This
** preview**
has intentionally

**sections.**

*blurred***to view the full version.**

*Sign up**This preview shows
pages
1–3. Sign up
to
view the full content.*

1.
Correlation and Regression (Chapters 13 and 14 of Kiess)
You should understand:
⇒
What regression and correlation are, and be able to explain the meaning of slope,
intercept, and correlation coefficient.
⇒
When to use regression and when correlation, and should know that if a given set
of points is standardized around the means of X and Y, the correlation coefficient
is equal to the slope of the regression line.
⇒
The best-fitting regression line passes through the mean of X and Y, and that it
minimizes the squared errors of estimation.
⇒
The regression line represents the central tendency of a bivariate distribution
much as the mean does for a univariate distribution, and that the sum of squared
errors of estimation represent the variance of the bivariate distribution.
⇒
Correlation does not imply causation. You should understand the factors that
affect the size of r, including nonlinearity, restriction in range, and outliers.
You should be able to:
⇒
Find the best-fitting regression line (slope and intercept) for a small set of points
by hand (using one formula).
⇒
Use regression coefficients to write an equation for a regression line.
⇒
Explain why a variable should be designated as X or Y in a regression equation.
⇒
Use the regression equation to calculate predicted values of Y for any value of X.
⇒
Examine a scatter plot and estimate what the correlation coefficient might be
(e.g., is it positive, negative, near zero?)
⇒
State a null hypothesis regarding the slope, and an alternative hypothesis. You
should be able to test the hypothesis using a non-directional or directional t test
(as appropriate).

This
** preview**
has intentionally

2. Nonparametric Statistical Tests (Chapter 15 of Kiess)

This is the end of the preview. Sign up
to
access the rest of the document.

Ask a homework question
- tutors are online