Final psy 3213C
Best Illustrates the concept of covariance:
"Students who score below the class mean on the midterm also score below mean on the final exam."
T OR F: The assumptions for a two-way between-subjects ANOVA apply only to main effect tests, and not the interaction test.
FALSE
T OR F: When the same participants are observed in each cell or group, we use the two-way between subjects ANOVA to analyze the data.
FALSE
T or F: Matching on more than one extraneous variable gets easier as the number of experimental groups increases
FALSE
The effect of adding fluoride to a community's water supply was tested by comparing annual records of cavities per patient in local dentists' offices for six years before and four years after the fluoride was added. This is an example of which type of research?
Interrupted Time series
A ___ is the effect of one IV on the DV that changes at different levels of another IV, while a ___ is the effect of an IV on the dependent variable within a factorial research design.
Interaction Effect; Main Effect
The challenges associated with generalizing particular research findings in quasi-experiments relate to issues of:
Internal Validity
T or F: Quasi-experiments examine relationships between the manipulated variable(IV) and the DV while experiments can examine causation between these variables
TRUE
The main purpose of conducting Post-Hoc pairwise comparisons:
Test research hypothesis
Most likely to reduce the internal validity of a non-equivalent groups design in which both groups are exposed to pre- and post-test (a "difference-in-differences" study)?
Testing
Dave plans to follow children in experimental math classes and children in regular math classes. He wants to test the children four times during the school year but is worried that asking children the same questions multiple times will bias his results. What kind of confound is Dave worried about?
Testing Confound
Assumptions for computing a two-way between-subjects ANOVA:
The population being sampled from is normally distributed. One observation has no effect on the likelihood of another observation. There is homogeneity of variance.
Common weakness of experimental research?
The presence of threats to internal validity posed by confounds.
The concept of statistical power is MOST directly related to the concept of ___.
Type II Error
Within a college course, a __ error occurs if an instructor decides a student ___ when he/she in fact ___.
Type II; has not cheated; did cheat.
Random assignment to a group is a requirement of:
A True Experiment
When r= +.60, what z-score in X leads us to predict that Y= mean of Y?
0
How could a researcher interpret an eta squared value of .35?
35% of the variance in one variable is explained by another variable.
In a 3 x 5 factorial ANOVA, there will be __ marginal means for the second factor.
5
The Hawthorne Effect
A phenomenon in which participants change their behavior simply because they are in a study and have the attention of researchers.
A within-Subjects research design is:
A research design in which each participant appears in all levels or categories of the IV.
The scatter-plot for a perfect correlation would show:
A straight line.
Suppose a researcher wishes to investigate the effect of biofeedback on performance under conditions of time pressure (treatment group) vs. no time pressure (comparison group). The researcher decides to control for wide variation in the biological responses used for the biofeedback by exposing participants in the no time pressure condition to the biofeedback pattern received by a participant in the time pressure condition. Which of the following procedures does this describe?
A yoked control design
Dr. Louis, a researcher investigating a group of deaf infants, reports that infants' speech production significantly improved following cochlear implantation (p < .01). However, he reports an effect size of .02. How would we interpret his results?
Accept his findings; standards for what constitutes as a meaningful effect size vary by research discipline.
The ___ hypothesis establishes whether a one-tailed or two-tailed significance test will be used.
Alternative
"Any variable that is not of immediate interest to a researcher but which may pose a threat to validity because it compromises the interpretation of research findings. This is usually because it obscures the measurement of processes of interest."
An Extraneous Variable
Correlation studies are often used when:
An experimental study would be unethical. It is not possible to manipulate the variable(s) of interest. As a pilot study to examine trends before an experiment in conducted.
Researchers must be aware of and address the possibility of _____, where participants drop out from the experiment.
Attrition
Mark is conducting a longitudinal study of children in a small village in Dominica. He has been taking measurements on the same children for the last five years. Though none of the children have died, quite a few have moved from the village to the nearest city. Mark is worried that the loss of participants may create what kind of confound?
Attrition/ Mortality Confound
A researcher hypothesizes a relationship between police satisfaction and one's level of fear. If she reports "F (3, 30) = 4.25, p <.05", her research hypothesis has ______ because the null hypothesis was _____.
