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;
19 Cards in this Set
- Front
- Back
Independent samples
|
When two ore more groups consist of completely different individuals
|
|
Dependent samples
|
Matched pairs - when the same individuals are tested more than one time
|
|
Parametric data
|
Data are normally distributed - greatly increased power of statistical analysis
|
|
Nonparametric data
|
Not normally distributed - categorical data (nominal or ordinal)
|
|
What are the qualifications for parametric data?
|
1. Measured variable must be normally distributed
2. Samples must be drawn at random 3. Variances in samples being compared must be equal 4. Data must be measured on interval or ratio scales (ie: continuous) |
|
Concept of Robustness
|
If a sample is large enough, parametric tests can withstand slight variations from parametric assumptions.
|
|
T-tests
|
2 groups only, 1 dependent and independent variable
|
|
Independent + Parametric (2 groups only)
Independent + Nonparametric (2 groups only) |
Independent t-test
Mann-Whitney U (ranks) |
|
Dependent + Parametric (2 groups only)
Dependent + Nonparamentric (2 groups only) |
Paired t-test
Wilcoxon (ranks) |
|
Why can't you run mulitple t-tests?
|
Increases chance for type I error (easier to find significant result)
|
|
What are ANOVAs?
|
Mean differences between more than 2 groups
|
|
Independent + Parametric (+2 groups)
Independent + Nonparametric (+2 groups) |
One-Way ANOVA
Knuskai-Wallace (ranks) |
|
Dependent + Parametric (+2 groups)
Dependent + Nonparametric (+2 groups) |
Repeated Measures ANOVA
Friedman's (ranks) |
|
Post-Hoc Tests definition
|
Only if an ANOVA is significant, can post-hoc tests be used to determine which specific groups are different using pairwise comparissons
|
|
Post-Hoc Tests examples
|
Tukey test, Newman-Keuis (NK) test, Bonferroni t-test (Dunn's), Scheffe's comparisons
|
|
Correlation coefficient (r)
|
Used to determine the strength of relationship - goodness of fit line
-1 = perfect negative relationship 1 = perfect positive rel 0=no relationship) |
|
Regression
|
Used to describe a predictive relationship between a dependent (Y) variable and an independent (X) predictor variable
|
|
Regression line
|
Y = a +b(x)
b-slope of line, rate of change in Y with one unit of X |
|
Distributions
|
chi square - nonparametric statistic used to determine if a distribution of observed frequencies differs from expected frequencies of population
|