Deck 7: Experimental Designs

Question 1

Which of the following is valuable but non-essential feature of a true experiment?

A) random assignment to two groups
B) assessment of change in the dependent variable
C) complete control over the experiment's context
D) an experimental group and a control group

Answer 1

C

Question 2

Which of the following is a requirement for the pretest in an experiment?

A) The pretest must be a previously validated instrument.
B) The pretest cannot refer to the treatment to be tested.
C) The pretest must be given only to the comparison group, not to the experimental group.
D) The pretest must be the same as the posttest.

Answer 2

D

Question 3

Randomization in experimental research ______.

A) results in a sample that can be generalized to a population
B) removes the operation of systematic bias from the assignment process
C) guarantees that the experimental and comparison groups are identical at the start of the experiment
D) precludes the detection of relatively small outcome differences with reasonable confidence

Answer 3

B

Question 4

How do randomization and random sampling differ?

A) Both use of a random selection procedure, for example, flipping a coin.
B) They are the same.
C) The purpose of random sampling is to increase generalizability and the purpose of randomization is to decrease spuriousness.
D) Randomization is only used in quasi-experiments.

Question 5

When matching is used to assign subjects to an experimental and a comparison group, the resulting research design is:

A) Equivalent to when random assignment is used
B) Weakened if it is combined with random assignment
C) Superior to assignment on the basis of chance
D) Weakened by the possibility of additional differences between the groups

Question 6

A researcher is interested in examining the effect that participation in D.A.R.E. has on future drug use. In her experiment, she identifies one student who has completed the D.A.R.E. program and matches the student to another who is similar in most respects except he or she has not completed D.A.R.E. This is an example of ______.

A) randomization
B) random selection
C) individual matching
D) aggregate matching

Question 7

The key feature of an experimental design that distinguishes it from a quasi-experimental design is the ______.

A) longitudinal measurement of the dependent variable
B) random assignment to two or more groups
C) comparison of the dependent variable before and after the independent variable
D) elimination of error due to chance

Question 8

A researcher wants to study the effectiveness of a midnight basketball program in reducing juvenile delinquency. She identifies a community that already has a program in place and compares delinquency rates to a community that does not. This is an example of which type of quasi-experimental design?

A) non-equivalent control group design
B) before-and-after design
C) ex post facto control group design
D) this cannot be a quasi-experiment.

Question 9

Before-and-after designs are distinguished from other quasi-experimental designs by which of the following?

A) the absence of a traditional comparison group
B) a comparison group designated after the treatment is administered
C) a posttest-only comparison group
D) random assignment to a comparison group

Question 10

Which of the following is the reason that ex post facto control group design should be considered nonexperimental rather than quasi-experimental?

A) It does not employ random assignment to the experimental and comparison groups.
B) It lacks a posttest.
C) It lacks internal validity.
D) The comparison group is designated only after the intervention has occurred.

Question 11

A quasi-experimental design is just as strong on which of the following criterion as an experimental design?

A) nonspurious relationship between variables
B) association between the independent and dependent variables
C) time order of effects of one variable on another
D) none of these

Question 12

A researcher is seeking to maximize generalizability of results. Which research design should she use?

A) laboratory experiment
B) field experiment
C) non-experiment
D) qualitative

Question 13

In a study examining the effectiveness of an education program, subjects are given a pretest. Their scores on the posttest increased because they learned information in the pretest and not as a result of the education program. This is an issue of:

A) Testing
B) Regression
C) External event
D) Maturation

Question 14

In a study examining the effectiveness of an SAT prep course, the researcher is considering how at prettests some students might have been having a bad day and their scores were artificially low. This is an issue of ______.

A) testing
B) maturation
C) regression
D) subject attrition

Question 15

A researcher is conducting a study examining residents' attitudes toward police. During data collection, news emerges that many police officers were involved in corruption. This is an issue of ______.

A) testing
B) maturation
C) external event
D) subject attrition

Question 16

Multiple pretest and posttest measures are valuable in before-and-after designs because they decrease the possibility of internal invalidity due to ______.

