Ch. 11 Using Specialized Research Designs
An advantage of quasi-experimental designs is that they: A. allow you to evaluate the impact of a quasi-independent variable under naturally occurring conditions. B. include sufficient controls so that internal validity is preserved. C. are sufficiently general to allow you to ignore issues of external validity. D. All of the answers are correct.
a
Dr. Kent studies the development of emotions by following a group of 100 participants over a period of 10 years, taking measures every year. The design being employed in this scenario is the: A. longitudinal design. B. cross-sectional design. C. cohort-sequential design. D. Solomon four-group design.
a
In _____ designs, you chart changes in behavior as a function of some naturally occurring event. A. interrupted time series B. time series C. time samples D. Solomon four-group
a
In the _____, you select several participants from each of a number of age groups and create groups based on the chronological ages of the participants at the time of a study. A. cross-sectional design B. longitudinal design C. cohort-sequential design D. Solomon four-group design
a
In the nonequivalent control group design, _____. A. a time series component is included along with a control group that is not exposed to your treatment B. observations are made only after introducing your treatment C. observations are made only before introducing your treatment to a control group D. periods of observation are alternated with repeated introductions of your treatment
a
Subject mortality, a problem in one of the developmental designs, arises when: A. participants do not complete a study because they have lost interest in the study. B. subject loss is random. C. participants die but not when they leave on their own accord. D. None of the answers is correct.
a
When evaluating the impact of a new policy on the productivity of factory workers, you would most likely use a: A. pretest-posttest design. B. cross-sectional design. C. cohort-sequential design. D. None of the answers is correct.
a
According to Campbell and Stanley (1963), which of the following is a remedy for pretest sensitization? A. Using a cohort-sequential design B. Simply eliminating the pretest C. Allowing several weeks to elapse between the pretest and posttest D. None of the answers is correct.
b
By including a covariate in your experimental design, you can effectively: A. include multiple quasi-independent variables in a single experiment. B. subtract out the influence of a covariate from your dependent variable. C. add the influence of a covariate to your independent variable. D. None of the answers is correct.
b
Dr. Jonathon conducts a cross-sectional study to assess the changes in intelligence across the life span of participants. She observes a decline in intelligence test scores and concludes that intelligence deteriorates with age after middle adulthood. This conclusion: A. is totally correct and logical. B. may be incorrect because of generation effects. C. is incorrect because cross-sectional research lacks external validity. D. None of the answers is correct.
b
Dr. Jones conducts an experiment on visual perception. He has reason to believe that the gender of a subject is an important factor in perception. Consequently, he records the gender of his subjects and includes it as a variable in his data analysis. Assume that he finds a significant effect of gender in his data analysis. Dr. Jones is most likely to conclude that: A. the gender of the subject causes changes in perception. B. the gender of the subject is correlated with changes in perception. C. the gender of the subject has no effect on perception. D. the gender of the subject causes changes in perception and is also correlated with changes in perception.
b
In the _____ design, a treatment is administered and withdrawn repeatedly. A. nonequivalent control group B. equivalent time samples C. interrupted time series D. Solomon four-group
b
In the interrupted time series design, the naturally occurring event is a(n): A. pseudo-independent variable. B. quasi-independent variable. C. independent variable. D. simulated independent variable.
b
In time series designs, _____. A. participants are not exposed to any treatment B. observations are made before and after the introduction of a treatment C. time is an independent variable D. both experimental and control groups are included
b
One way to combine experimental and correlational variables is to include: A. a between-subjects and a within-subjects factor in an experimental design. B. a quasi-independent variable. C. multiple dependent variables in an experiment. D. All of the answers are correct.
b
The essence of the nonequivalent control group design is that: A. its validity remains unaffected even if both groups differ on some important variable before a study begins. B. a comparable group of subjects is chosen and observed for the same period as the group for which a treatment is introduced. C. when selecting groups on the basis of extreme scores on a pretest, a shift of posttest scores toward the less extreme values is because of the effect of a treatment. D. groups are created based on the chronological ages of the participants at the time of a study.
b
To ensure internal validity of a pretest-posttest design, you must: A. include a large sample of subjects. B. include control groups that are not exposed to your treatment. C. conduct your research in your subjects' natural environment. D. All of the answers are correct.
b
When using naturally occurring events as quasi-independent variables, one of the drawbacks is that: A. such events rarely occur. B. one has no control over when the event will occur. C. the events are usually too subtle to have a great impact on behavior. D. None of the answers is correct.
b
Which of the following is an advantage of a cross-sectional design? A. It remains unaffected by the influence of generational differences in experience. B. It permits one to obtain useful developmental data in a relatively short period. C. It effectively determines whether factors other than age influence the observed changes in behavior. D. It lets one evaluate the impact of a quasi-independent variable under naturally occurring conditions.
b
Which of the following statements is true about longitudinal designs? A. They are plagued by the problem of generation effects. B. In these designs, multiple observations of the same participants are made across time. C. They help determine factors other than age that influence the observed changes in behavior. D. In these designs, conclusions drawn from the study of a particular generation easily apply to another generation.
