Deck 7: Randomized Experiments and Causal Inference

Question 1

An experimental design that administers one or more levels of one independent variable in combination with two or more levels of another independent variable is called a

A) between-subjects design.
B) within-subjects design.
C) simple randomized design.
D) simple factorial design.

Answer 1

D

Question 2

Because he is limited in the number of subjects he can recruit, Tom decides to test his hypothesis that studying to music improves test performance by comparing individuals' test performance after studying to music to their test performance after studying in silence. This is an example of which type of design?

A) simple factorial
B) between-subjects
C) within-subjects
D) nested

Answer 2

C

Question 3

Within-subjects designs are also called

A) crossed designs.
B) nested designs.
C) counterbalanced designs.
D) matched designs.

Answer 3

A

Question 4

Between-subjects designs are also referred to as

A) nested designs.
B) matched designs.
C) counterbalanced designs.
D) crossed designs.

Question 5

The activating force or event that is responsible for an effect is the __________ cause of that event.

A) efficient
B) final
C) formal
D) material

Question 6

Between-subjects designs in which participants are assigned to treatment conditions by means other than randomization are referred to as

A) cross-lagged designs.
B) longitudinal designs.
C) nonequivalent-groups designs.
D) single-case experimental designs.

Question 7

Which of the following is NOT a reason given for incorporating random assignment into one's research design?

A) Random assignment permits the use of certain statistical tests.
B) Random assignment prevents possible biases in determining which treatment a subject receives.
C) Random assignment increases the reliability and validity of one's experimental design.
D) Random assignment distributes the characteristics of the sampling units over all experimental conditions.

Question 8

Notions of causality that refer to the objective or end purpose of an event are referred to as __________ causes.

A) efficient
B) material
C) formal
D) final

Question 9

In order to avoid bias in how subjects are assigned to the different treatment conditions, experimenters should use

A) random assignment.
B) counterbalancing.
C) random sampling.
D) self-selection procedures.

Question 10

Notions of causality based on the substance or substances necessary for the movement or coming into being of the effect are referred to as __________ causes.

A) formal
B) material
C) final
D) efficient

Question 11

Experimental designs that expose subjects to more than one treatment condition each are referred to as

A) expectancy control designs.
B) repeated-measures designs.
C) nested designs.
D) between-subjects designs.

Question 12

An explanation of causality based on the plan or development that gives meaning to the event is an example of which type of causality?

A) formal
B) efficient
C) final
D) material

Question 13

Randomized experiments are experimental designs that involve

A) the manipulation of variables.
B) the use of humans as research participants.
C) the use of random assignment to treatment conditions.
D) the strict adherence to ethical guidelines.

Question 14

Tom believes that studying while listening to music will improve test performance. To test this hypothesis, Tom has one group of participants listen to music while studying for a spelling test. A second group of participants study for the test in silence. Tom then compares the spelling test performance for both groups. Tom's study is best described as which type of experimental design?

A) between-subjects
B) within-subjects
C) repeated measures
D) simple factorial

Question 15

Dave explains that what caused there to be a new library on campus was the implementation of the architectural plans. According to Aristotle, Dave is focusing on the __________ cause for why there is a new library on campus.

A) material
B) formal
C) final
D) efficient

Question 16

To minimize problems in repeated-measures designs that can result from systematic differences between successive measurements, a researcher should use a method called

A) random assignment.
B) counterbalancing.
C) nesting.
D) expectancy control.

Question 17

The order in which the treatments are administered can unintentionally create problems for the experimenter who is employing which research design?

A) simple factorial designs
B) repeated-measures designs
C) between-subjects designs
D) nested designs

Question 18

When asked to explain what caused there to be a new library on campus, Dave explains that the mortar and bricks are responsible. According to Aristotle, Dave is focusing on the __________ cause for why there is a new library on campus.

A) material
B) formal
C) final
D) efficient

Question 19

When asked what caused there to be a new library on campus, Dave explains that there was a desire to enhance the quality of the students' education. According to Aristotle, Dave is focusing on the __________ cause for why there is a new library on campus.

