apple seeds placed in their ears or not.
14.	One issue that almost always concerns researchers when they employ a repeated measures design is that the order of presentation of conditions will affect performance in addition to the independent variables. This is known as:
	A)	carryover effects.
	B)	incomplete counterbalancing.
	C)	indifference effects.
	D)	systematic counterbalancing.
15.	When researchers observe an improvement in performance due to repeated exposure to a task, this is called a ________ effect.
	A)	reactive
	B)	fatigue
	C)	nonreactive
	D)	practice
16.	When researchers observe a decline in performance due to repeated exposure to a task, this is called a ________ effect.
	A)	reactive
	B)	fatigue
	C)	nonreactive
	D)	practice
17.	To control for order effects in a repeated measures design, researchers should make use of:
	A)	counterbalancing techniques.
	B)	independent groups designs.
	C)	reversal designs.
	D)	short time intervals between conditions.
	E)	All of the above.
18.	A technique to control for order effects that does not require using all possible orders is called a:
	A)	pretest-only design.
	B)	partial randomization design.
	C)	complete counterbalancing design.
	D)	balanced Latin square design.
19.	A researcher is most likely to use a matched pairs design when:
	A)	the sample size used is very large.
	B)	both the matching and dependent variables are unrelated.
	C)	only a small number of participants is available for the study.
	D)	using naturally paired volunteers (e.g., romantic couples, twins, business partners, etc.).