Chapter 9-14
Experimenter *bias* refers to
differences in *how the experimenter treats the participants.*
A *confounding* variable is an example of ___________, whereas an *extraneous* variable is an example of __________.
*systematic* error; *random* error
A linear relationship that moves from the top left corner of the scatterplot to the lower right corner of the scatterplot would be which of the following?
Negative
Which of the following is likely to decrease experimentwise alpha in a factorial design?
Using post hoc rather than a priori comparisons
Before administering a manipulation designed to improve the mood of research participants, each participant completes a mood questionnaire. After the mood change is created, the participants then complete another mood questionnaire. The original questionnaire is used as a(n) __________.
baseline measure
*Time-series* designs
include *repeated measurements*.
A researcher who uses only college students in her research may be trying to
increase *experimental control*.
Differences on a dependent measure across the levels of one factor that ignore differences on all other factors are called
main effects.
A *post* hoc comparison
makes comparisons that were *not planned ahead of time*.
A 5 x 6 x 2 design has
sixty conditions and three factors.
The Pearson product-moment correlation coefficient is used to assess the association between
two quantitative variables.
Which of the following is an advantage of *experiments*?
The *experimenter's ability* to draw conclusions about the causal relationship between variables is maximized.
Which of the following is least likely to be a threat to validity in a *single- group* before-after design?
*Selection*
A schematic diagram is used to
*diagram* the specific predictions of the research hypothesis.
A researcher found a positive correlation between income and job satisfaction in a sample of one hundred adults. The researcher should conclude
*he cannot tell which.*
A design in which behavior is initially measured during a baseline period, and then measured again after the intervention of interest begins, is called a
*single group before-after* design.
A study that assesses how a patient with a leg fracture responds to a *new physical therapy program* is an example of a
*single-participant* design.
The goal of treating all experimental participants in *exactly the same way* with the simple exception of the manipulation itself is known as
*standardization* of conditions.
*Experimental* realism is the extent to which
*the experimental manipulation* involves the participants in the research.
A disadvantage of using convenience samples, such as college students, is that
*the findings might not be generalizable* to other samples.
A crossover interaction occurs in a 2 x 2 factorial design when
*the simple effects are opposite* in direction.
An experiment produces a between-groups variance estimate (mean square) of 25 and a within groups variance estimate (mean square) of *5*. The F value is therefore
5
Research that uses *two groups of individuals*, one of which has had and one of which has not had the experience, is known as which of the following?
A *comparison-group design*
When there are many different levels of the independent variable in a repeated-measures design, it may not be possible to counterbalance using every order of conditions. Which of the following can be used in such a case?
A Latin square design
A researcher finds that height and intelligence are correlated in children. The age of the child is probably which of the following?
A common-causal variable
Which of the following is a table showing the *correlations* of many variables with each other?
A correlation matrix
Marginal means indicate the means of which of the following?
A main effect
When we say that there is a relationship between two variables but that this relationship only occurs under certain conditions, then the variable that represents the conditions under which the effect does or does not occur is known as which of the following?
A moderator variable
Which of the following analyses might be used to study the relationships in a longitudinal research design?
A path analysis
Patricia is told by the experimenter that the lights in the experimental room may put people in a positive mood. Although the lights do not actually have any effect on mood, Patricia nevertheless begins to feel a positive mood. The change in mood probably represents which of the following?
A placebo effect
Which of the following is NOT a quasi-experimental research design?
Matched-participant
A *complex* comparison is
a means comparison in which *more than two means* are compared at the same time.
A scientist hypothesizes that divorce can lead to depression and that depression can then lead to poor physical health. In this case, depression can be considered
a mediating variable.
Differences on the dependent measure between the levels of one factor within one level of another factor are known as
a simple effect.
The possibility that a confounding variable, *rather* than the independent variable of interest, caused changes on the dependent measure represents
an *alternative* explanation.
A baseline measure is
an *initial measurement* of the dependent variable in a before-after research design.
Findings that do not generalize
are still useful because *they may provide information* about the limitations of prior studies.
Structural *equation* analysis is a technique that can be used to
assess causal relationships using *correlational* data.
Longitudinal designs are to cross-sectional designs as
before-after designs are to comparison-group designs.
Whereas the main effects compare marginal means, the simple effects compare
cell means
In general, ______________ are more problematic than ________________ because they can produce spurious relationships.
common-causal variables; extraneous variables
Results of a study show that children who watch a nonviolent cartoon act more aggressively when they are hungry than when they are not hungry, but that children who watch a violent cartoon act more aggressively when they are not hungry than when they are hungry. This finding represents a
crossover interaction.