Been supported; Rejected
"Systematic change to an independent variable where different participants are exposed to different levels of that variable by the experimenter
Between-Subjects Manipulation
The issue of unintentional experimenter bias can often be attended to quite straightforwardly - simply by using an experimenter who is___to the condition in which participants are placed.
Blind
Most conservative out of the three common post-hoc Analysis
Bonferroni Correction
The Impact of the treatment or measurement lasts longer than the time between different conditions
Carryover
Which statistic is an estimate of the magnitude of the difference between the means of two groups measured in standard deviation units?
Cohen's d
A study tested whether students who were required to do community service in college were more or less likely to continue volunteering after graduation than those who were not required. Since this was not a randomized experiment, the students who were not required to do service are called the ________.
Comparison group
A ____ Hypothesis is a statement about the relationship between theoretical concepts. They cannot be measured until they are operationalized.
Conceptual Hypothesis
Assumptions of Parametric Tests:
Data collected on these variables from samples of the population are normally distributed. Data from a sample is being used to estimate the population parameter. They assume that the distribution of scores for a variable is normally distributed in the larger population. NOT: they do not require variable to be normally distributed.
Appropriate statistical test to use if your participants all experience both of two conditions?
Dependent-samples t test
A study on the effectiveness of a virtual reality system for treating anxiety disorders finds that creating conditions in the virtual world that are similar to the conditions that induce anxiety in patients tends to work best in terms of treatment outcomes. Which of the following forms of validity does this BEST reflect?
Ecological
After conducting a study looking at test performance after an exam-taking preparation course, the researcher concludes mistakenly that the learning strategies taught in the course helped students do better on a major test, when in reality students did better because they studied the relevant material. The study lacks:
External Validity
T OR F: Because of the nature of a mixed ANOVA with repeated measures on one factor design, one may always assume that condition/group variances are equal.
FALSE
T OR F: If a researcher increases the sample size in a study from 30 to 50, she is likely to decrease the power of the study.
FALSE
If a researcher wanted to examine all of the possible combinations of two or more independent variables, she would want to use a ___ research design.
Factorial
The primary Advantages of a ____ design over simpler experiments is that a researcher can be more efficient in testing the effects of multiple IV's in one experiment and can also examine the effects of the interactions of those IV's on the DV.
Factorial
T OR F: It is inappropriate to compute effect size for interactions and main effects.
False
Characteristics of Dependent-group design when compared to Independent groups design
Greater Sensitivity to changes in the measured variable. Assurance of homogeneity of variance Decreased chance of confounds. NOT:LARGER SAMPLE SIZE
If an F Ratio is ___ the critical value, I will ___ the null hypothesis and conclude that ___.
Greater than; Reject; The groups are not all equal to each other.
The mean of each level of a factor that ignore the other factor is called:
Group means
Proper way to state the alternative hypothesis for the interaction effect?
H1: An interaction effect exists.
Arnold compares the means of three groups and calculates a value of -2.01 for the F-ratio. This implies ___.
He has made an error in his calculations.
Pam is running a field experiment. She wants to know if women in her introductory psychology class change their attitudes toward gun control after being exposed to National Rifle Association brochures. She measures the women's attitudes in a pretest and gives them the literature to take home and read. Before she has a chance to measure their attitudes again, the local newspaper runs a front-page story of about a student's child who is killed while playing with a gun. What kind of confound is this likely to cause?
History Confound
The regression equation measures
How far each data point deviates from the line that most closely fits the data.
The process of determining the probability of obtaining a particular result:
Hypothesis Testing
DV AND IV:A researcher conducts a study in which participants are asked to administer electric shocks to a participant in what they believe to be a memory experiment, but which in fact is a study of obedience. In a number of different conditions the proximity of the researcher to the participant is varied to see if this has an impact on participants' willingness to administer shocks of different intensity
IV: Proximity DV: Shock intensity
Most Likely to calculate a correlation coefficient?
IV=Number of friends; DV= Hours a day spent on the internet
The main concern with lowering the probability of making a Type 1 Error is that it:
Increases the probability of making a Type II Error.
