Stats Flashcards
Output variable
Dependent variable
Input variable that is manipulated
Independent variable
What method may be used for reducing experimenter bias?
Double-blind technique
Statistical regression
Tendency of an extreme score to fall closer to the mean after re-testing
T/F: In correlational research. The researcher does not manipulate variables
True - variables are measured, not manipulated
When do you use a factorial ANOVA?
When two independent variables are being used
Two-way ANOVA or three-way ANOVA if 3 independent variables are being used
Main effect (ANOVA)
The effect of one independent variable by itself
Interaction effects (ANOVA)
The effects of an independent variable at differing levels
What analysis is used when a study involves two or more dependent variables and one or more independent variables?
MANOVA
What methods were developed to be used in cases where a researcher knows nothing about the parameters (mean or standard deviation) of a variable?
Measures on an ordinal or nominal scale
Non-parametric
When processing data of “low quality” from small samples, or on variables about which nothing is known concerning their distribution, which statistical procedure would be most appropriate?
Non-parametric methods
What type of tests are used when the data are measured on an interval or ratio scale?
Parametric tests
A correlation coefficient has what value range?
-1.0 and + 1.0
What is the most commonly used correlation coefficient?
Pearson r
T/F: The Pearson r assumes linearity and homoscedasity
True
AB Design
Single-subject design
A = baseline B = treatment phase
When would an ABAB design be used?
When treatment is re-applied
A = baseline B = treatment A = baseline 2 B = treatment 2
In an ABAB design, the first treatment administered is withdrawn until the behavior…
Reverts to its original baseline level
THEN treatment is re-applied
From a practical and ethical standpoint, one potential problem with an ABAB design in single-subject research is what?
Getting the behavior to revert back to baseline
A test has a reliability coefficient of .90. What percentage of variability among examinees on this test is due to true score difference?
0.9
You always interpret reliability coefficient directly
How should a reliability coefficient be interpreted?
DIRECTLY!
In order to compare results from two or more different studies that measure the same concept but used different outcome measures, you would be most interested in interpreting what?
The effect size
When multiple predictors will be used to predict a score on a single criterion, what test should you use?
Multiple regression
The Solomon four-group design is used to evaluate what?
The effects of pretesting