Aspects of the research that allow the participants to guess the research hypothesis are known as
demand characteristics.
Having more than two levels of the independent variable is necessary to
detect *nonlinear* relationships.
In a study of the relationship between participant age and memory, the gender of the participant would represent a(n) _____ variable.
extraneous
Whereas a correlation matrix contains a series of Pearson correlation coefficients, a contingency table contains
frequencies
*Factor*ial designs are *advantageous* because
hypotheses can be tested that *could not be tested in one-way designs*.
The extent to which changes in the dependent variable can confidently be attributed to the influence of the independent variable is known as
internal validity.
The results of a study show that, regardless of whether they viewed cartoons or television shows, children who viewed a violent show acted more aggressively than children who saw a nonviolent show. This finding represents a
main effect.
Consider the following relationship between violent material and aggressive behavior: Violent TV -> arousal -> aggressive behavior. In this case, arousal can be described as a
mediating variable
A researcher who studies the outcomes of prior experiments by statistically analyzing effect size measures is conducting a
meta-analysis.
In terms of learning about generalization across participants,
no sample of research participants can provide definitive information about how a finding will generalize.
An experimenter is interested in the influence of taking a one-semester-long test preparation course on performance on the Scholastic Aptitude Test (SAT). She contacts high school students who have and who have not taken the course and compares their scores on the SAT exam. She finds that the students who have taken the course score significantly higher on the test. The appropriate statistical test is
one-way *ANOVA*
Which of the following topics would be most appropriate to study as an independent variable in an experimental design?
participant *mood*
Which of the following r values shows the strongest correlation between two variables?
r = -.8
A researcher found that *conclusions* regarding his research were incorrect because a Type 1 error had been made. His error represents a type of
statistical *conclusion* invalidity.
In *experimental* research, the temporal priority of the independent variable is guaranteed through the use of
the *experimental* manipulation.
In an experimental research design, the research hypothesis is that
the *means* on the dependent variable will be different in the different levels of the independent variable.
The difference between the t test and the F test is that
the *t test* is used to compare *two group* means, whereas the F test is used to compare any number of means.
Reciprocal causation occurs when
the predictor variable causes the outcome variable and *vice versa.*
Whereas *internal* validity refers to _________, *external* validity refers to _________.
the relationship between the *independent* and the dependent variable; the extent to which the findings are *expected* to generalize
In an experimental design that has three levels of the independent variable, a significant F value indicates that
the three means are *not the same*, but we cannot tell which means are different from each other.
An experimenter studies the influence of ingesting sugar on memory. Children are randomly assigned either to eat a *low calorie* ice cream cone (low sugar condition) or a *Ben and Jerry's cone* (high sugar condition). They are then asked to learn ten vocabulary words that they did not previously know, and the number of trials it takes to learn them is measured. The independent variable in this experiment is
the type of ice cream cone.
The difference between a factorial experimental design and a *one-way* experiment is that
there is *more than one* independent variable in the factorial design.
Pairwise comparisons within factorial designs are problematic because
they increase the probability of a *Type 1 error*.
A 2 x 2 x 3 design has
three factors and twelve conditions.
A 2 x 3 x 4 factorial design has
three factors and twenty-four conditions.
An experimental design in which the dependent measure is assessed for one or more groups of individuals more than twice, at regular intervals, both before and after the experience of interest occurs is called a
time-series design.
*Manipulation* checks are particularly useful
when there is *no significant relationship found* between the independent and dependent variable.
In scientific research, *valid*ity refers to
whether the conclusions drawn about the research are *correct.*
Quasi-experimental designs are usually analyzed using
*ANOVA*
Which of the types of validity is threatened when a researcher commits a *Type 1 or Type 2 error*?
*Statistical* conclusion validity
Which of the following ways represents the correct reporting of results according to APA format?
*There were significant differences on rated aggression* across the levels of the cartoon condition, F(1,38) = 10.98, p < .01.
Which of the following represents the greatest limitation to drawing conclusions from a single-group design?
*They cannot be used to draw conclusions* about the how the experience influenced the participants.
Which of the following best describes confound checks?
*They help determine whether the manipulation* has unwittingly caused differences on other variables that we did not desire to manipulate.
How many main effects and interactions are there in a 2 x 4 x 3 design?
*Three* main effects and *four* interactions
In a repeated-measures design with *two* levels, the dependent measure will need to be measured on each participant how many times?