The inferential statistic used to analyze two-group designs is called the :
Independent-samples t test
A researcher decides to control for the effects of stress vulnerability on performance by matching participants in the stress condition with a participants in the control condition who have similar scores on a pre-task measure of stress. Which kind of matching procedure does this reflect?
Individual Matching
Suppose that a researcher found that as introversion (as measured on a standardized questionnaire) of participants goes up, number of social activities attended per week decreases. Which of the following relationships can explain this finding?
Introversion causes people to attend fewer social activities. Type of parenting received in childhood causes both introversion and attendance of social activities. Attending fewer social activities causes people to be introverted.
True about randomized partial counterbalancing?
It works best if you have a large number of conditions.
The Pearson correlation coefficient (r) assumes the relationship between two variables is ___.
Linear
You wish to learn whether married people with children differ from married people without children on a scale of "conservatism." You obtain scores for 50 people from each group and rank order them. Which test of significance would you use to test for a difference between the two groups on this dimension?
Mann-Whitney U Test
Natural changes that occur over time can be a threat to internal validity. This threat is called
Maturation
Researchers try to ___ between-groups variance and ___ within-groups (error) variance.
Maximize; Minimize
Used to determine the linear equation that "best fits" a set of datapoints?
Method least squares
A researcher reports a significant ANOVA as F(2, 12)=7.03, p<0.05. What is wrong with this report?
No effect size is reported.
In an interaction effect, the factor that changes the strength or direction of the relationship between a predictor and the outcome ( or the IV and DV in an experiment) is called a :
Moderator
A multi factorial design examines the effect of:
More than one independent variable on a DV.
When two variables both strongly predict a third, you expect the first two to be correlated. If you can't tell the influence of one independent variable free from the influence of the other with which it is correlated, you may be dealing with:
Multicollinearity
A study in which a researcher investigates the effect of time pressure on cognitive performance on individuals who are high vs. low in stress vulnerability is an example of ?
Multifactorial Designs
How can you maximize the chance of observing a very strong treatment effect in a quasi-experimental design?
Multiple time-series design
Researchers studied the labor market in Miami before and after the Mariel Boatlift in 1980, during which a large influx of Cubans arrived in Miami looking for jobs. The shock of the Mariel Boatlift on the local labor supply was exogenous to wages in Miami. This is an example of a(n)
Natural Experiment
You believe the farther you sit from an instructor, the less likely he or she will call on you. This is an example of a ___ relationship.
Negative
A psychologist wants to compare the effects of an improved forensic interview with the standard police interview (control group) on memory recall in abused and non-abused children. Memory for an event is pre-tested once in both groups before the interview and compared with information recalled during the interview in abused and non-abused groups. Which quasi-experimental design was used in this study?
Non Equivalent Control Group (NECG) Design
A Type II Error can occur when a researcher makes the decision to ___.
Not reject the null hypothesis.
The formula for the ___ of the t-test and F-ration concerns __ and involves group ___.
Numerator; Difference between groups; Means
The results of the chi-square test found that fear of crime depended on one's satisfaction with police. If the null hypothesis is rejected, which is the most appropriate conclusion that can be made?
One's fear of crimes is related to whether or not they are satisfied with the police.
A psychologist at a major university has an idea for a personality experiment. Before starting a major experiment the psychologist tries out a very basic version of the experiment using his introduction to psychology class. This pilot-testing of the larger study is also known as a:
Pre-Experiment
In a scatter-plot, the ___ variable is typically plotted on the x-axis, whereas the __ variable is typically plotted on the y-axis
Predictor; Criterion
Correlation
Prerequisite for Causation
An archeologist wanted to see whether his lecture increased students' knowledge about traditional food production among the highland Maya in 1000 AD. During the first 10 minutes of class, he administered a multiple choice quiz on the subject. After his lecture he re-administered the quiz. What kind of research design was the archeologist using?
Pretest-Posttest Design
A(n) ___ is a group design in which a researcher compares pre-existing or naturally occurring groups that are exposed to different levels of a variable of interest.
Quasi-Experiment
Results for Quasi-experimental design differs from that of an experiment because:
Quasi-Experiments examine relationships while experiments can examine causation
_____ Tend to have stronger generalizability, whereas ____ tend to provide stronger evidence of causation.
Quasi-Experiments, random experiments.
