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;
112 Cards in this Set
- Front
- Back
When the independent variable is manipulated and tested between subjects
|
Independent groups design
|
|
When the independent variable is manipulated in a between subjects design and equivalent groups are made through random assignment
|
Matched groups design
|
|
Groups are formed because of their already existing subject characteristics
|
Nonequivalent groups design or
Ex post facto |
|
When the independent variable is tested within subjects
|
Repeated-measures design
|
|
Different levels of the IV are administered to different groups of participants
Each participant has only one DV score, so they are not varied within each participant |
Between-subjects
|
|
Different levels of the IV are administered to each participant
Each participant has as many DV scores as there are levels Can vary both within and between subjects |
Within-subjects
|
|
What test do you use when you have 2 separate groups of participants and you want to compare means from these groups?
|
Independent samples t-test
|
|
What test would you use when you have one group of participants being tested 2 times and you want to compare the means at the 2 different times?
|
Dependent samples t-test
Within-subjects and or matched groups |
|
The variability of each of the sets of scores being compared ought to be similar
One of the assumptions in analyzing 2 level designs |
Homogeneity of variance
|
|
Single factor--two levels, between subject design
|
Independent Samples t-test
|
|
Single factor--two levels, within-subject design
|
Dependent samples t-test
|
|
Single factor--more than two levels, between-subject design
|
Regular One-Way ANOVA
|
|
Single factor--more than two levels, within-subject design
|
Repeated Measures One-Way ANOVA
|
|
A variable for which a number of intermediate values exist
Ex.: dosage level of a drug What type of graph is best to use? |
Continuous variables
Line graphs - estimate in-between effects |
|
A variable in which each level represents a distinct category and no intermediate points can occur
What is the appropriate graph to use? |
Discrete variable
Bar graph |
|
These members are led to believe they are receiving some treatment when in fact they aren't
Only measures the psychological effects |
Placebo control group
|
|
These are often used in research designed to assess effectiveness of some program or psychotherapy
They are experiencing the problem, but not the program designed to alleviate it They will receive treatment after |
Waiting list control groups
|
|
They experience the same events as the experimental group except for one slight difference
Each member is matched to a member of the experimental, so time spent on events is constant |
Yoked control groups
|
|
Indicate the amount of variability that occurred in each condition
Shown on bar graphs |
Error bars
|
|
This test deals with nominal categories and the differences between observed and expected
|
Chi Square
|
|
This design involves any study with more than one IV
|
Factorial Design
|
|
The term "levels" refers to what in the factor?
|
The number of levels of the independent variable
|
|
How many conditions are in a 2 x 3 design?
|
6
|
|
When the row means or column means do not equal each other
Requires combining all of the data for each of the levels of that factor |
Main effect
|
|
Occurs when the effect of one IV differs/depends on the level of another IV
|
Interaction
|
|
When there are both types of between-subjects and a within-subjects variable present
At least one must be tested between and one within subjects |
Mixed factorial design
|
|
When a between-subjects factorial include both a manipulated IV and a subject variable
They can yield an interaction between the type of person and the environment |
P x E designs
|
|
When a P x E design includes both a between-subjects and a within-subjects factor
|
mixed P x E factorial
|
|
Used to prevent confounding order effects
|
Counterbalancing
|
|
Variables that consist of some already existing attribute of a person (gender, age, trait)
Aren't really manipulated, but are of interest Unable to draw causal inferences |
Subject variables
|
|
Example of a P x E design, which measures both the nonmanipulated subject and manipulated environment
Interaction is the most interesting Ex. Females performance on math problems in groups of females or a minority |
ATI (Aptitude-Treatment Interaction) designs
|
|
There is no manipulation, so there can be no causation
|
Correlation
|
|
Both variables change in the same direction
High score on one relates to a high score on the other |
Positive correlation
|
|
Variables change in opposite directions
It is an inverse relationship |
Negative correlation
|
|
This correlation is used when both variables are interval or ratio
|
Pearson's r
|
|
This correlation is used when the variables are both ordinal
|
Spearman's rho
|
|
Provides a visual representation of the relationship shown by a correlation
|
Scatterplot
|
|
This weakens the correlation when this happens to one or both of the measured variables
|
Restricting the range
|
|
A portion of variability in one of the variables in the correlation that can be accounted for by variability in the second variable
Always positive Found by squaring Pearson's r |
Coefficient of determination r squared
|
|
Indicates the linear relationship between variables
|
Line of best fit
|
|
What are some caveats of correlations?