*Twice*
Which of the following is an advantage of before-after designs?
*Random error* is reduced.
Which of the following is NOT a method by which experimenters control for *extraneous* variables?
*Random* assignment to conditions
Which of the following is a possible disadvantage of *before-after* research designs?
*Retesting* effects
Amanda conducted a 2 x 2 factorial design in which she manipulated lack of food (high or low) and aggressive stimulation (high or low) in dogs. She then measured barking level. Amanda found that in the high food condition more aggressive stimulation reduced barking, whereas in the low food condition more aggressive stimuli increased barking. An interaction in which such *opposite effects* occur in different levels of different conditions might be which of the following?
A *cross*over interaction
Stephen and Pham (2007) measured *trust* in feelings by having participants list either 2 or 10 situations in which they had trusted in their feelings. The researchers then measured the extent to which the participants thought that they would trust in their feelings if they were to perform a task. The researchers used this measures as which of the following.
A *manipulation* check.
A means comparison in which any one condition mean is *compared* with any other condition mean is known as which of the following?
A *pairwise* comparison
To determine whether a beneficial influence of therapy that has been *found in a sample of women* would also be *found in a sample of men*, the researcher will need to use which of the following?
A *participant replication*
Elliot and Niesta (2008) tested the hypothesis that men would find a woman more attractive when she was surrounded by a red background than when she was surrounded by other colors. In their first study the researchers compared a *woman in front of a red background* and a *woman in front of a white background*. The possibility that in this study the white background was brighter than the red background represents which of the following?
An *alternative* explanation.
An artifact refers to which of the following?
An *aspect of the research methodology* that may produce confounding
A *repeated-measures* design refers to which of the following?
An experiment in which equivalence is created by having the same people participate in *more than one condition*
Which of the following variables causes only the outcome variable, but not the predictor variable?
An extraneous variable
The term *factor* refers to which of the following?
An independent variable in a *factor*ial design
If we find that there is a positive correlation between variable x and variable y, which of the following is true?
Any of the above possibilities may be correct.
Which of the following statements is true about quasi-experimental designs?
Because *the independent variable is measured, not manipulated,* they are best considered as correlational research.
Which of the following is true regarding the *participant-variable* design?
Because *the participant variable is measured*, causal conclusions about the effect of the participant variable on the dependent variable are *difficult to draw*.
Which of the following refers to a measured variable used to determine whether the manipulation has unwittingly caused differences on other variables that might be systematically related to the independent variable?
Confound check
Which of the following are necessary in an experimental design?
Random assignment and *experimental manipulation*
Parr, Kavanagh, Cahill, Young, and Mitchell (2008) conducted a meta- analysis to study the effects of different types of treatments on the success of tranquilizer withdrawal. The dependent measure in this study is which of the following?
The *effect size of the studies*
Which of the following statistics is used as the dependent measure in a meta-analysis?
The *effect size*
A researcher studies the cognitive development of children by measuring *kindergarten*, *fourth grade*, and *seventh grade* children. Which type of design is being used?
Cross-sectional
When the light level of a computer screen is moderate, task performance scores are high. When the light level of the computer screen is either very low or very high, task performance scores are low. The relationship between the light level and performance is
Curvilinear
In order to compare the means of levels of task performance when mood state is manipulated experimentally, a researcher would likely use which of the following statistical tests?
F test
Which of the following statistics are used to show the relationships between variables in a path analysis?
Regression coefficients
A flight instructor notices a particularly unusual phenomenon. Every time one of her students has a particularly excellent flight, she praises them for it. And she notices that the next time the student flies, he or she ends up doing worse. On the other hand, every time one of her students has a particularly poor flight, she criticizes them, and the next time they do better. What phenomenon might the instructor be seeing?
Regression to the mean
A student gets a 99 out of 100 on her first quiz in a psychology class, but gets only a 91 on the next quiz. This is a likely example of which of the following?
Regression to the mean
An experimenter is interested in the influence of taking a one-semester-long test preparation course on performance on the Scholastic Aptitude Test (SAT). She contacts high school students who have and who have not taken the course and compares their scores on the SAT exam. She finds that the students who have taken the course score significantly higher on the test. Which of the following represents the principal threat to the internal validity of this experiment?
Selection
The different situations that create the independent variable in an experimental research design (for instance, providing participants with either a *high* or a *low* dose of a drug) are known as which of the following?
The *levels* of the independent variable
Which of the following is true of an internal analysis?
The *manipulation* check is *correlated* with the *dependent* measure.
