Another name for the conditions that make up a single factor (or variable).
What are levels?
These kinds of statistical tests can only be run when certain assumptions are met about the population parameters (e.g., the population has a normal distribution)
What are parametric tests?
A factorial design that results in two main effects (A level 1 vs. level 2, B level 1 vs. level 2) and one interaction (A vs. B).
What is a 2x2 design?
A test statistic used to calculate a correlation when the relationship between the two variables is linear.
What is Pearson's r?
Y
What is the criterion variable, or the variable you are trying to predict?
To avoid compounding error from running too many pair-wise T-tests (comparing two means).
What is the reason for running an ANOVA instead of multiple T-tests
These kinds of statistical tests are called "distribution free" meaning they make few, if any assumptions about the population parameter.
What are non-parametric tests?
You compare these means when analyzing a main effect.
What are the marginal means?
A test statistic used to calculate a relationship between two variables when the relationship is non-linear.
What is Spearman's Correlation?
X
What is the predictor variable (or the variable you have data for on an individual)?
The variance that goes in the numerator of the ratio used to calculate an F-statistic.
What is the between-treatments variance?
This type of Chi square test determines how well an obtained sample proportion fits the population proportion.
What is a chi-square test for goodness of fit?
You subtract these means going in a row-wise or column-wise direction when analyzing an interaction.
What are the cell means?
An r-statistic that equals 1.37.
What is an impossible to achieve r-statistic?
Regression is done by plotting this kind of line through a data-set.
What is a line of best fit or a regression line?
A measure of variance caused by random, unsystematic differences.
What is the error term or the within-treatments variance?
P1 = P2 = P3 = P4
For a Chi-square test that is testing the proportion of people that prefer 4 different brands of cereal.
What is the null hypothesis?
An interaction.
What is an effect where one factor shows a "main effect" (differs from level 1 to level 2), but only at one level of the other factor.
E.g., "Women --> perform equally at reading and picture comprehension tasks, men --> perform better on picture comprehension task vs. reading task"
The numerator in the Pearson's correlation formula,
What is the sum of the cross-product deviation scores for variables X and Y.
Σ (X-Mx)(Y-My)
a
What is the y-intercept in the linear equation?
SStotal - SSwithin = ______
What is the SSbetween?
The denominator in the Chi-square formula.
What is the expected frequency (E)?
DFa X DFb = ______
What is the formula for calculating the degrees of freedom for an interaction between Factor A and Factor B.
The proportion of variability in one variable (X) that can be determined from its relationship with another variable (Y).
What r-squared tells us?
The reason for standardizing your regression coefficients (slopes) (from b to B).
What is a multiple linear regression that contains multiple predictors on different scales?