|
Only applicable for linear relationships between variables
May be misleading if the range is restricted |
|
Have to be careful of this problem in correlations, for causation cannot be explained
|
Third variable problem
|
|
These attempt to statistically control for the third variable
|
post facto or partial correlations
|
|
Used to predict one variable based on the other
|
Regression
|
|
What is the equation for a regression line?
|
y = mx + b
y = bx + a |
|
The variable that is being predicted in a regression (Y)
|
Criterion variable or predictor variable
|
|
The existence of the correlation by itself does not allow one to decide about the direction of causality
|
Directionality problem
|
|
Using this increases confidence about directionality in a correlation
Investigates correlations at several points in time |
Cross-lagged panel correlation
|
|
Correlating between the items from two halves of a subtest
Should be high if test is reliable |
Split-half reliability
|
|
Correlation between scores of test taken at two different times on the same measure
|
Test-retest reliability
|
|
The ability of a test to predict some future event
Tests what it is designed to measure |
Criterion validity
|
|
Involves one criterion variable and a minimum of two predictor variables
|
Multiple regression
|
|
A multivariate technique where a large number of variables is measured and correlated with each other
|
Factor analysis
|
|
What differentiates true experiments from quasi experiments?
|
Degree of control
True experiments have random assignment of particular conditions and manipulated IVs while Quasi may have measured or manipulated IVs and probably lack random assignment |
|
These exist whenever causal conclusions cannot be drawn because there is less than complete control over the variables in the study, usually because random assignment is not feasible
|
Quasi-experiments
|
|
Design where groups are not equal to each other at the start of the study and they also experience different events in the study itself
They are effective for evaluating treatments where random assignment is impossible |
Nonequivalent control group design
|
|
Measures are taken for an extended period before and after the event expected to influence behavior
OOOOOTOOOOO |
Interrupted time series design
where T is the interruption |
|
What is the main advantage of a time series design?
|
It allows researchers to evaluate trends over time
|
|
Predictable patterns of events that occur with the passing of time
|
Trends
|
|
What are two ways to vary an interrupted time series design?
|
OOOOOTOOOOO
OOOOO OOOOO-- Combining the features of a nonequivalent control group design (control group) with the interrupted time series design (long-term trend analysis) OR OOOTOOOOOOO OOOOOOOOTOO Interrupted time series with switching replications |
|
The treatment is introduced at two different points in time for the two different groups
More confident in generality |
Interrupted time series with switching replications
|
|
Refers to information that has already been gathered for some reason aside from the research project at hand
Educational, hospital or commercial records No problem with reactivity for it requires no consent |
Archival data
|
|
Systematic examination of qualitative information in terms of predefined categories
Some archival information needs to undergo this |
Content analysis
|
|
What are some problems with archival research?
|
Information on some part of the population may be missing
Experimenter bias - only selecting supporting records |
|
What are some advantages of archival research?
|
The amount of information available is virtually unlimited
No reactivity |
|
Knowledge that their behavior is being observed can influence their behavior in ways that yield a distorted result
|
Reactivity
|
|
Why is archival research quasi-experimental?