Which of the following statements is correct? (marginal)
The *marginal* means show the *main effects*, whereas the cell means show the simple effects.
Which one of the following would NOT be found on the path diagram of a multiple *regression* analysis?
The *regression* line
Which of the following represents the best interpretation of the beta weights in multiple regression?
The beta weights *control* for the effects of the other predictor variables.
Which part of the correlation matrix is usually included in the research report?
The upper triangle
Which of the following is NOT true about meta-analyses?
They are *less objective* than narrative literature reviews.
Which of the following best describes *manipulation* checks?
They are *usually given* after all of the dependent variables have been collected.
Which of the following is NOT true regarding *field* experiments?
They are always more *externally valid* than lab experiments.
Which of the following is true about *correlational* research designs?
They provide no evidence about the causal *relation*ships among the measured variables.
Why would a researcher most likely use a "blind" experimenter?
To *decrease experimental bias*
Eisenberger, Lieberman and Williams (2003) studied the extent to which being excluded by others created brain responses similar to those created by physical pain. The participants, who were all studied while they were in an fMRI scanner, first experienced a baseline session, then a session in which they were included in a game, and then a session in which they were excluded in a game. Which of the following designs was used?
Within participants
A variable other than the independent variable that is found to *vary* systematically among the conditions is known as
a *confounding* variable.
A potential problem with random *assignment* to conditions is that
unless blocked random *assignment* is used, confounding can still occur.
Which of the following is a likely result of *restriction* of range?
coefficient is *reduced.*
The goal of *random* assignment to conditions is to
ensure that the participants in the different conditions of the experiment are, *on average*, equivalent before the experiment begins.
A threat to internal validity that refers to the fact that the researcher knows the research hypothesis and thus causes an invalid confirmation of the research hypothesis is known as
experimenter bias.
The idea of generalization is most closely related to
external validity
The major issue underlying external validity is that of
generalization
Which of the following sets of the sequential assignment of participants to four conditions represents the use of blocked random assignment to conditions?
*1,3*,2,4,4,2,1,3
How many orders of conditions would be needed to counterbalance an experiment that had *two* conditions?
*2*
Which of the following is an example of a single-group research design?
*A group of patients who are spending time* in a mental health facility are interviewed.
Which of the following are threats to internal validity of the single-group before-after design?
*All of the above are threats* in this design.
Which of the following three characteristics are used in experiments to make inferences about the *causal* relationship between the independent and the dependent variable?
*Association*, temporal priority, and control of common-causal variables
A student who is participating in a long-term research project decides to *stop participating* and does not inform the experimenter of his decision. This is an example of which of the following threats to validity?
*Attrition*
Which of the following procedures can be used to help ensure standardization of conditions?
*Automated experiments* and the experimental script
Jarrod hypothesizes that getting plenty of sleep leads to lower *blood pressure*, and that lower blood pressure reduces stress. In this hypothesis, which of the following is the mediating variable?
*Blood pressure*
Which of the following statements is true? (correlational)
*Both* the predictor and the outcome variables are measured in correlational research.
Which of the following is NOT used to help make inferences about *causal*ity?
*Causal* correlation
Buchanan, Tranel, and Kirschbaum (2009) *compared three groups* of participants, one of which had damage to the hippocampus, one that had similar brain damage but not to the hippocampus, and a control group of participants. The researchers assessed responses to the three groups both before and after they were exposed to stress. The hippocampus group showed less change in cortisol response to stress than the other two groups. What research design was used in this research?
*Comparison group* before-after.
Which of the following types of validity refers to the extent to which the measured variables actually measure the conceptual variables?
*Construct* validity
Which of the following refers to a *false or misleading statement* given by the experimenter about what is being studied that is used to reduce the possibility of demand characteristics?
*Cover story*
A study of parenting styles across generations samples groups of participants at ages *five*, *fifteen*, and *twenty* at *the same time*. Which of the following terms best describes this design?
*Cross-sectional*
*Cover* stories are most likely to be used to prevent which of the following?
*Demand* characteristics
An experiment that is conducted in situations that are similar to *everyday life experiences* is said to have which of the following?
*Ecological* validity
The degree to which relationships among conceptual variables can be demonstrated in a *wide variety of people* and using a wide variety of manipulated or measured variables is known as which of the following?
*External* validity
Which of the following represents a difference between an ANOVA summary table from a *f*actorial design and an ANOVA summary table from a one-way experimental design?
*F-values* are not labeled as "within groups" in a factorial design.
Which of the following represents an example of regression to the mean?
