The correlation coefficient is a measure of
Association
While you can use the correlation coefficient as its own test statistic, what is the other appropriate test statistic often used to examine the significance of a correlation?
T-test
This chapter illustrates that you can also incorporate ____ into the correlation coefficient
Statistical significance
Correlation coefficients examine
The relationship between variables
Correlation coefficients can test this many variables at a time:
Only two
The appropriate test statistic to use is the:
T-test for the correlation coefficient
Correlations can be:
Directional or nondirectional
Significant correlations are not able to indicate:
Casuality
What is another term for a positive correlation?
Direct
What is another term for a negative correlation?
Indirect
What is the most important chacteristic of a correlation coefficient?
Absolute value
Which of the following is an example of a null hypothesis for testing a correlation coefficient?
H0:Pxy=0
If you predict that a relationship between two variables would be either positive or negative, what type of test should you use?
One-tailed
If you do not predict that a relationship between two variables would be either positive or negative, what type of test should you use?
Two-tailed
The level of risk or Type I error typically set for testing the level of significance of a correlation coefficent is which of the following?
.05
Which of the following is the appropriate method for calculating the degrees of freedom associated with a correlation coefficient?
n-2
In the formula for calculating degrees of freedom for a correlation coefficient, what does the n represent?
Number of Pairs
What is the name of the Greek letter p?
Rho
Which of the following Greek symbols is used to represent the population estimate for a correlation coefficient?
p
Which of the following represents the test statistic for a correlation coefficient?
R
Which of the following is another use of correlation coefficients?
Estimating reliability
Which of the following types of reliability correlates scores on a single measure on two different occasions?
Test-retest
If the correlation between two variables is .496, how much of the variance has not been accounted for?
75.4%
If the correlation between two variables is .496, what is the coefficient of determination?
.246
What does the statement rxy=/ 0 represent?
Research hypothesis
If a research hypothesis does not predict the direction of a relationship, the test is__.
Two-tailed
If a research hypothesis does predicts that there is a direct relationship between two variables, the test is__.
One tailed
In the equation r(65) = .45, p<.05, what does r represent?
Test Statistic
When computing a correlation coefficient, if you have a degrees of freedom of 27, your sample size must be:
29
When computing a correlation coefficient, if you have a degrees of freedom of 55, your sample size must be:
57
If a simple Pearson correlation value=.512, what percentage of variance is accounted for?
47%
If a simple Pearson correlation value=.362, what percentage of variance is accounted for?
87%
If a simple Pearson correlation value=.75, what percentage of variance is accounted for?
44%
Correlation coefficients examine the relationship between variables.
True
A correlation coefficient can inly test one variable at a time.
False
Regression equation:
In regression, an equation that defines the line that has the best fit with your data.
Regression line:
The line that best fit that is drawn (or calculated) based on on the values in the regression equation.
Line of best fit:
The regression line that best fits the data and minmizes the error in prediction.
Error in prediction (aka error of estimate):
The difference between the actual score and the predicted score in a regression.
Criterion or dependent variable:
The outcome variable, or the variable that is predicted, in a regression analysis.
Predictor or independent variable:
The variable that is used to predict the dependent variable in a regression analysis.
Y prime:
The predicted value of Y, the dependent variable, written as Y'
Standard error of estimate:
The average amount that each data point differs from the predicted data point.
Multiple regression:
A type of regression in which more than on independent variable is included in the analysis.
Linear regression uses correlations as its basis.
True
Linear regression can be used to predict values of the dependent variable for individuals outside of your data set.
True
The higher the absolute value of your correlation coefficient, the worse your predictive power is.
False. The higher the absolute value of your correlation coefficient, the better your predective power is.
When using multiple regression, it is always best to include as many predictor variables as possible.
False. Careful judgement should be used whendeciding which deciding which predictor variables to include.
When using multiple regression, it is best to select independent variables that are uncorrelated with one another but are all related to the predicted variable.
True