A) formal
B) efficient
C) material
D) final

Question 20

Experimental designs that expose subjects to only one treatment condition each are referred to as

A) randomized designs.
B) repeated-measures designs.
C) within-subject designs.
D) between-subjects designs.

Question 21

Ruling out plausible rival hypotheses is central to satisfying which criteria for establishing efficient causality?

A) internal validity
B) covariation
C) temporal precedence
D) external validity

Question 22

While conducting a year-long study on friendship patterns in the sixth grade, a researcher discovers that some of the students had reached puberty during that year while others had not. Which potential threat to internal validity would this example illustrate according to Campbell and his colleagues?

A) selection
B) instrumentation
C) maturation
D) history

Question 23

An educator is interested in whether including a physically challenged student in a kindergarten class will enhance the other students' level of compassion. The educator assesses the students' level of compassion at the beginning of the year, before the physically challenged student was assigned to the classroom. At the end of the school year, the educator reassesses the students' level of compassion and notes whether overall compassion has increased or decreased. This is an example of which type of research design according to Campbell and his colleagues?

A) the posttest-only control group design
B) the one-shot case study design
C) an expectancy control design
D) the one-group pre-post study

Question 24

Mill's method of agreement corresponds to which group in a simple research design?

A) the placebo group
B) the control group
C) the experimental group
D) the randomized group

Question 25

Which of the following is an example of a preexperimental design?

A) the one-group pre-post study design
B) the Solomon design
C) the pre-post control-group design
D) the posttest-only control-group design

Question 26

Which of the following evidence is particularly difficult to obtain when one is trying to determine efficient causality based on relational research?

A) internal validity
B) covariation
C) temporal precedence
D) external validity

Question 27

Mill's method of agreement implies that X is a(n) __________ condition of Y.

A) necessary
B) sufficient
C) important
D) activating

Question 28

Dave explains that what caused there to be a new library on campus was the large donation by a wealthy alumnus of the university. According to Aristotle, Dave is focusing on the __________ cause for why there is a new library on campus.

A) formal
B) final
C) efficient
D) material

Question 29

Which of the following issues makes inferences of causation more risky in quasi-experimental designs?

A) lack of treatment manipulations
B) nonrandom assignment to treatment conditions
C) unreliable outcome measurements
D) the use of both human as well as nonhuman sampling units

Question 30

By establishing that the independent variable and dependent variable are correlated with each other, one has satisfied which criteria for establishing efficient causality?

A) internal validity
B) temporal precedence
C) random assignment
D) covariation

Question 31

Which of the following is NOT a characteristic of quasi-experimental designs?

A) the use of treatment conditions
B) the use of randomization to allocate sampling units
C) the use of outcome measures
D) the recruitment of sampling units

Question 32

In a drug study, one group of subjects receives a sugar pill rather than the pill containing the ingredients under investigation. This "sugar pill" group is referred to as the __________ control group.

A) placebo
B) experimental
C) zero
D) expectancy

Question 33

All of the following are criteria for determining efficient causation EXCEPT

A) temporal precedence.
B) external validity.
C) covariation.
D) internal validity.

Question 34

Research designs that do not employ the use of randomization are referred to as

A) nonrandomized experimental designs.
B) correlational designs.
C) quasi-experimental designs.
D) pseudo-experimental designs.

Question 35

While participating in a timed experiment, the participants in the experimental group had to leave the room because of a fire drill. Which potential threat to internal validity would this example illustrate according to Campbell and his colleagues?

A) maturation
B) selection
C) instrumentation
D) history

Question 36

To assess the effectiveness of a special lesson plan, a teacher implements the lesson plan and then evaluates the effectiveness of the plan by analyzing the students' test scores upon completion of the learning unit. This is an example of which type of research design according to Campbell and his colleagues?

A) simple factorial design
B) one-group pre-post study
C) one-shot case study
D) posttest-only control-group design

Question 37

I selected my 7^th grade classmates as the control group and a group of 7^th graders from another school as the treatment group. What bias has a good chance of impacting this study?

A) bias due to history
B) the posttest-only control-group design
C) bias due to maturation
D) biased selection

Question 38

Mill's method of difference implies that X is a(n) __________ condition of Y.

