• Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off
Reading...
Front

Card Range To Study

through

image

Play button

image

Play button

image

Progress

1/75

Click to flip

Use LEFT and RIGHT arrow keys to navigate between flashcards;

Use UP and DOWN arrow keys to flip the card;

H to show hint;

A reads text to speech;

75 Cards in this Set

  • Front
  • Back
cross-over interaction
When, on a graph plotting effects of variables, the lines plotting the effects of the variables cross. This means the effect of one of the variables is positive at one level of the second variable, but negative at another level of it.
mixed design
two-way design with a within- or -between subjects factor.
Explain the difference between mediation and moderation. Your response should be 2–3 sentences long. (2 points)
Moderation is an interaction between two variables to influence a third variable. It occurs when the effect of one independent variable on a dependent variable depends upon the level of a second independent variable. Mediation occurs when one independent variable has its effect on a dependent variable either partially or entirely through its effect on a second independent variable.
Explain what is meant by “unconfounded comparisons.” Your response should be 2–3 sentences long. (2 points)
An unconfounded comparison is one that compares cell means that differ along only one factor. This type of comparison compares cells from either from only the same column or the same row.
A social scientist is interested in how race and geographical location influence membership in the Libertarian party. What statistical test would this scientist use to evaluate these relationships? Your response should be 1–2 sentences long. (1 point)
She would use a two-way chi square.
When we combine all the levels of one factor with all the levels of the other factor in a two-way design, we have a
complete factorial design.
A 3 × 3 design is a design with
two factors. Each factor has three levels.
If the interaction between two independent variables is different, depending on the level of a third independent variable, a _____ is said to be present
three-way interaction
If the dependent variable in a two-way between-subjects design is measured on an ordinal scale, the appropriate statistical analysis is a(n)
two-way ANOVA, whereas a(n) two-way chi square is appropriate if the dependent variable is measured on a nominal scale.
Statistics based on one dependent variable are known as
univariate statistics
Statistics based on more than one dependent variable are known as
multivariate
The proportion of variance accounted for in the dependent variable is one measure of
effect size, or an indication of how dramatically an independent variable influences a dependent variable.
Y′
'Y prime/hat': when using a regression line and looking up a value on the X axis, it's the predicted value on the Y axis.
standard error of the estimate
Describes the 'average' prediction error when using the linear regression procedure.
coefficient of determination
The square of r (the coefficient): finds the variance of the Y variable that can be attributed to the X variable.
test–retest reliability
Participants, when tested repeatedly, produce same scores.
double-barreled question
Questions with more than one component.
Explain what the observed correlation of .60 means about the relation between diet soda consumption and weight gain. Your response should be 1–3 sentences long. (1 point)
Firstly, it doesn't mean that diet soda consumption plays a direct causal role in weight gain. It does, however, suggest that there is a relationship beyond moderate strength between diet soda consumption and weight gain. This might mean that soda consumption affects weight gain, weight gain affects soda consumption, or that some third variable affects both.
Based on this correlation, can we reasonably conclude that diet soda consumption causes increases in body weight? (½ point)
Hell no. It might be the cause that F (say, weight gain) precedes E (say, diet soda consumption), or that some third variable precedes both (say, stress—if diet soda consumption is your answer to stress).
A(n) _______ is a measure of association between two variables measured with ordinal scales, whereas a(n) ________ is a measure of association between two variables measured with interval or ratio scales.
Spearman correlation coefficient, Pearson correlation coefficient
______ research is conducted using data collected after an event has occurred. ______ is an example of this type of research.
Ex post facto, Archival research
A researcher interested in studying the relationship between family history of anxiety disorders and smoking is interested in examining these variables across multiple generations in order to control for any effects that could be due to the eras in which people were raised. She decides to collect data from three groups of people. Group A includes people born in 1950, Group B includes people born in 1970, and Group C includes people born in 1990. This is an example of a ______.
cross-sectional design
Suppose I am interested in studying the relationship between amount of credit card debt and optimism. I hypothesize that people with more credit card debt will report less optimism. Because I can’t randomize participants to have differing amounts of debt, I decide to study people who fit in one of three groups. Group A will have $0 to $1,000 in debt, Group B will have $5,000 to $10,000 in debt, and Group C will have $15,000 or more in debt. These amounts of debt represent three ______ of the independent variable.