*Furniture* from famous people tends to sell at similar prices to those from non-famous people.
Which of the following types of validity is threatened when *self presentation* occurs?
*Internal* validity
Which of the following is true about a significant interaction?
*It only gives support* for the research hypothesis if the observed means are also in the expected direction.
Which of the following could not represent a participant variable?
*Manipulated mood* state
In a single-group before-after research design, *changes that occur in the participants over time* but which are not related to the independent variable are known as which of the following?
*Maturation*
In research that is conducted over a period of time, the participants naturally get *older*. If the researcher is not interested in the effects of age on the dependent variable, then the changes in age pose which of the following threats to the validity of the research?
*Maturation*
A correlational study with *two predictor* variables and *one outcome* variable will be analyzed using which of the following statistics?
*Multiple* regression
Wisneski, Lytle, and Skitka (2009) assessed the associations among measured *religiosity*, measured *moral conviction*, and measured *trust in authority*. The researchers used an analysis that controlled for the influence of each of the predictor variables on the outcome measure. Which of the following statistical analysis did they use?
*Multiple* regression
A scientist who decides to compare condition means *after* the experiment has already been run will need to use which of the following?
*Post* hoc comparisons
When specific comparisons have *not been planned ahead of time*, increases in experimentwise alpha that occur when many comparisons are made can be reduced by using which of the following?
*Post* hoc comparisons
In behavioral research, the *expected* causal relationship between the independent and the dependent variable is assumed to be which of the following?
*Probabilistic*
Which of the following would NOT be useful in reducing extraneous variables?
*Providing different information* to each participant
Which of the following is NOT a method by which experimenters control for extraneous variables?
*Random assignment* to conditions
Which of the following is used to be certain that the participants in the different conditions of a research design are, *on average*, equivalent before the experiment begins?
*Random assignment* to conditions
Consider a 2 x 2 experiment design in which both study time and sleep time have been manipulated and learning has been measured. If a main effect of *sleep time* is found, this can be interpreted to mean which of the following?
*Sleep time* affects learning, controlling for study time.
Which of the following statements is true? (control)
*Some* experiments do not have a control condition.
The possibility of making a *Type 2 error* in an experiment would most likely be reduced by using which of the following procedures?
*Standard*izing conditions
Incorrect conclusions about research that occur because a *Type 1 or Type 2 error* has been made lead to which of the following?
*Statistical conclusion* invalidity
In a three-way factorial design, the three-way interaction can be interpreted to mean which of the following?
*Two* of the *two-way* interactions have different patterns of means.
In order to confirm a research hypothesis in a *factor*ial design there must be both
*a significant interaction* and a predicted pattern of observed means.
It is most accurate to say that
*although they will generally* have some threats to internal validity, quasi-experiments can sometimes be used to draw inferences about the causal relationships among variables.
If a researcher conducted a meta-*analysis* on the effects of performance enhancing drugs within professional sports, she would be most likely to
*analyze the data* on this topic from many prior studies.
In a *matched* group research design, the participants are
*assigned to conditions* on the *basis of information collected* prior to the experiment.
A researcher is unable to retest some of his participants in a before-after design because they have *dropped out* of the program. This is an example of a(n) ________threat.
*attrition*
A limitation of meta-analyses is that
*because only significant studies* are published, a relationship may be overestimated.
An advantage of a single-participant design is that
*behavior changes can be observed* over time.
In a *factor*ial design, the researcher may find
*both main effects* and interactions.
A replication that investigates the relationship between the same *conceptual* variables that were studied in previous research but tests the hypothesis using different operational definitions of the independent variable and/or the dependent variable is known as a(n)
*conceptual* replication.
Amalia conducted a study of how the media affects political affiliation. She showed participants a five-minute clip from CNN and then asked them to fill out a political ideology survey. Edward later conducted a similar study but had participants *read an article* from the Washington Post *instead* of *watching a TV broadcast*. Edward's study is an example of a
*conceptual* replication.
Complex comparisons are usually made using
*contrast* tests.
Within repeated-measure designs, experimenters may arrange the order of conditions so that each condition has an *equal chance of occurring* after each other condition. This method is called
*counterbalancing*
Tova is participating in a study involving media awareness and conformity. After viewing a short film she fills out a conformity questionnaire. As Tova fills out the questionnaire, the questions lead her to guess that the research hypothesis is that the film will *increase conformity*. In this case the questions serve as a
*demand* characteristics.
The primary goal of replication is to
*determine the extent* to which an observed relationship generalizes across different tests of the research hypothesis.
