Ch. 4 Choosing A Research Design
A demonstration is a kind of research design that resembles an experiment but lacks one of the crucial features of a true experiment: an independent variable.
t
A threat to internal validity comes in the form of confounding variables.
t
A tightly controlled laboratory experiment affords the researcher a relatively high degree of internal validity but reduces external validity.
t
Causal relationships can be clearly established only with experimental research.
t
Control over extraneous variables is an important characteristic of experimental research.
t
In experimental research, you manipulate independent variables and look for related changes in the value of the dependent variable.
t
In laboratory research, you have a high degree of control over the variables that could affect your results.
t
One way to control the effects of extraneous variables is to randomize their effects across treatment groups.
t
Sometimes, steps taken to ensure high levels of internal validity lead to reductions in the levels of external validity.
t
Subjects in the control group are not exposed to your experimental treatment.
t
Subjects in the experimental group are exposed to your experimental treatment.
t
The first step in designing a simulation is to familiarize yourself with the real-world phenomenon to be simulated.
t
The third-variable problem is one reason why you should not infer causality from correlational data.
t
The value of a dependent variable is determined by a subject's behavior.
t
Your ability to distinguish causal from correlational relationships depends on the degree of control you have over the variables.
t
Your choice of a research design affects the quality of the conclusions you can draw from your data.
t
In a _____ relationship, changes in one variable produce changes in another. A. correlational B. hypothetical C. confounded D. causal
d
In a simple two-group experiment, the group of subjects receiving the treatment is called the _____ group. A. demonstration B. treated C. control D. experimental
d
In the context of a research design, which of the following is a factor affecting external validity? A. Instrumentation B. Experimental mortality C. History D. Reactive testing
d
The main difference between correlational and experimental research is that in experimental research you: A. do not manipulate independent variables. B. measure more than one dependent variable. C. do not manipulate more than two independent variables. D. manipulate independent variables and look for changes in the dependent variables.
d
Using participants who have been selected on the basis of their extreme scores on some measure introduces the possibility of _____ as a source of confounding. A. history B. maturation C. biased selection D. statistical regression
d
Which of the following is listed in the text as a factor affecting internal validity? A. Participants growing older or becoming fatigued B. Changes in the calibration of instruments or observers used C. Biased selection of subjects D. All of the answers are correct.
d
A confounding variable poses little threat to internal validity if: A. its effect on the dependent or criterion variable is known to be small. B. the experimenter knows that the confounding variable exists. C. its effect on the dependent or criterion variable cannot be determined. D. the experimenter believes that its effect on the dependent or criterion variable is small.
a
A difference between correlational and experimental research is that in experimental research you: A. try to hold extraneous variables constant during the course of your experiment. B. include extraneous variables in your experiment to increase the chances of finding causal relationships. C. worry less about the effects of extraneous variables on your results. D. do not manipulate independent variables.
a
Although tight control over experimental conditions gives you high levels of _____, it may result in low levels of _____. A. internal validity; external validity B. external validity; internal validity C. reliability; validity D. validity; reliability
a
An experiment explored the effect of alcohol consumption (1, 3, or 5 drinks) on simulated driving. The simulated driving constitutes the _____ variable. A. dependent B. independent C. correlational D. extraneous
a
Conducting your research in the laboratory provides an advantage because you: A. can exercise tight control over extraneous variables. B. need not worry about internal validity. C. have a high level of external validity. D. need not worry about external validity.
a
Dr. Smith varies the temperature of her laboratory (70F, 80F, or 90F) and notices that her subjects' aggression increases with temperature. This illustrates: A. a causal relationship between temperature and aggression. B. a correlational relationship between temperature and aggression. C. an unsystematic relationship between temperature and aggression. D. no relationship between temperature and aggression.
a
If not controlled, extraneous variables can: A. produce uncontrolled changes in your dependent variable. B. produce uncontrolled changes in your independent variable. C. cause your subjects to drop out of your experiment. D. All of the answers are correct.
a
If the effects of an extraneous variable cannot be held constant, it is best to: A. randomize its effects across treatments. B. totally redesign your experiment. C. ignore it and hope that it does not affect your results. D. inform your subjects about the extraneous variable and tell them to ignore it.