manipulations (conditions)
A group of researchers is interested in the relation between caffeine consumption and typing speed. First, they bring in a group of participants and measure their baseline typing speed. Then, they administer a standard dose of caffeine and measure typing speed a second time. The researchers notice that the two fastest typists at baseline typed a little more slowly after consuming the caffeine, whereas the two slowest typists typed a little more quickly after consuming caffeine. This probably is not due to an effect of the caffeine; it is likely due to ______.
regression of the mean
A(n) ______ combines elements of the longitudinal design and the cross-sectional design, and it follows individuals from multiple age groups over time.
cohort design
What is an observational study? Describe at least two advantages of observational research and two disadvantages of observational research. Your response should be 3–5 sentences long. (3 points)
Observational studies are a form of descriptive research that observes participants in a less obtrusive way. There are three forms of observational studies: naturalistic observation, systematic naturalistic observation, and participant observation. The first observes participants in an unstructured and unsystematic way, the second observes participants in an unobtrustive but in a more structured and systematic way, and the third is when a researcher is an active member of the group of participants being observed. Excellent!
Imagine you are a researcher interested in whether physical exercise affects the age of onset of Alzheimer’s disease. Describe a one-group pretest posttest design to examine this relation. Be sure to describe what sort of sample you would use and how you would conduct the study. Your response should be 3–5 sentences long. (3 points)
One-group pretest posttest designs measure a target quality before and after an event. In this case, the researchers are measuring the effect of physical exercise on the onset of Alzheimer’s. To do this, they would need to find a sample of older people, test (i.e., 'measure') them for Alzheimer’s, and then start them on an exercise regimen. After the exercise regimen is implemented and had time to take effect (they would have to get more rigorous in what that would look like), the researchers would them re-measure the participants for Alzheimer’s. A non-equivalent control would be a group of older people who haven't been placed on the physical exercise regimen. This would give the researchers a much better idea of physical exercise's affect on the onset of Alzheimer’s.
Define the terms “longitudinal study” and “cross-sectional study.” Briefly discuss the advantages and disadvantages of both of these types of studies. Make sure to include at least two advantages and disadvantages of each study type. Your response should be 3–5 sentences long. (3 points)
A longitudinal study measures the effect the passage of time has on some variable; the same participants are repeatedly measured over a period of time. The benefits of longitudinal studies are that they provide strong evidence for causation (based on an established temporal sequence of events), and that the participants can act as their own controls across time. The disadvantages are the large expense of putting on the study, and the burden to the participants in it (making it difficult for researchers to keep the participants in the study). A cross-sectional design studies multiple participants of different age groups simultaneously. The benefits of a cross-sectional design are that they are much cheaper than longitudinal studies, and they are much cheaper; disadvantages, however, are that they are more likely to have confounding variables (e.g., cohort effects), and that a particular testing procedure might not be appropriate for different age groups.
Hawthorne effect
The bias in a person's behavior that results from any special interest or treatment shown by, or given by, a researcher.
snowball sampling
A form of non-probability sampling that is generally used with incredibly small or difficult to reach population. A researcher finds out one or two individuals from the population and finds others through them (either by actually finding others, or by gathering data from those individuals on the others).
pilot study
A smaller version of a larger study that is designed to get an estimate of the effect size a researcher could expect to find in the population of interest.
baseline
The level of performance on the dependent variable when the independent variable isn't present.
What is the difference between stratified random sampling and quota sampling? Your response should be 1–3 sentences long. (2 points)
Stratified random sampling first divides the population in terms of a characteristic of interest, and then randomly samples from within each division. Quota sampling is when quotas are set for a particular characteristic in the population, which has been divided into groups based on that characteristic, and then the researchers sample from each group until a quota is reached. The difference is that quota sampling stops when the researchers reach their desired quota; stratified sampling doesn't. Right and this makes stratified random sampling a form of probability sampling while quota sampling is a form of non-probability sampling.