A contingency table is a table that
*displays* the number of individuals
In a factorial design in which the factors are crossed,
*each level of one factor* occurs with each level of the other factor.
The extent to which the *experimental* manipulation involves the participants in the research is called
*experimental realism*.
If a research finding was found to hold up only in one sample of individuals or *only in one laboratory*, the research would be said to have low
*external* validity
In comparison to the single-group before-after design, the comparison- group before-after design
*helps control for threats* caused by history.
A study involving the effect of physical activity on self-esteem has a p- value of .001 and has correctly *ruled out all obvious alternative hypotheses*. However, this study *cannot be applied to any real-life settings*. Thus it can be said to have __________.
*high internal* validity, *low external* validity
Threats to internal validity that occur due to the potential influence of changes in the social climate during the course of a study are called
*history threats.*
When the experimental manipulation creates the *hoped-for* changes in the conceptual variable, it is said to have
*impact*
In comparison to lab experiments, *field experiments* have the *advantage* of
*increased ecological* validity.
The *absence* of confounding variables assures
*internal* validity.
A *limitation* of program evaluation research is its
*lack of random assignment* to conditions.
A research design in which the same individuals are measured more than once over a time period *long* enough that changes in the variables could occur is known as a
*long*itudinal design.
In comparison to laboratory studies, field experiments are likely to have
*lower internal* and *higher external* validity.
In a study *manipulating* levels of aggression, participants are asked to complete a Likert Scale indicating their current state of aggression. This measure is given after the dependent measure has been assessed. This measure will likely serve as a
*manipulation* check
When participants are *measured* on the variable of interest *before* the experiment begins and are then *assigned* to conditions *on the basis of their score*, the design is called a
*matched-group* design.
Toby is conducting an experiment in which he predicts an interaction between appearance and partner sex (same or different). *One-half* of the participants complete the study dressed in casual clothes, and the *other half* complete the study dressed in formal attire. All participants are observed in interaction with a same-sex partner and then with an opposite-sex partner. This is an example of a ___________factorial design.
*mixed*
A variable that produces an *interaction* of the relationship between two other variables such that the relationship between them is different at different levels of that variable is a
*moderator* variable.
In a one-way experiment testing the influence of *mood* states on performance, the manipulation check will be a measure of
*mood*
Factorial experimental *designs* can have
*more than* one independent variable.
Wisneski, Lytle, and Skitka (2009) correlated measured religiosity and measured *moral* conviction with measured trust in authority. The researchers found that highly relious people showed _____ trust in authority whereas the highly moral people showed ____ trust in authoiry.
*more*; less
The technique of predicting a single outcome variable using a *number of predictor variables* is known as
*multiple* regression.
Researchers who *do not know* the research hypothesis are known as
*naive* experimenters.
An experimenter conducts a study in which participants are randomly assigned to study in either a *quiet condition* or a *noisy condition* before their learning is tested. This research design has
*one independent* variable and *two levels*.
Experimental designs are called *one-way* designs if they have
*one* independent variable.
In an experiment studying the effect of attraction on sexual behavior, the *independent variable of attraction* is manipulated with two conditions: *high attractiveness* and *low attractiveness*. This design has ____________ independent variable with _________level(s).
*one, two*
A priori comparisons, comparing only means for which specific differences are *predicted* by the research hypothesis, are also called
*planned* comparisons.
Research designed to study intervention programs with the goal of determining whether the *programs* are effective in helping the people who make use of them is called
*program evaluation* research.
In comparison to one-way experiments, factorial experiments
*provide information* about interactions between and among variables.
Extraneous variables
*reduce* power and *increase* the likelihood of a Type 2 error.
In a multiple *regression* analysis, the statistics that indicate the relationship between each of the predictor variables and the outcome variable are known as
*regression* coefficients.
A collection of experiments in which a topic of interest is systematically *studied* through conceptual and constructive replications over time is known as a
*research* program.
A document that discusses the research in a given area with the goals of *summarizing* existing findings, making inferences about the conditions under which relationships do and do not occur, linking research to findings in other areas, and suggesting further research is called a
*review* paper.
Exact replications are most likely to be *conducted* because the *researcher* wants to
*see if an effect* that has been found in one *laboratory* or by one *researcher* can be found.
Regression to the mean causes interpretive difficulties when
*there is not initial equivalence* between the groups.
Researchers in the behavioral sciences may be content to use college students as research participants because
*they can be assumed* to have the same characteristics as any other sample of individuals.