If a researcher conducts a dependent-groups (within-subjects) experiment in which the participants experience all levels of the IV, what type of random assignment should he use?
Random assignment to order of conditions.
Using flip of coin to assign participants to groups:
Randomized Experiment
In a study to examine the relationship of Extraversion to cognitive performance, a researcher divides participants into two groups (high vs. low on the trait) based on a cutoff score on a measure of Extraversion. This type of research uses
Regression Discontinuity
To study the effects of stress (time pressure vs. no time pressure) on cognitive performance, a researcher wants to evaluate performance for those who are high vs. low in stress vulnerability. Which of the following designs would be BEST for this purpose?
Regression-Discontinuity Design
A Type 1 Error occurs when researchers make the decision to ___ when they ___.
Reject the null hypothesis, should not.
Your value of a statistic such as the t-test is greater than the critical value for α = .05 but less than the critical value for α = .01; consequently, you ____ the null hypothesis and state the level of significance as ____.
Reject;p<.05
If alpha = .01, the null will be :
Rejected 1% of the time when it is true.
Presence of a confounding variable means you have not met which casualty:
Ruling our alternative explanations
A researcher conducts a study examining the effect of exercise on mood. She recruits participants from a health club to exercise for 30 minutes and then report their mood. She then recruits participants from a book club to sit for 30 minutes and then report their mood. Because the groups were different prior to the study, ________________ is a threat to the study's internal validity.
Selection
A psychologist is investigating the effects of a training program for computer security on the performance of Soldiers who are high vs. low in cognitive ability. Using a non-equivalent groups design with pre-post scores on a computer security test, the psychologist finds that the training program increased scores more for the high ability group compared to the low ability group. However, it is possible that those with high cognitive ability were able to learn faster than those lower in cognitive ability. Which confound does this example BEST illustrate?
Selection-Maturation Interaction
When the null hypothesis is rejected, the results of the analysis are generally described as ___
Significant
A ___ is the effect of one IV at one of the levels of another independent variable.
Simple Effect
A cohen's d Value of .25 may be considered to be a ____ effect.
Small
The average error between predicted Y values and the actual Y values is called
Standard Error of the Estimate.
More susceptible to Mortality/Attrition than Single group pre-post designs
Static Group
A researcher is interested in studying a sample population, and then applying the findings to the population. This is referred to as :
Statistical Inference
T OR F: As the homoscedasticity of a given bivariate distribution increases, the accuracy of the correlation test to evaluate the linear relationship increases.
TRUE
T OR F: For linear regression with no predictor variable, beta will equal r.
TRUE
T OR F: The Spearman correlation is a non-parametric technique.
TRUE
T OR F: The predictor variable is a known value that is used to predict the value of another variable.
TRUE
T or F: In independent designs the participants in each level of the IV are unrelated to one another.
TRUE
A not-so-brilliant chemist is convinced that he has a drug that will make people smarter. To test his hypothesis he gives participants an IQ test before taking the drug and the same IQ test after taking the drug. Participants performed better on the second IQ test. Assuming the drug did not work, what might account for the improved IQ test performance?
Testing Effects
To calculate the Confidence Interval (CI) of a mean you need to know:
The CI level, the number of people in a sample, Mean and SD.
A group of parents were given two tests that asked about their children's eating habits. The pre-test was designed as a baseline before an educational seminar on the benefits of eating rutabagas. Researchers were concerned that their post-test would be influenced by information in their pre-test, thus displaying the:
The Hawthorne Effect
A researcher computed a two-way between-subjects ANOVA and finds a significant A x B interaction and two main effects. Which significant effect should be analyzed first?
The Significant Interaction
Most common measure of effect size for the correlation coefficient:
The coefficient of determination
A common drawback to time series study designs?
The control group discovers the treatment, and thus seeks treatment themselves.
Where does the power of the dependent-groups design come from
The decrease in random error that is created by participant characteristics.
In testing the difference between two means, what is implied by the alternative hypothesis (H1)?
The difference between the two population means is not equal to 0.
Statement Inappropriate to use when interpreting significance level of a variable
The difference was found to be significant at the p <.001 Level, making it highly significant
The correlation coefficient indicates:
The extent of the relationship between two variables.