What is one argument for the use of small N designs? Your response should be 1–3 sentences long. (2 points)
Using smaller samples is incredibly useful when the equipment required for the study is extraordinarily expensive to operate per use, or when it is being rented at an extraordinary price.
What would cause a researcher to conclude that genetic effects and not environmental effects were present in an adoption study? Your response should be 1–3 sentences long. (2 points)
When the individual is taken out of her original family environment, and still expresses the variable of interest (e.g., depression). Specifically, when the adopted child shares the characteristic of interest with the biological parents and not the adoptive parents.
Provide two examples of vulnerable populations. Your response should be 2–3 sentences long. (2 points)
One example is a mentally challenged person, who may not be able to provide consent (though can provide assent). Another example is of a baby, who cannot freely choose to participate in a study.
What is a multiple baseline design, and what is the reasoning behind the use of this design? Your response should be 3–4 sentences long. (3 points)
A multiple baseline design shows the effect of a variable by demonstrating a change in a target behavior only when the treatment is present. Researchers examine more than one baseline, in this design. A baseline is the level of performance on the dependent variable when the independent variable isn't present; then, it's compared to performance when it is present. By examining more than one baseline, researchers are able to eliminate potential confoundings by demonstrating that the target behavior changes only when the treatment is introduced.
In order to determine whether a trait is familial, researchers first conduct _____. Then, in order to determine to what extent the familiality is due to genetic influences, ______ or ______ are conducted.
family studies, adoption studies, twin studies
In a ____, the researcher examines several independent variables and their interactions. In a ____, all levels of one factor are combined with all levels of the other factor
Multifactor experiment, complete factorial design
In a ____ , there is no matching or repeated measures of participants in conditions of either of the two independent variables. In a ____, matched groups or the same repeatedly measured participants are tested in all conditions of two independent variables. In a ____, one within-subjects and one between-subjects factor is examined
Two way between subjects ANOVA, two way within subjects ANOVA, two way mixed design
In any two way ANOVA, an ___ is computed for the ____ of each favtor and for the ____
F, main effect, interaction
A ___ of a factor is the effect that changing the levels of that factor has on dependent scores, while collapsing across--ignoring--all other factors in the study
Main effect
An ____ is the influence that the combination of levels from the factors has on the dependent scores. In a ___, the relationship between one factor and the dependent scores is different for and depends on each level of the other factor. When the cell means of a significant interaction are graphed, the lines are ___.
Interaction effect, two way interaction, not parallel.
A ____ is the effect of one factor at one level of a second factor within the interaction. ___ comparisons in an interaction should involve only ____ - comparisons of cell means that differ along only one factor
simple main effect, post hot, unconfounded comparisons
Because an interaction contradicts the overall pattern suggested by the main effects, the interpretation of a study usually focuses on the ____
significant interaction
A ___ produces three main effects, three two way interactions, and a three way interaction. In a ___, the two way interaction between two factors changes as the levels of the third factor change
Three way design, three way interaction
The ___ is used when counting the frequency with which participants fall into the categories of two variables. If X^2 is significant, the frequency of participants in the different categories of one variable depends on their ctegory mebership along the other variable
two way chi square procedure
For any significant main effect or interaction, the ___ indicates how dramatically the variable influences the dependent scores
effect size
The ___ in dependent scores indicates how much all of the scores in the study differ, and it equals the error variance plus the systematic variance. ____ refers to the differences in scores that occur with, or are associated with, changes in the conditions
Total variance, systematic variance
Calculating effect size as the ___ indicates the proportion of total variance in dependent scores that is systematic variance. It reflects the proportional impreovement in predictions that is achieved when we use a relationship to predict scores, compared with if we do not use the relationship.
Proportion of variance accounted for
The effect size is calcuted as 1) the ____ when performing a t-test, or 2) ____ for each significant factor or interaction effect when performing ANOVA
squared point-biserial correlation coefficient, eta squared
The effect size in a significant 2 x2 chi square is the ___. With two way design that is not a 2 x 2, effect size is the ____.
Squared phi coefficient, squared contingency coefficient
