Deck 12: Quasi-Experimentation

Question 1

If you investigated the problem solving ability of children of different ages and your study consisted of comparing the performance of a group of six-year olds to that of a group of eight-year olds,your design would be:

A) cross-sectional
B) longitudinal
C) cross-sequential
D) time lag

Answer 1

cross-sectional

Question 2

One way to increase the internal validity of an interrupted time series design is to:

A) use deviant case analysis
B) include a nonequivalent control
C) randomly assign variables to subjects
D) use cross-lagged panel correlations

Answer 2

include a nonequivalent control

Question 3

Each of the following is a subject variable except:

A) age
B) sex
C) religious beliefs
D) memory ability
E) learning instructions

Answer 3

learning instructions

Question 4

A subject variable in a quasi-experiment is a characteristic of the subject that:

A) the researcher selects
B) the researcher uses to define the dependent measure
C) cannot be directly measured
D) directly manipulated by the researcher
E) does not interact with other characteristics of the subject

Question 5

Which is generally considered to be the poorest quasi-experimental design due to the lack of a baseline or control group?

A) multiple baseline
B) one-shot case study
C) interrupted time-series
D) natural treatment

Question 6

The purpose of matching in designs with subject variables is to:

A) directly measure the influence of matched variables across groups
B) avoid the possibility of selection bias
C) increase the sensitivity of the dependent variable
D) minimize confounding

Question 7

When the combined effects of two or more variables are not additive (i.e.,they interact),this could represent a demonstration of ____.

A) regression to the mean
B) an artifact
C) a nonequivalent control
D) synergism

Question 8

All of the following problems are associated with matching in a research design using subject variables except the notion that:

A) it may be impossible to match the groups
B) measures of some relevant matching variables may be difficult to take
C) interactions among matched variables may occur
D) regression artifacts can occur
E) it typically increases the likelihood of confounding with the variable of interest

Question 9

A quasi-experiment differs from a true experiment because in a quasi-experiment:

A) the researcher directly manipulates the independent variable
B) the laboratory environment is never used
C) there is greater internal validity
D) researcher selects levels of the independent variable

Question 10

A nonequivalent control for a one-shot case study is:

A) a randomly selected member of the population
B) other individuals who have received the same treatment
C) a very similar individual who did not receive the treatment
D) a deviant relative of the subject in the case study

Question 11

Results from one-shot case studies should be interpreted cautiously because:

A) the subjects are deviant
B) the treatment is long
C) the external validity is low
D) there is no baseline

Question 12

A researcher selected three groups on the basis of I.Q.(smart,average,dopey).All subjects were matched on the basis of age,weight,and health.The smart subjects ran a mile faster than the average ones,who ran faster than the dopey ones.The researcher can conclude that:

A) the time to run a mile is related to intelligence
B) being smart allows you to run a mile quickly
C) being stupid causes you to run a mile slowly
D) running speed is determined by intelligence

Question 13

If you compared the problem solving ability of a group of six-year olds to their own performance when they were eight years old,your design would be:

A) cross sectional
B) longitudinal
C) cross-sequential
D) time lag

Question 14

In the observation-treatment-observation situation,a typical threat to validity is the fact that:

A) natural treatments are likely to produce long-term carry-over effects
B) changes in the subjects' behavior result from experiences of the participants prior to the first observation period
C) a nonequivalent control group always results in selection bias
D) participants are randomly assigned to treatment and control conditions

Question 15

Which of the following is a problem with using an interrupted time-series design?

A) It is difficult to establish a matched control group.
B) Participants often fail to complete the study.
C) The effect of the treatment can be delayed by other variables.
D) All of the above are potential problems.

Question 16

In an interrupted time series design,there are all of the following except:

A) multiple observations before a naturally occurring treatment
B) multiple observations after a naturally occurring treatment
C) interruptions caused by tightly controlled treatments
D) interruptions caused by uncontrolled variables other than the treatment

Question 17

The phenomenon of regression to the mean implies that if your parents are slightly more intelligent than the average,your intelligence is probably:

A) above average and greater than your parents'
B) above average but less than your parents'
C) average
D) below average

Question 18

Which of the following might occur as a result of regression to the mean?

