Module 18
Module 19
Module 20
Naked Stats
Random
100
A correlation coefficient can vary between ____ and ____ set of scores.
-1.00 and +1.00
100
What is the most commonly used correlation coefficient and what is it usually referred as?
Pearson product-moment correlation coefficient This is usually referred to as Pearson's r
100
What do you call the equation for the best-fitting line for a data set?
A regression line
100
For any regression coefficient, list three things that you should check.
sign,size and significance
100
Why are outliers a disadvantage?
- lessen the strength of the relationship - can either lead to type I or type II error
200
Describe a positive correlation and negative correlation.
Positive: a relationship in which the two variables move in the same direction Negative: a relationship in which an increase in one variable is accompanied by a decrease in the other variable
200
True of False Is r^2 applicable for every type of correlation coefficient?
True!
200
" A type of regression analysis that involves combing several predictor variables into a single regression equation" is called a ....
multiple regression
200
True or False (naked stats) "When the number of degrees of freedom gets large, the t-distribution converges to normal distribution."
True!
200
Why use a scatterplot?
- to summarize bivariate data - can be used to see how values of two variables tend to move with each other
300
What does a curvilinear relationship graph look like?
points are tightly clustered in a inverted U shape
300
What is the coefficient of determination?
(r^2) a measure of the proportion of the variance in one variable that is accounted for by another variable
300
Why use linear regression?
Explanation (why and how the variables are related) and prediction (the ability to predict scores from one or more variables)
300
What are two keys of true experiments?
Randomized assignment and manipulation of IV
300
In the Pearson Product Moment Correlation, what is r? List 3 things "r" does not tell us...
R is a measure of linear association R does not tell us if y is a function of x R does not tell us if x causes y R does not tell us if y causes x
400
Explain the two assumptions of causality and directionality (when misinterpreting correlations) and give an example of a directional problem
Example from the book (p.318) - Researchers observed a negative correlation between eye movement patterns and reading ability in children. The researchers assumed that poor oculomotor skills caused poor reading and proposed programs for "eye movement training." However, later research found that the relationship is actually reverse - poor reading causes more erratic eye movements.
400
When one or more of the variables is measured on an ordinal ranking scale, the appropriate correlation coefficient is _________.
Spearman's rank-order correlation coefficient
400
What is the equation for a regression line and what does each symbol mean
Y^1 = bX+a Y^1= predicted value on the Y variable b= slope of the line X= individual's score on the X variable a= y-intercept
400
In regression analysis, what methodology do you use to determine the best line which describes a linear relationship between the two variables?
A methodology called ordinary least squares, or OLS
400
Why use correlational research?
-Ethical and practical considerations -Discover new relationships -More external validity (outside of laboratory) -Good at predicting behavior because of higher external validity
500
" Finding a strong positive relationship between birth control and number of electrical appliances" is an example of what type of misinterpretation?
Third variable
500
What are the four *main* properties of the correlation coefficient
1. indicates the degree of consistent relationship 2. indicates variability 3. indicates our ability to predict "y" scores 4. indicates the spread of the scores
500
Why use multiple regression analysis?
Because in the real world, it is highly unlikely that one variable is affected by only one other variable
500
List 4 of the most common abuses of regression analysis
1. using regression to analyze a nonlinear relationship 2. reverse causality 3. correlation does not equal causation 4. omitted variable bias other options 5. highly correlated explanatory variables 6. extrapolating beyond the data 7. data mining (too many variables)
500
What are factors that distort correlations?
1. range 2. outliers 3. reliability
M
e
n
u