A 2 x 2 x 2 factorial design has (main effects)
*three* main effects, *three* two-way interactions, and *one* three-way interaction.
In a *two*-way factorial experiment design, ________ will be manipulated and _______ will be measured
*two* independent variables; *one* dependent variable
A technique in which participants are told that they will be participating in *two separate experiments*, but in reality the experimental manipulation is presented in the first experiment and the dependent measure is collected in the second experiment, is called the
*unrelated*-experiments technique.
Manipulation checks are
*useful for interpreting* the results when there is no significant relationship found between the independent and dependent variables.
A mixed *factor*ial design is one that
*uses both* between-participant and repeated-measures factors.
Daniel is testing the effects of watching cartoons on children's play. He has one group watch *violent* cartoons and another one watch non-violent cartoons. He then measures the children's aggressive play behaviors. What are the levels of Daniel's independent variable?
*violent* and non-violent cartoons
A participant replication should be conducted
*when there is a specific reason to believe* that a result will not generalize.
If the correlation between variables x and y is r = .50, then the *coefficient of determination* is
.25
If the correlation between variables x and y is r = .50, then the *effect size* is
.50
In a correlational research design, Jonas finds that r = .82. Jonas's data demonstrates which of the following?
A positive linear relationship
The term experiment refers to which of the following? (manipulated)
A research design in which the *independent* variable is *manipulated*.
Which of the following provides a visual image of the relationship between two quantitative variables?
A scatterplot
Buchanan, Tranel, and Kirchbaum (2009) studied three groups of participants, one of which had damage to the hippocampus, one that had similar brain damage but not to the hippocampus, and a control group of participants. The researchers assessed responses to the three groups both before and after they were exposed to stress. The hippocampus group showed a decrease in cortisol response to stress, whereas the other two groups shoed increases. Which of the following statistical tests was used to test the hypothesis?
An interaction.
A researcher is studying the relationship between a person's gender and his or her preference for three new premium television channels. Each participant is asked to choose the one channel that he or she prefers. The researcher will use which of the following statistical tests?
Chi square
Which of the following serve to make meta-analysis objective?
Inclusion criteria and *analysis of the effect size statistic*
Which of the following is an advantage of a *two-way factorial* design over a one-way experimental design?
It *allows studying* the effects of *more than one independent variable* simultaneously.
Which of the following statements is true about interpreting a 2 x 2 factorial experiment?
It is misleading to interpret the *main effects* when there is a significant interaction.
Which of the following is true?
It is useful to conduct constructive replications because they provide information about *the exact circumstances* in which an effect does or does not occur.
Which of the following procedures would be likely to reduce experimenter bias?
Keeping experimenters *"blind to condition"*
A researcher studies the cognitive development of a single group of children by measuring them in kindergarten, fourth grade, and seventh grade. Which type of design is being used?
Longitudinal
Differences on the dependent measure in the different levels of any one factor, controlling for all other factors in the experiment, are known as which of the following?
Main effects
Which of the following is NOT a necessary goal for the creation of a valid experimental manipulation?
Matching of groups
Which of the following statements is true?
Only the *dependent* variable is measured in experimental research.
Stephen and Phan (2007) measured trust in feelings by having participants list either 2 or 10 situations in which they had trusted their feelings. Which of the following results did the researchers find?
Participants in the high trust in feelings condition made *less generous* offers.
An experimenter wishes to know if his research procedure is going to work as he hoped it would. What procedure might he use?
Pilot testing
Which of the following refers to an artifact that occurs when participants' expectations about what effect an experimental manipulation is supposed to have influences the dependent measure independently of the actual effect of the manipulation?
Placebo effect
*Inclusion criteria* refer to which of the following?
Specific rules about which studies to *include* in a *meta-analysis*
Anita has collected a large correlation matrix in which each of several conceptual variables has been measured more than once. She wishes to assess both the reliability of her measurements and the correlations among the conceptual variables in one analysis. Which of the following statistical procedures is appropriate?
Structural Equation Analysis
When conducting statistical analyses in an experimental design, in order to confirm the research hypothesis, the researcher hopes that which of the following is true?
The *between-groups variance* is greater than the within-groups variance.
Consider a 2 x 2 experiment design in which both study time and sleep time have been manipulated and learning has been measured. An interaction between *study time* and sleep time can be interpreted to mean which of the following? (Interaction)
The *effect of study time* on learning is different in the different levels of sleep time.
The F value is calculated as which of the following?
The between-groups *variance* divided by the within-groups *variance*
Which of the following statistics would be used to assess the relationship between two nominal variables?