A) deviant cases might be found to be non-deviant
B) cohorts might be found to be of different ages
C) matched subjects might be found to be mismatched
D) experiments might be found to be quasi-experiments

Question 19

You are interested in surveying attitudes about gambling as a function of age.Assume that you decide to conduct a poll today about gambling expenditures.This poll will consist of an equal number of people at 18,24,30,and 60 years of age.This is an example of which age-related method?

A) cross-sectional
B) longitudinal
C) time-lag
D) cross-sequential

Question 20

In designs that use subject variables,matching refers to:

A) matching subjects on the subject variable
B) matching subjects on likely confounding variables
C) matching subjects on known unconfounded variables
D) matching subjects on the task used in the study

Question 21

The most serious confounding that occurs in the cross-sectional design is between:

A) age and treatment
B) age and generation
C) age and time of test
D) generation and time of test

Question 22

In quasi-experiments,the independent variable is not under direct control of the experimenter.

Question 23

Unlike the one-shot case study,the internal validity of the interrupted-time-series design is threatened by mortality.

Question 24

The major difference between the ABA design and the observation-treatment-observation design is that the former design permits a return to baseline.

Question 25

The performance of two groups of Research Methods students (matched for age)is compared.The first group is from the Spring 2003 term,whereas the second group is from Spring 2004 term.This represents an example of a ____ design.

A) time-lag
B) longitudinal
C) cross-sectional
D) cross-sequential

Question 26

The most serious confounding that occurs in the longitudinal design is between:

A) age and treatment
B) age and generation
C) age and time of test
D) generation and time of test

Question 27

A problem with matching on the basis of subject variables is synergism with other variables.

Question 28

The advantage of the time-lag design is that it provides the researcher a chance to discover:

A) confounding between age and generation
B) the influence of the time of testing
C) confounding between age and cohorts
D) the distinction between age and maturation

Question 29

One primary limitation of a cross-sectional method is:

A) the fact that it can confound age and the generation of birth
B) there is a greater likelihood of participant mortality
C) that regression to the mean could occur
D) that it is time-consuming and cumbersome to use

Question 30

In the cross-sequential design:

A) two or more groups are tested at the same age
B) two or more groups are tested at different ages
C) the same subjects are tested at different ages
D) two or more groups of different ages are tested at two or more time periods

Question 31

Phillips' studies on sporting events indicate that imitation of prize-fight winners accounts for an increase in homicide rates.

Question 32

The major threat to validity in the one-shot case study is the small number of subjects that are examined.

Question 33

If an experimental treatment had different effects on subjects of different ages,we would conclude that:

A) there was a confounding in the study
B) there was an interaction
C) the study lacked external validity
D) a longitudinal design is more appropriate

Question 34

If a schizophrenic performs poorly on an intelligence test,it is likely that schizophrenia is a cause of the poor performance.

Question 35

Which age-related design could be described as potentially cumbersome and time-consuming,but tends not to suffer from confounds (e.g.,of age and time,and selection bias)that plague other age-related designs?

A) cross-sectional
B) longitudinal
C) time-lag
D) cross-sequential

Question 36

An interaction between age and an experimental treatment means that:

A) the results of the study are ambiguous
B) the experimental treatment influenced age
C) age influenced the experimental treatment
D) the effect of the experimental treatment differed at various ages

Question 37

The nonequivalent control group design involves randomly assigning subjects to conditions.

Question 38

Synergism is a synonym for interaction.

Question 39

Although case studies have shaky internal validity,they are good at predicting future behavior.

Question 40

In the time-lag design:

A) subjects of different ages are tested at the same time to determine generation effects
B) subjects of the same age are tested at different times to determine effects of time of testing
C) the same subjects are tested at different times to determine the effects of age
D) the same subjects are tested at different ages to determine the effects of time of testing

Question 41

Regression to the mean implies that if your parents are taller than average that you will be very tall as well.

Question 42

Age is the most common true independent variable used in developmental research.

Question 43

The most common developmental research design is the cross-sectional design,which is a within-subjects design.

Question 44

The cross-sequential design confounds age with time of testing.

Question 45

The longitudinal design used in developmental research is a kind of within-subjects design.