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;
26 Cards in this Set
- Front
- Back
Slope efficient
Correlation coefficient |
Correlation coefficient represents the direct effect of the associated IV on Y
|
|
Independence
|
Variables independent if unrelated to each other
|
|
Beta weights
|
Standardized partial slopes
|
|
Partial slope coefficient
|
Estimates the difference in DV associated with a one unit difference in an IV when controlling for the effects of the other IVs
|
|
Example
Ŷ = 5078 + 732 X (bivariate regression equation) |
The slope estimate (which in this case is a value of 732) indicates the AVERAGE change in Y associated with a UNIT change in X
For our example: if we had a person who had 0 years of formal education, their income would be, on average, = to $5078.00 |
|
where our slope coefficient = 732
|
that mean, on average, for each additional year of education, the person will see an increase in their income of $732
|
|
So if we have NO relationship, what should our slope coefficient be?
|
ZERO…meaning no effect on the DV at all.H0:X is NOT associated with Y (therefore the slope = 0 in the population)
H1:X IS associated with Y, therefore the slope is NOT zero in the population |
|
If we CAN reject the null,
|
what we are saying is that the slope estimate IS significantly different from 0 at a .05 level (95% level of confidence)
|
|
Purpose of one way ANOVA and how it’s executed (logic)
|
Designed to be used with interval ratio-level dependent variables.
The test compares the amount of variation between categories with the amount of variation w/i categories. The greater the differences between categories (means) relative to the differences w/i categories (sd)the more likely the null hypothesis of no difference is false and ca be rejected. |
|
Intercept (slope)
|
the point where the regression line “intercepts” or intersects the Y-axis
-- It estimates the average value of Y when X is equal to zero |
|
R – square
(Coefficient of determination |
accounts for the proportion of variation in the dependent variable “explained” by the independent variable
The range of R2 values: 0.0 to +1.0 |
|
R – square
(Coefficient of determination) In our example, R2 = .56 |
We can say: our model, where education is the independent variable, accounts for 56% of the variation in salary, the dependent variable
|
|
What does model fit mean in regression analyses?
|
used to model the relationships between a response variable and one or more predictor variables,
|
|
What is the value of measure of association when assessing variable relationships?
|
Measure of association help us trace causal relationships among variables, and they are our most important and powerful statistical tools for documenting, measuring, and analyzing cause-and effect relationships
|
|
What are the strengths and weaknesses of multiple regression analysis
|
Muliple regression and correlation are powerful tools for analyzing the interrelationship among three or more variables.
They are not cheap. They assume that each independent variable has a linear relationship with the dependent variable. They assume there is no interaction among the variables in the equation. They assume the independent variables are uncorrelated with each other. |
|
What is the difference between a true experiment and a quasi experiment?
|
True- uses randomized choice, selecting subjects and methods in a way that prevents bias in results Quasi-doesn’t use proper random assignment, recruitment of people can cause bias.
|
|
What are the strengths and weaknesses of a cross sectional research design approach?
|
association by itself is not proof that a causal relationship exists. Causation and association are two different things.
|
|
Within group variance
|
Pattern of variation within each category
|
|
Between group variance
|
Measure of variation in score between categories
|
|
Bivariate relationship
|
relationship between two random variables
(zero order) |
|
Partial Correlation
|
Examines how a bivariate relationship (i.e. a relationship between two variables) is affected by a third variable
Distinctions between direct relationships, spurious or intervening relationships, interactive relationships |
|
If the partial correlation coefficient differs from the zero-order coefficient
|
we conclude the third variable has an effect on the bivariate relationship
|
|
Partial Correlation
when you introduce a THIRD variable to check to see |
see if you have a direct or a spurious relationship
|
|
Partial correlations with a control variable added are called
|
first-order correlations and are symbolized as ryx.z where the variable to the right of the dot is the control variable
|
|
Interpreting the partial correlation result:
The first order partial correlation (ryx.z = 0.43) is lower than the zero-order correlation (ryx = 0.50), but |
the difference is slight
This suggests a direct relationship: regardless of SES, husbands’ household work increases with the number of children in the house |
|
Interpreting the partial correlation result:
If the partial correlation is much different (e.g. lower) than the original zero-order correlation, ...If the partial correlation is much higher than the zero-order r value, |
then we likely either have a spurious relationship between X and Y, or an intervening relationship...another causal structure may be at work
|