|
It does not allow for random assignment in between-subject designs
|
|
Applied research that attempts to assess the effectiveness and value of policies or specially designed programs
|
Program evaluation
|
|
A set of procedures for predicting whether a population of sufficient size exists that would benefit from the proposed program, whether that program could solve a clearly defined problem, and whether people would use it
|
Needs analysis
|
|
Someone in the community who has a great deal of experience and specialized knowledge about the problem at hand that is otherwise unavailable to the researcher
|
Key informant
|
|
Small group of individuals who respond to a set of open-ended questions about some topic, such as the need for a particular program
|
Focus group
|
|
Aimed at improvement and monitors the progress of some program and how people are responding
|
Formative evaluation
|
|
Overall assessments of program effectiveness
More threatening |
Summative evaluation
|
|
The actual costs of a program are monitored in relation to those costs to the effectiveness of the program's outcomes
|
Cost-effectiveness analysis
|
|
How people perform and especially interface with equipment
|
Ergonomics
|
|
Name 5 categories of human error
|
Omission
Commission Due to extraneous acts (distractions/ interruptions) Sequential errors - wrong order Time errors |
|
Know a lot about the program and are closer to the problem but can be biased
|
In-house auditors
|
|
Are in need of the details about a specific program, but are free of local pressures
|
Outside auditors
|
|
Deals with developing theory
Broad explanations of behavior that can be generalized |
Basic research
|
|
The goal is to use this to solve real-world problems
|
Applied research
|
|
Examines a small number of participants, in some cases only one
|
Small N designs
|
|
The extent to which the findings for the group apply to an individual
|
Individual-subject validity
|
|
What do small-N designs provide that large-Ns don't
|
Individual-subject validity
|
|
What are some practical problems with large-N designs?
|
Too expensive
Too many people Too much time Not needed to draw conclusions |
|
Any procedure that uses behavioral, especially operant, principles to solve real-life problems
|
Applied behavior analysis
|
|
How are single-subject designs carried out?
|
No control group
The subject is compared to him/herself ABAB design with baseline measure/withdrawal (A) and treatment (B) |
|
Provides an in-depth look at a representative situation
Adds reality and meaning |
Case studies
|
|
Baseline measures are established and then treatment is introduced at different times
|
Multiple baseline design
|
|
A behavior is developed by reinforcing successive approximations to the desired behavior
|
Shaping
|
|
In this design, the target behavior is too difficult for the person to accomplish all at once so it must be shaped in small increments
Establish baseline, then criterion is made increasingly stringent Good for health-related behaviors such as exercise |
Changing criterion design
|
|
Refers to whether a particular applied behavior analysis program has value for improving society, whether its value is perceived as such by the study's participants and the extent to which the program is actually used by participants
|
Social validity
|
|
Used when comparing the effectiveness of more than one type of treatment for the same individual
The treatments are given in random order to avoid any potentially biasing order effects |
Alternating treatments design
|
|
Studies the behaviors of people or animals as they act in their everyday environments
|
Naturalistic observation
|
|
Characteristics of doing research in the real world
|
Greater validity, generalizability
Less control for extraneous variables Less ability to differentiate IVs Absence of control Observer reactivity |
|
What are some characteristics of doing research in the lab?
|
Greater ability to structure and control the experiment/extraneous variables
Problems with the contrived setting Ability to control participant selection Problems with generalizability |
|
The researcher joins the group being observed, or at least make their presence known
|
Participant observation
|
|
Having preconceived ideas about what will be observed and having those ideas color one's observations
|
Observer bias
|
|
Predefined behaviors that observers are trained to spot
|
Behavior checklists
|
|
Measured in terms of the percentage of times that observers agree
Helps reduce bias |
Interobserver reliability
|
|
Behavior is sampled at predefined times and only at those times
|
Time sampling
|
|
Only a specific set of events are recorded for observations
|
Event sampling
|
|
Measures taken of behavior, either indirectly or directly, when the subject is unaware of the measurement being made
Can reduce reactivity |
Unobtrusive measures
|
|
Structured set of questions or statements given to a group of people to measure their attitudes, beliefs, values, or tendencies to act
|
Survey
|
|
Good surveys require what type of sampling and not this type of sampling?
|
Probability
Convenience |
|
Each member of the population has some definable probability of being selected for the sample
|
Probability sampling
|
|
The proportions of important subgroups in the population are represented precisely in the sample
|
Stratified sample
|
|
What is the best feature of interviews?
|
It yields highly detailed one-on-one information
|
|
What is a big problem with written surveys?
|
Since they are sent by mail, the return rates are often low
Those who have strong opinions on the issue will most likely respond |
|
A tendency to agree with statements
Can be biased if all items are worded favorably |
Response acquiescense
|
|
A question that is structured so that it is likely to produce an answer desired by the asker
|
Leading question
|
|
When the question is asking two different things, but it is impossible to answer differently between them
|
Double-barreled question
|
|
The complete list of individuals from whom the sample will be drawn
|
Sample frame
|