The chi-square
Which of the following would NOT be found in the ANOVA summary table?
The condition means
Which of the following statistics serves as the dependent variable in a meta- analysis?
The effect size
The level of the independent variable in which the situation of interest was created is frequently known as which of the following?
The experimental condition
Experimental *control* refers to which of the following?
The extent to which unwanted *influences* on the dependent variable have been eliminated.
Which of the following is a potential *difficulty* of cross-cultural replications?
The findings are *difficult* to interpret.
Which of the following represents the null hypothesis about the means of the dependent variable in a one-way ANOVA with *two levels?* (=)
The mean in level A is *equal* to the mean in level B.
Which of the following is NOT a disadvantage of repeated-measures designs over between-participant designs?
The potential for *lowered* statistical power
In a multiple regression analysis, which of the following are most similar to correlation *coefficients*?
The regression *coefficients*
Consider a 2 x 2 experiment design in which both study time and sleep time have been manipulated and learning has been measured. An interaction between study time and sleep time can be interpreted to mean which of the following? (simple)
The simple effects of study time are *different* in each level of sleep time.
Which of the following statements is true? (t test)
The t test is a *special case* of the F test.
In a correlational research design, a spurious relationship is caused by
a common-causal variable.
Monique manipulates self-esteem to be either high, moderate, or low. She finds that performance is significantly better at moderate levels of self- esteem than it is at either high or low levels. Monique has found an example of
a curvilinear relationship.
When a participant's expectation about what effect an experimental manipulation is supposed to have influences the dependent measure independently of the actual effect of the manipulation, the change in the dependent measure is called
a placebo effect.
The goal of the ANOVA is to
compare the means on a dependent variable across levels of an *independent variable.*
A researcher tests the effects of a skills enhancement program by measuring the change in social adjustment of a group of handicapped children before and after they have completed the program against a change in adjustment over the same time period for an equivalent group of children who had not participated in any therapy. The children who went through the program showed significantly greater increase in social adjustment functioning over time. The research design is
comparison-group before-after.
An experimenter decided to test the effects of a new training program by comparing the motivation of twenty employees who had participated in the program against an equivalent group of twenty employees who had not gone through any training at all. The employees were measured when they had completed the program and it was found that those who had gone through the program had much greater on-the-job motivation than those who had not. It was concluded that the difference in motivation was due to the content of the training program. The research design is
comparison-group.
An experimenter is interested in the influence of taking a one-semester-long test preparation course on performance on the Scholastic Aptitude Test (SAT). She contacts high school students who have and who have not taken the course and compares their scores on the SAT exam. She finds that the students who have taken the course score significantly higher on the test. The research design is
comparison-group.
Researchers compared college students who reported they had smoked marijuana at least fifteen days a month with similar students who said they smoked five days a month at most. A day after going without marijuana, the heavy users performed significantly worse on tasks that involved sustaining and shifting attention. The research design is
comparison-group.
Researcher Franklin tests the limiting conditions of a previous finding by adding a new control condition to the design. Franklin's experiment is a(n)
constructive replication.
One difficulty of repeated-measures designs is that, due to the possibility of fatigue, *practice*, or carryover, the researcher may need to use
counter*balancing*
A design in which behavior is initially measured during a baseline period, measured again after the intervention of interest begins, and then measured once more after the intervention is removed is called a
reversal design.
Julia compares means between two conditions of her factorial experiment within one level of another factor. These comparisons are known as
simple effects.
A researcher finds that the number of times people carry an umbrella in a month is related to how fast the grass is growing. This relationship is most likely to be
spurious
Jason finds a correlation of .85 between monthly homicide rates and monthly ice cream sales. However, he believes that the correlation is caused by monthly changes in temperature. In this case, temperature represents a(n) _____ variable.
spurious
R2 is to the multiple *correlation coefficient* as the coefficient of determination is to
the Pearson *correlation coefficient.*
*Carry*over is said to have occurred in an experiment when
the effects of one level of the manipulation are *still present* when the dependent measure is assessed for another level of the independent variable.
Quasi-experimental research designs are so named because
the independent variable *involves a grouping*.
Pilot testing before conducting an experiment helps to be sure that
the independent variables will be *interpreted as expected*.
A 2 x 2 x 2 factorial design has
three factors with two levels each.
An experiment is testing the effect of violent radio broadcasts on aggressive behavior. Each participant listens to a violent broadcast and then completes a survey. The participant then listens to a non-violent radio broadcast and completes a second survey. This is an example of a
within-participants design.