a
In an experiment, internal validity means: A. showing that changes in your independent variable caused observed changes in your dependent variable. B. that most extraneous variables have been identified and controlled. C. showing that changes in your criterion variable relate only to changes in your predictor variable. D. None of the answers is correct.
a
In the context of a research design, a demonstration differs from an experiment in that a demonstration: A. exposes a group of subjects to only one treatment condition. B. involves the manipulation of an independent variable. C. aids in identifying relationships among variables. D. is unable to show what happens under a specified set of conditions.
a
In the context of research design, exploratory data collection and analysis involves: A. identifying the variables involved in explaining the behavior of interest. B. a "trapping" process after information has been collected about a behavior. C. forming potential explanations for the behavior of interest. D. testing tentative explanations for the behavior of interest.
a
Performance changes observed over time due to age or fatigue may confound changes to the variable being studied. This source of confounding is called _____. A. history B. maturation C. statistical regression D. biased selection
a
The directionality problem: A. means that sometimes you cannot clearly specify the direction of causality of a relationship observed in a correlational study. B. differs between experimental and correlational research. C. is limited to experimental research. D. deals with the effects of extraneous variables on an experiment.
a
The term _____ refers to the degree to which your research design evaluates the hypothesis that it was intended to evaluate. A. internal validity B. external validity C. internal consistency D. reliability
a
The values of a(n) _____ variable are set by the experimenter. A. independent B. dependent C. extraneous D. confounding
a
When specific events occur between multiple observations taken over time, confounding due to _____ may result. A. history B. maturation C. statistical regression D. biased selection
a
Which of the following is an obstacle in the way of drawing clear causal inferences from correlational data? A. The directionality problem B. Probability sampling C. The extraneous variable problem D. Nonresponse bias
a
_____ aims at classifying behaviors in a given research area and identifying the variables involved in explaining the behavior of interest. A. Exploratory data collection and analysis B. Inferential statistics C. Statistical analysis D. Naturalistic observation
a
A(n) _____ relationship is one in which values of variables covary in some way, but not necessarily causally. A. confounded B. experimental C. correlational D. causal
c
According to the text, _____. A. studies should be conducted so that the results can be applied to real-world behavior B. the results from research need not always predict behavior in the real world C. when conducting research, you should not concern yourself with matters of external validity D. much laboratory research is designed to tell us what actually happens in the real world
b
According to the text, confounding variables: A. are a greater threat to the internal validity of experimental research than to that of correlational research. B. are a greater threat to the internal validity of correlational research than to that of experimental research. C. threaten experimental and correlational research equally. D. are not a serious threat to either experimental or correlational research.
b
According to the text, correlational research is useful: A. when you want to establish causal relationships but cannot manipulate independent variables. B. in the early stages of research to identify variables that might be causally related. C. when you do not have the time or money to conduct an experiment. D. All of the answers are correct.
b
According to the text, the time to be concerned about internal validity is: A. when you conduct statistical tests. B. during the design phase of your research. C. after you have run your study. D. All of the answers are correct.
b
According to the text, the two defining characteristics of experimental research are: A. measuring two or more variables and determining whether they covary. B. manipulating an independent variable and controlling extraneous variables. C. randomly assigning subjects to conditions and manipulating an independent variable. D. None of the answers is correct.
b
Dr. Smith conducts a correlational study of the relationship between religious affiliation and attitudes toward abortion. In this example, religious affiliation would most likely be used as a(n) _____ variable and attitudes toward abortion as a _____ variable. A. independent; dependent B. predictor; criterion C. criterion; predictor D. experimental; correlational
b
Dr. Wong conducts a correlational study and is able to determine that variation in his criterion variable relates solely to changes in his predictor variable. Dr. Wong's study is high in: A. external validity. B. internal validity. C. construct validity. D. predictive validity.
b
Hypothesis testing normally begins: A. when all variables affecting the behavior of interest have been identified. B. after you have collected enough information about a behavior to begin developing supportable explanations. C. during the preliminary exploratory stages of research. D. None of the answers is correct.
b
If a causal relationship exists between variables, then: A. the value of your independent variable will depend, in part, on the value of your dependent variable. B. the value of your dependent variable will depend, in part, on the value of your independent variable. C. there will be no strong relationship between your independent and dependent variables. D. None of the answers is correct.