A) selection
B) external events
C) compensation
D) interaction of testing and treatment

Question 17

In medical research, the placebo effect occurs when subjects ______.

A) feel no different in the experimental group than in the control group
B) complain about being members of the control group
C) act passively throughout the experiment
D) feel better just because they think they are being given a drug

Question 18

Since social programs are delivered by human beings, ______ can be very difficult to control.

A) expectancy effects
B) regression
C) maturation
D) history

Question 19

A Hawthorne effect occurs when ______.

A) members of the control group change because they feel disadvantaged compared with the treatment group.
B) members of the treatment group change due to positive expectations of the experimental staff.
C) members of the treatment group change because their participation in the study makes them feel special.
D) members of the control group are influenced by the experience of testing at baseline.

Question 20

The features of experimental designs that make them so useful for establishing causation ______.

A) enhance the cross-population generalizability of a study
B) reduce the sample generalizability of a study
C) limit the internal validity of a study
D) prove to be a disadvantage in measurement validity

Question 21

A randomized field experiment is designed to test the usefulness of a new DNA test for solving crimes. Crime clearance rates in police departments that use the new DNA test will be compared with crime clearance rates in police departments that use traditional DNA tests. The fact that some departments will not be using this new DNA test to solve crimes during the experiment may be ethically justifiable when ______.

A) citizens do not know about the new DNA test
B) there is no way of knowing whether the DNA test is of any value to solving crime
C) DNA tests have become affordable to police departments
D) the research is conducted in a high-crime area

Question 22

A researcher is testing the effectiveness of a new boat paddle by holding a race between two teams. When the control team realizes that it received traditional paddles instead of the new paddles that the experimental team received, the control team members feel as though they have no chance of winning and put forth little effect in the race. This is an example of ______.

A) external events
B) endogenous change
C) demoralization
D) compensatory rivalry

Question 23

All of the following are accurate statements about pretests and posttests EXCEPT ______.

A) in order to be an experiment, there must be a posttest
B) in order to be an experiment, there must be a pretest
C) some experiments have multiple posttests and pretests
D) pretests provide a direct measure of change over time

Question 24

Which design has no comparison group?

A) nonequivalent
B) before-and-after
C) ex post facto
D) experiment

Question 25

A time series (repeated measures) design is a type of ______.

A) nonequivalent control group
B) before-and-after
C) ex post facto control group
D) randomization

Question 26

Which of the following is NOT a source of internal invalidity?

A) selection bias
B) external events
C) contamination
D) statistical control

Question 27

All of the following are types of endogenous change EXCEPT ______.

A) testing
B) maturation
C) regression
D) selection bias

Question 28

In the process of conducting an experiment, a researcher administers a survey before the treatment is applied, in order to provide a baseline against which to judge the results. This survey is a ______.

A) controlling variable
B) pretest
C) posttest
D) selection bias

Question 29

Which of the following is NOT required for a true experiment?

A) random assignment
B) control group
C) posttest
D) pretest

Question 30

A researcher conducts a study on the effectiveness of a new prisoner rehabilitation program. All of the elements of an experiment are present except that the groups were not randomly assigned. This study is an example of a(n) ______.

A) nonequivalent control group design
B) before-and-after design
C) ex post facto control group design
D) regression-discontinuity design

Question 31

All of the following are accurate statements about factorial surveys EXCEPT which of the following?

A) Factorial surveys have many of the elements of an experiment.
B) Factorial surveys are utilized to increase generalizability.
C) Factorial surveys are quasi-experimental as they lack a control group.
D) Factorial surveys randomly alter elements of the vignettes.

Question 32

In a study of the effectiveness of a new cancer drug, the researcher administering the medication is unaware of which patients are receiving the new drug and which are receiving a placebo. This is an example of ______.

A) the self-fulfilling prophecy
B) a double-blind procedure
C) the Hawthorne effect
D) generalizability

Question 33

Which of the following is an accurate statement about the Solomon four-group design?