b
Which of the following statements is true about nested designs? A. They are less economical than factorial designs. B. They increase the generality of experimental results. C. They are also referred to as split-plot designs. D. They should be avoided when more than one task needs to be included under a level of an independent variable.
b
You have six different classes available for your experiment. You expose the first three classes to one level of your independent variable and the remaining three classes to the other level. In this experiment, you are: A. counterbalancing classes across treatments. B. nesting classes under different levels of your independent variable. C. using class as a correlational variable. D. None of the answers is correct.
b
Both mixed and nested designs: A. combine experimental and correlational variables in a single experiment. B. combine multiple independent and dependent variables in a single experiment. C. combine between-subjects and within-subjects manipulations in a single experiment. D. compare the results obtained from between-subjects experiments with the results obtained from separately conducted within-subjects experiments.
c
By including a covariate in your experimental design, you _____. A. can look at both between-subjects and within-subjects effects B. can assess the impact of carryover effects C. reduce error variance and improve the sensitivity of your experiment to the effect of your independent variable D. All of the answers are correct.
c
Campbell and Stanley (1963) point out that although the two-group pretest-posttest design ensures a degree of internal validity, _____. A. identifying causal relationships is difficult B. establishing the statistical significance of your results is difficult C. it does not preclude problems with external validity D. None of the answers is correct.
c
Subject mortality, testing effects, and the cross-generational effects are drawbacks associated with the: A. cohort-sequential design. B. cross-sectional design. C. longitudinal design. D. None of the answers is correct.
c
The Solomon four-group design: A. is used to eliminate the sensitization effects of a pretest. B. is used to control history effects in a pretest-posttest design. C. is used to test for any possible sensitization effects of a pretest. D. eliminates the need for a pretest.
c
The main disadvantage of including a quasi-independent variable in an experimental design is that: A. carryover effects become uncontrollable. B. the results obtained are difficult to analyze statistically. C. the results obtained are often misinterpreted. D. the resulting design cannot be carried out easily.
c
Which of the following designs is a quasi-experimental strategy and is most appropriate when the effects of a treatment are temporary or transient? A. The Solomon four-group design B. The cohort-sequential design C. The equivalent time samples design D. The cross-sectional design
c
Which of the following designs permits researchers to see developmental changes clearly and as they occur, thereby making the design worth the rather large investment of time it takes to collect data? A. The cohort-sequential design B. The interrupted time series design C. The longitudinal design D. The Solomon four-group design
c
You conduct a quasi-experiment to assess the impact of raising the speed limit from 55 to 65 miles per hour. You find out that there were more accidents at higher speeds. Although it is tempting to say that raising the speed limit caused higher accident rates, you must be careful because: A. you did not actually manipulate an independent variable. B. drivers may have exceeded the 65-mile-per-hour speed limit. C. other variables (for example, cheaper gasoline or the season of the year during which the change was instituted) may also be affecting accident rates. D. None of the answers is correct.
c
You include a set of four tasks under one level of your independent variable and a different set of four tasks under the other level. Your experimental design is a _____ design. A. Latin square B. combined C. nested D. completely counterbalanced
c
A major drawback of quasi-experimental research is that: A. you can study only one variable at a time. B. tight experimental controls often disrupt naturally occurring behavior. C. the results obtained often have too much external validity. D. you have no control over potential confounding variables.
d
Including a quasi-independent variable in your research allows you to: A. test the generality of your findings across the levels of the quasi-independent variable. B. reduce error variance by segregating the data into groups of participants who responded in similar fashion to the manipulation. C. analyze within-subjects and between-subjects factors in the same design. D. test the generality of your findings across the levels of the quasi-independent variable and reduce error variance by segregating the data into groups of participants who responded in similar fashion to the manipulation.
d
Which of the following statements is true about the cohort-sequential design? A. In this design, the naturally occurring event is a quasi-independent variable. B. In this design, one includes a time series component along with a control group that is not exposed to any treatment. C. This design is a variation on the basic time series design, and it charts changes in behavior as a function of some naturally occurring event. D. This design lets one evaluate the degree of contribution made by factors such as generation effects.
d
_____ allow you to mix within-subjects and between-subjects components. A. Mixed designs B. Nested designs C. Combined designs D. Both mixed and nested designs
d
A mixed design is a type of factorial design that combines between-subjects and within-subjects factors.
t
A mixed design is also called a split-plot design, and the term "split plot" comes from agricultural research in which the design was first developed.
t
A particular problem with the pretest-posttest strategy is that participants may be sensitized by the pretest, causing them to perform differently.
t
In a time series design, you make several observations of behavior over time prior to introducing a treatment and again after.
t
Many time series designs employ archival data.
t
Subject mortality and multiple testing effects are problems that a researcher should consider when choosing a longitudinal design.
t
The Solomon four-group design allows to test for pretest sensitization effects.
t
The cohort-sequential design embodies the features of both the cross-sectional and the longitudinal designs.
t
The major problems with quasi experiments relate to issues of internal validity.
t