A) activating
B) sufficient
C) necessary
D) adequate

Question 39

Mill's method of difference provides the rationale for including a(n) __________ group in an experimental design.

A) treatment
B) experimental
C) control
D) expectancy control

Question 40

The control group was measured using a different measuring tape than the treatment group. This is an example of which potential bias?

A) bias due to testing
B) instrumentation bias
C) bias due to history
D) biased selection

Question 41

What is a one-shot case study? Why is it inadequate for demonstrating efficient causation?

Question 42

How do Mill's methods form the logical basis of all simple randomized designs?

Question 43

Describe six threats to internal validity identified by Campbell and his colleagues.

Question 44

A researcher used an expectancy control design in her study of the effectiveness of a new teaching method. Students were randomly assigned to conditions, and the following group means were found (where higher means reflect better performance): Assuming that all of the differences are statistically significant, what can the researcher conclude?

A) The old teaching method is more effective.
B) The condition with the expectation that the method does not work performed better.
C) The new teaching method, combined with the expectation that the method works, performed the best.
D) Expectations did not have an impact on student performance.

Question 45

One strategy that researchers can use to identify potential demand characteristics in their experiment is to employ the use of

A) good subjects.
B) quasi-control subjects.
C) expectancy control designs.
D) double-blind procedures.

Question 46

One way of minimizing the experimenter expectancy effect is to employ the use of __________ experimenters.

A) blind
B) control
C) neutral
D) quasi

Question 47

What is the difference between within-subjects and between-subjects research designs? Give an example of each type of design.

Question 48

Unintended cues that can influence a subject's behavior in an experiment are referred to as

A) expectancy cues.
B) demand characteristics.
C) residual effects.
D) quasi-experimental manipulations.

Question 49

After conducting an experiment, Linda realizes that she smiled more often when interacting with the experimental group than when interacting with the control group. Which artifact may Linda have introduced into her experiment?

A) the double-blind procedure
B) the good subject effect
C) the Hawthorne effect
D) the experimenter expectancy effect

Question 50

Which of the following refers to a finding that results from conditions other than those intended by the experimenter?

A) selection bias
B) artifact
C) good subject effect
D) experimenter expectancy effect

Question 51

The expectancy control design is an example of which type of research design?

A) repeated measures design
B) quasi-experimental design
C) preexperimental design
D) simple factorial design

Question 52

Describe three reasons why one should use random assignment of subjects to treatment conditions.

Question 53

The notion that subjects are often motivated to "look good" when participating in an experiment is because they are

A) trying to be good subjects.
B) being employed as quasi-control subjects.
C) experiencing experimenter expectancy effects.
D) experiencing evaluation apprehension.

Question 54

Describe the three criteria for establishing efficient causation.

Question 55

A researcher notices that the participants assigned to the experimental group were less likely to attend the second session in his experiment than those in the control group. The researcher is concerned that __________ might represent a potential threat to the internal validity of her study.

A) history
B) attrition
C) instrumentation
D) maturation

Question 56

How do randomized experiments differ from other types of experiments?

Question 57

A researcher becomes aware that the judges he is using to make observations are becoming more proficient at their task. This improvement among the judges might suggest that __________ represents a potential threat to the internal validity of the study.

A) instrumentation
B) history
C) maturation
D) selection

Question 58

The observation that sometimes subjects are extremely willing to complete a meaningless task in order to help the experimenter illustrates the idea of

A) the experimenter expectancy effect.
B) the subject selection bias.
C) the good subject effect.
D) the Hawthorne effect.

Question 59

What are the four kinds of causation one may refer to when explaining events? Which type of causation do scientists refer to when they say that one thing "caused" something else?

Question 60

Why is counterbalancing important when using a within-subjects design with repeated treatments and measurements?

Question 61

How can the experimenter expectancy effect bias the results of an experiment? How can this bias be minimized?

Question 62

What is a quasi-control subject? How can this assist in discovering demand characteristics in an experiment?

Question 63

What is an artifact? Why are artifacts of concern to researchers?

Question 64

What is the good subject effect? How can this effect be minimized in a study?