b
Internal validity is threatened by: A. running your experiment under highly artificial conditions. B. the operation of extraneous, uncontrolled variables. C. the presence of quasi-independent variables in your study. D. choosing the wrong strategy to answer your questions.
b
The degree to which your research results extend beyond your study refers to the study's: A. internal validity. B. external validity. C. general validity. D. None of the answers is correct.
b
The great strength of experimental research is that: A. you can identify variables that vary together. B. you can identify and describe causal relationships. C. you control extraneous variables. D. extraneous variables can be totally eliminated.
b
To make a demonstration into a true experiment, you must: A. choose subjects to be exposed to a single treatment at random from a population. B. have an independent variable. C. measure more than one dependent variable. D. make sure that all extraneous variables are held constant within a single treatment group.
b
Which of the following is an advantage of interpreting results of a study using a correlational rather than an experimental approach? A. It involves easy manipulation of variables to determine their effects on other variables. B. It helps in determining how naturally occurring variables relate in the real world. C. It incorporates a high degree of control over the variables of a study. D. It allows one to investigate and manipulate extraneous variables that affect an experiment.
b
_____ is defined as a scientific activity that consists of forming tentative explanations for behavior and then validating those tentative explanations. A. Demonstration B. Hypothesis testing C. Statistical analysis D. Simulation
b
_____ is defined as the ability of a research design to adequately test a hypothesis. A. Demonstration B. Internal validity C. Experimental mortality D. Instrumentation
b
A common complaint about using animals such as rats as research subjects is that: A. animals are often too dumb to successfully complete experiments. B. the results obtained may lack internal validity. C. the results obtained may not apply to the behavior occurring under natural conditions in the "real world." D. None of the answers is correct.
c
A design in which subjects are exposed to only one treatment condition is called a(n): A. correlational design. B. uncontrolled experimental design. C. demonstration. D. illustration.
c
A weakness of experimental research is that: A. causal relationships cannot be discovered. B. extraneous variables cannot be controlled. C. sometimes you cannot manipulate the variables of interest. D. All of the answers are correct.
c
According to the text, if during the design phase of your research you discover a problem with internal validity, you should: A. abandon the research completely and start over. B. ignore the problem and go ahead with your study because maintaining internal validity is not that important. C. identify the source of the problem and redesign your study. D. wait until after you have run your study to determine if internal validity was seriously threatened.
c
According to the text, the objective of research is to: A. tell us what actually happens in the real world of behavior. B. tell us what ought to happen in the real world of behavior. C. gain insight into the underlying mechanisms of behavior. D. discover relationships that apply to normal conditions in the real world.
c
An advantage of field experimentation is that: A. you can control extraneous variables more tightly than you can in a laboratory experiment. B. you can be assured of high levels of internal validity. C. your results may have more external validity than the results from a laboratory experiment. D. you can control extraneous variables more tightly than you can in a laboratory experiment and you can be assured of high levels of internal validity.
c
Correlational research belongs to a broader category called _____. A. relational research B. a demonstration C. nonexperimental research D. None of the answers is correct.
c
Dr. Ames conducts a correlational study of the relationship between maternal age and attachment. He finds that the two variables are significantly related. Dr. Ames can safely conclude that the two variables are: A. causally related. B. causally related, but the direction of causality is not clear. C. related in some way, but no causal inference should be made. D. None of the answers is correct.
c
Dr. Harrison writes down the names of her subjects on slips of paper and puts them in a box. She shuffles them and begins to draw names one at a time. She assigns every other subject to the experimental group of her experiment. This is an example of: A. holding extraneous variables constant across the levels of the independent variable. B. systematically assigning subjects to the conditions of the experiment. C. randomly assigning subjects to the conditions of the experiment. D. None of the answers is correct.
c
Experimental mortality refers to a source of confounding in which: A. age or fatigue affects the experimental results. B. subjects are selected on the basis of their extreme scores on a measure. C. subjects drop out of a study more often in one condition of the experiment than in another because of the effects of the condition. D. uncontrolled events occur between multiple observations and affect the experimental results.