A) Not all of the groups receive the pretest.
B) All of the groups receive the pretest.
C) Not all of the groups receive the posttest.
D) All of the groups receive the posttest.

Question 34

A worthwhile, conservative strategy for minimizing the possibility of invalid causal conclusions due to differential attrition is ______.

A) ensuring at the outset of the experiment that subject assignment is truly random
B) recruiting only subjects with stable residential histories and ongoing institutional ties
C) combining matching with randomization
D) basing the outcome analysis on comparing all subjects initially assigned to each condition

Question 35

______ is a defining feature of a true experiment.

A) Comparison group
B) Random assignment
C) Dependent variable
D) Independent variable

Question 36

The ______ can be considered as the variable that is manipulated in an experiment.

A) dependent variable
B) control variable
C) treatment variable
D) outcome variable

Question 37

Multiple ______ can be adminstered to examine how long a treatment effect may be observed.

A) prettests
B) experiments
C) posttests
D) random assignments

Question 38

Random assignment removes ______ from the assignment process.

A) bias
B) confusion
C) randomness
D) none of these

Question 39

An example of a variable that is often used in matching procedures would include which of the following?

A) age
B) gender
C) year in school
D) all of these

Question 40

In the Sherman and Berk (1984) domestic violence experiment, which of the following was considered the treatment condition?

A) no arrest
B) separation
C) arrest
D) none of these

Question 41

Matching is a more preferable strategy than randomization.

Question 42

Attrition can cause experimental and control groups to differ over time.

Question 43

In a factorial survey design, all of the respondents complete some version of the survey.

Question 44

Process analysis is one way to minimize treatment misidentification.

Question 45

In the Solomon four-group design, some groups do not receive a pretest.

Question 46

A true experiment requires a pretest and a posttest.

Question 47

Sherman and Beck's study of the effect of arrest on domestic violence is an example of a field experiment.

Question 48

Random assignment eliminates systematic bias.

Question 49

A quasi-experiment fares better in achieving generalizability than a true experiment.

Question 50

Treatment misidentification is when either the experimental group or the comparison group is aware of the other group, and as a result, influences the posttest.

Question 51

True experiments can have multiple control groups.

Question 52

Random assignment guarantees that the groups are perfectly identical at the beginning of the experiment.

Question 53

It is difficult, if not impossible, to establish a spurious relationship between an independent and dependent variable in a nonexperimental research design.

Question 54

Quasi-experiments have more internal validity than true experiments.

Question 55

True experiments have greater measurement validity than quasi-experiments.

Question 56

Matching procedures involves random assignment.

Question 57

Field experiments refer to experiments that are performed in a laboratory.

Question 58

Individuals that were born in 1980 would be considered a cohort.

Question 59

In a nonequivalent control group design, the individuals serve as their own control group.

Question 60

A time-series design does not involve a prettest.

Question 61

Discuss the degree of control over conditions in Sherman and Berk's experiment on domestic violence. To what extent did the degree of control over conditions affect the internal validity of each experiment?

Question 62

Why is it difficult to conduct true experiments in the social sciences?

Question 63

What can a researcher do to improve the external validity of a true experiment??

Question 64

Imagine that you are the lead member of research team and you are concerned with treatment misidentification. What policies could you implement to minimize this concern?

Question 65

Provide an example of a type of endogenous change. What can researchers can do to minimize these threats?

Question 66

Explain what the ethical concern regarding the selective distribution of benefits.

Question 67

What is the main difference between experimental and quasi-experimental designs?

Question 68

Why is random assignment so important to experimental research design?

Question 69

What are the advantages of using a Solomon four-group design over an experimental design with one control group and one experimental group?

Question 70

How can matching be used by researchers to improve the quality of their research?

Question 71

Describe a criminal justice research project where a time series design would be appropriate.

Question 72

Define attrition and differential attrition, and discuss their impact on experiments.

Question 73

Describe several reasons why Sherman and Berk's (1984) domestic violence experiment experienced concerns with generazability.

Question 74

List and describe three types of endogenous changes that can occur in an experiment and influence its results.