____ are inferential statistical procedures for a study involving more than one dependent variable
Multivariate statistics
____ involves statistical procedures for combining, testing, and describing the results from different studies
Meta-analysis
In a ____, the researcher measures participants' scores on two or more variables to ovbserve a relationship. Causality cannot be inferred because participants are not randomly assigned to the levels of a variable, and which variable occured first is unknown.
Correlational design
The ___ is computed when the X and Y scores are normally distrib uted and measured using an intervial or ratio scale. The ___ is computed when both variables are measured using an ordinal scale.
Pearson correlational coefficient, Spearman correlation coefficient
A ___ will range between +1 and -1. The sign indicates the type of relationship. The smaller the absolute value of the coefficient, the greater the variability in the Ys at each X, the greater the vertical width of the scatterplot, and the less accurately Y scores can be predicted from X scores.
Linear correlation coefficient
____ is used for predicting scores on one variable based on the linear relationship with another variable. The ____ is used to predict a Y score, called ____, for a particular X. The ____ summarizes a linear relationship, with the values of Y' falling on the line.
Linear regression, linear regression equation, Y', linear regression line
The ____ communicates the "average" amount that the Y scores in a sample differ from the corresponding predicted Y scores. The ____, called the ____, is the proportion of variance in Y scores that is accounted for by the relationship with X.
standard error of the estimate, squared correlation coefficient, coefficient of determination
The size and power of a correlation coefficient is increased by by ____, _____, ____, and ___.
minimizing error variance, avoiding restricted range, testing a large N, and using parametric, lineaer procedures when appropriate
____ indicates that participants tend to obtain the same score when repeatedly tested. ____ indicates that their scores on some trials ocnsistently match their score son other trails.
Test-retest reliability, split-half reliability
With ____ a procedure appears to be valid. With ____ a procedure is correlated with other procedures that are accepted as valid. With ____ a procedure is not correlated with procedures that measure other things.
Face validity, convergent validity, discriminant validity
With ____ the scores from a procedure correlate with an observable behavior. With ____ a procedure correlates with a present bvehavior. With ____ a procedure accurately predictes a behavior.
Criterion validity, concurrent validity, predictive validity
The goal of psychological questions is to _____ and ____ discriminate between individuals on the variable of interest
reliabiliy and validly
With ___ participants select from alternative provided by the resarcher. With ___ participants determine the alternative to choose from.
Closed-ended questions, open-ended questions
With ___ participants create a description or interpretation of an ambiguous stimulus, thus projecting their hidden feelings or attributes
projective tests
___ is the procedure used to score open ended questions by looking for specific words or themes
Content analysis
In a ____, participants are asked specific, predetermined questions. In an ____, the questions are less rigidly predetermined
Structured interview, unstructured interview
We should avoid asking _____, which have more than one component, ____, which are biased so that there is only one obvious response; ____, which are global truisms to which everyone responds in the same way; and questions with ____.
double-barreled questions, leading questions, Barnum statements, undefined terms
A ____ is the number and type of choices provided for each question
response scale
____ are general questions that lead to more specific follow up questions. ____ are general questions for determining whether more detailed follow up questions should be asked
Funnel questions, filter questions
____ of a questionnaire contain differently worded questions that measure the same variables. They are especially imporant when testing in a repeated measures design.
Alternate forms
___ and ___ predict scores on one Y variable using multiple X variables. ____ categorizes subjects along a qualitative criterion variable using several predictor variables. ____ determines the correlation between two variables while keeping the influence of other variables constant. ____ indentifies a common underlying construct by determining which questions are correlated.
Multiple regression, multiple correlation. Discriminant analysis, partial correlation, factor analysis
A ____ is an experiment conducted in a real life setting in order to increase external validity.
Field experiment
With ____, every potential participant has an equal likelihood of being selected. With ____, participants are selected in a random fashion. With ____, every nth individual is selected from the population. With ____, participants are selected from within each important subgroup in proportion to the population. With ___, specific groups are randomly selected and then all members of each group are observed.
Probability sampling tec hniques, simple random sampling, systematic random sampling, stratified random sampling, cluster sampling