c
If you are interested in conducting basic research, you might be more concerned with _____ than with _____. A. confounding; establishing causal relationships B. establishing causal relationships; confounding C. internal validity; external validity D. external validity; internal validity
c
In a simple two-group experiment, the group of subjects not receiving the treatment is called the _____ group. A. demonstration B. treated C. control D. experimental
c
In correlational research, you: A. manipulate only two independent variables. B. manipulate one dependent variable and look for changes in a second. C. observe two or more variables and determine what relationships exist between them. D. typically do not develop specific hypothesis.
c
In correlational research, your main interest is to: A. uncover clear causal connections between variables. B. re-create the real world as closely as possible in the laboratory. C. find out whether two or more variables covary and to establish the direction of any observed relationship. D. All of the answers are correct.
c
Manipulating an independent variable involves: A. exposing subjects to at least one level of the independent variable. B. measuring two or more behaviors and determining whether they covary. C. exposing subjects to at least two levels of the independent variable. D. exposing subjects to at least three levels of the independent variable.
c
One way to control the influence of extraneous variables is to: A. pretend they do not exist when analyzing your data. B. tell your subjects to ignore any extraneous variables. C. hold extraneous variables constant across the levels of your independent variable. D. All of the answers are correct.
c
The third-variable problem means that: A. a correlational study with fewer than three variables is invalid. B. if two predictor variables are shown to be important predictors of the value of a criterion variable, then there is probably a third important predictor variable as well. C. in a correlational study there may be a third, unobserved variable that actually causes changes in the observed variables. D. None of the answers is correct.
c
To manipulate an independent variable, one must expose participants to at least two levels of that variable. The specific conditions associated with each level are called the _____ of the experiment. A. controls B. simulations C. treatments D. measures
c
Unobserved changes in criteria used by observers or in instrument calibration contribute to confounding due to: A. history. B. maturation. C. instrumentation. D. experimental mortality.
c
Variables are said to be _____ if their effects cannot be separated. A. extraneous B. confused C. confounded D. None of the answers is correct.
c
When you cannot manipulate independent variables or want to observe naturally occurring variables, the strategy of choice is a(n): A. experimental research. B. simulation. C. correlational research. D. None of the answers is correct.
c
_____ can sometimes improve the generality of results from laboratory research. A. Including a quasi-independent variable B. Tightly controlling confounding variables C. Conducting a simulation D. All of the answers are correct.
c
A weakness of the experimental strategy is that: A. causal relationships cannot be discovered. B. extraneous variables cannot be controlled. C. experiments can be done only in the laboratory. D. tight control over extraneous variables may limit the generality of your results.
d
A(n) _____ variable is a variable that may affect the behavior you are investigating but is not of interest to you. A. independent B. control C. correlational D. extraneous
d
According to Mook (1983), with regard to external validity, much laboratory research is designed to tell us what _____ happen, rather than what _____ happen in the real world. A. ought to; can B. does; can C. does; ought to D. can; does
d
According to the text, _____. A. confounding variables are a greater problem in experimental research than in correlational research B. confounding variables are of little concern because they do not pose a great threat to internal validity C. confounding variables can easily be eliminated in most types of research D. it is often impossible to eliminate all sources of confounding in a study
d
According to the text, nonexperimental research: A. includes only research intended to find a relationship between variables. B. always includes more than one independent variable. C. cannot be used to find a relationship between variables. D. includes designs not specifically aimed at identifying relationships between variables.
d
Dr. Jones conducts a correlational study of the relationship between the preference for violent television programming and aggression in children. She finds that the two variables are correlated and concludes that an increasing preference for violent television programs causes more aggression. Which of the following is most likely to be a potential obstacle in drawing a clear, causal inference from such correlational data? A. Probability sampling B. The extraneous variable problem C. Nonresponse bias D. The third-variable problem
d
Dr. Jones is interested in studying the effect of the number of bystanders present at an emergency on the number of subjects who help the victim. She varies the number of bystanders (1, 3, or 5) and records how many subjects help under each condition. In this example, the number of bystanders is the _____ variable. A. dependent B. extraneous C. correlational D. independent
d
If an extraneous variable provides an alternative explanation for your results, then a(n) _____ hypothesis exists. A. alternative B. conflicting C. duplicate D. rival
d
If you are interested in conducting applied research, you might be more concerned with _____ than with _____. A. confounding; establishing causal relationships B. establishing causal relationships; confounding C. internal validity; external validity D. external validity; internal validity
d