The hypothesis for the AxB interaction effect take into consideration:
The interaction effects
The effect of each IV on each DV is called
The main effect
Aspects of the relationship between main effects and interactions
The presence or absence of main effects provide no indication of whether an interaction effect is present or absent. The presence or absence of an interaction effect provides no indication as to whether a main effect is present or absent. Whether and how one interprets main effects depends on the presence or absence of interaction effects.
What distinguishes the experimental method from the quasi-experimental method?
The random assignment of participants to conditions in the experimental method.
When the manipulation of an IV inadvertently manipulates a second variable that is actually responsible for observed effects on a DV
The second variable is a threat to the study's internal validity
In multiple regression, the coefficient R can be interpreted as:
The strength of a relationship between the dependent variable and a set of independent variables.
If a researcher finds that treatment and comparison groups are not equivalent on a pre-test measure what could reduce the internal validity of the study by interacting with the pre-test variable?
The treatment
A week after participating in a workshop on driving safety, we asked a sample of 30 new drivers to report their frequency of texting while driving. We then compared their scores to the national norms for this behavior and found that t(29) = 2.95, p < .01, η2 = .08. We conclude that:
There is a significant difference between our sample and the population and the effect size is weak.
You have designed a new fitness workout that you think will help college sprinters decrease their times. You decide to test the new program on the fastest and slowest runners on several track teams. You expect that the new workout regime will decrease the times for all runners. After three months of training, you are surprised to find that the slowest sprinters have decreased their times significantly, but that the faster sprinters increased their times. What is the simplest explanation for this result?
This is probably an example of regression toward the mean.
A researcher examines how both education level and political beliefs relate to environmental values. An interaction hypothesis might be:
Those with a college degree will have stronger environmental values than those without a college degree, and this relationship will be stronger for liberals than conservatives.
Confounds that must be controlled so that a cause-effect relationship can be demonstrated
Threats to internal Validity
An experiment is conducted in which one group of participants is asked to perform math tasks while watching television, a second group of participants is asked to perform math tasks while listening to the radio, and a third group of participants is asked to perform math tasks while doing nothing else. How many Conditions and What is the IV
Three Conditions, nature of distraction is the IV
T OR F: A factor is another word for a predictor variable or IV
True
T OR F: In natural experiments, investigators cannot randomly assign individuals to groups. Therefore, selection biases are always a threat to the internal validity of a natural experiment.
True
A pharmaceutical company has asked you to test the effectiveness of their cure for the common cold. What is the most effective kind of test to determine if the drug is affects the likelihood of people getting the cold?
Two-Tailed
The appropriate analysis for a dependent-groups factorial design
Two-way Within-Subjects ANOVA
Within a marriage, a ___ error occurs if someone decides their spouse __ when in fact he/she __.
Type 1; Has been unfaithful; Has not been unfaithful.
A large sample of 10,000 marathon runners was given glucosamine supplements to see if it reduced joint pain. Significance was found at the p < 0.00001 level, but the resulting effect size was very small. As a result, the glucosamine supplement is most likely:
Unlikely to have an effect large enough to be of practical significance.
In a regression-discontinuity design the cutoff score is:
Used to assign participants to groups based on a co-variate measure.
In the regression equation, a=
Y-intercept
Dependent Variables
Whatever the investigators think might change as a result of the experimental intervention
The relationship between X and Y is non-linear. Knowing this, which of the following can you correctly conclude?
r will underestimate the strength of the relationship.
An instructor finds she is completely unable to predict student's performance in her class from her initial impression of them. What correlation best reflects the relationship between her impressions and their performance?
r= .00
Pairs of novice weight lifters from Clem's Gym have been matched for height, weight, and experience. One of each pair is assigned to one group and the other to a second group. Both groups work out for an hour and a half every second day, using the same routine. However, Group #2 receives special instructions on effective bench pressing strategies. Two months later, both groups are tested for improvement in the number of kilograms they are able to bench press. What test should be used to test for the difference in improvement between the two groups?
t-test for the difference between two dependent means.
The regression line always passes through zero when the data are in which form?
z-score
