Term
| To what does the term correlation refer? |
|
Definition
| The association or relationship between variables in interval/ratio level data |
|
|
Term
| What is a bivariate correlation? |
|
Definition
| The relationship between one independent and one dependent variable |
|
|
Term
|
Definition
| The graphical display of two bivariate interval/ratio-level variables |
|
|
Term
| What is a positive correlation between two variables? |
|
Definition
| When the two variables have corresponding scores (e.g. when one goes up, so does the second) |
|
|
Term
| What is a negative correlation? |
|
Definition
| When high scores on one variable correlate to low scores on another variable |
|
|
Term
| What does the term covariation mean? |
|
Definition
| Scores on two or more variables that vary together |
|
|
Term
| What is a regression line? |
|
Definition
| A line drawn through the center of the data represented in a scatterplot |
|
|
Term
| If the pattern of a set of data points and the line drawn through them has an upward slope, it has a ________ correlation |
|
Definition
|
|
Term
| The closer data points cluster around a line, the _______ the relationship between x and y |
|
Definition
|
|
Term
| What is the Pearson correlation coefficient? |
|
Definition
| The statistic used to measure the correlation between two interval/ratio level variables |
|
|
Term
| What does Pearson's r measure? |
|
Definition
| The strength of the linear correlation between two continuous (interval/ratio) variables |
|
|
Term
|
Definition
| The sample estimate of the correlation between the two variables in the population |
|
|
Term
|
Definition
| The population correlation coefficient |
|
|
Term
| Pearson's r indicates correlation using what range of numbers? What do the numbers mean? |
|
Definition
-1 through 1
-1 is a perfect negative relationship, 0 is no relationship, and 1 is a perfect positive relationship |
|
|
Term
| What are conditional distributions? |
|
Definition
| The distribution of possible y scores for a fixed x score |
|
|
Term
| What is the conditional mean of y? |
|
Definition
| The calculated mean of y for every value of x |
|
|
Term
| The y intercept is the point where... |
|
Definition
| The regression line crosses the y axis (the value of y when x = 0) |
|
|
Term
| The slope (aka the ___ ______ or the ______ _________) measures the amount of change produced in the y variable by a 1 unit change in the x variable |
|
Definition
| Beta coefficient; regression coefficient |
|
|
Term
| What is the residual when referring to regression analysis? |
|
Definition
| The error between predicted and actual data |
|
|
Term
| What is the main difference between the slope coefficient (b) and the correalational coefficient (r)? |
|
Definition
-b informs us about the way a relationship looks (if the relationship is negative or positive and how slanted the line is)
-r standardizes b so that you can tell the strength of a relationship and compare this to other cases or variables |
|
|
Term
| What are the 5 assumptions for testing hypotheses about the population slope and correlation coefficient? |
|
Definition
1- The observations were randomly selected
2- Both variables have normal distributions
3- The two variables are measured at the interval/ratio level
4- The variables are related in a linear form
5- assumption of homoscedasticity |
|
|
Term
| What is the assumption of homoscedasticity? |
|
Definition
| The error component is independent of and therefore uncorrelated with the independent or x variable, is normally distributed, and has an expected value of zero, and has a constant variance across all levels of x |
|
|
Term
| How can the assumptions of homoscedasticity and linearity be assessed? |
|
Definition
| By looking at the scatterplots. If the variance of y is the same for it's corresponding x, the conditions are met |
|
|
Term
|
Definition
| The finding that a change in one variable leads to or produces a change in another variable, all other things being equal |
|
|
Term
|
Definition
| The relationship between two variables is not caused by a third factor or factors related to them both |
|
|
Term
| The multiple regression equation aims to... |
|
Definition
| Estimate the effect of several independent variables on a dependent variable |
|
|
Term
| What is a partial slope (or regression) coefficient? |
|
Definition
| The partial slope of the linear relationship between each independent variable and the dependent variable |
|
|
Term
| What is the additional assumption that is made for the multivariate regression model? |
|
Definition
| The independent variables are not highly correlated among themselves |
|
|
Term
| What is multicollinearity? |
|
Definition
| Independent variables that are highly correlated with each other |
|
|
Term
| Can you determine which IV has the strongest effect on the DV by comparing unstandardized partial slope coefficients? |
|
Definition
| No, you must first standardize the slopes using beta weights |
|
|
Term
|
Definition
| The change in the amount of variance explained when the second variable is entered into the regression model |
|
|
Term
| What is a dichotomous variable? |
|
Definition
| A variable that has only two outcomes (e.g. yes or no, did happen or did not happen) |
|
|
Term
| What is a binary variable? |
|
Definition
| A dichotomous variable that is coded as either 0 or 1 |
|
|
Term
| When is the ordinary least-squares regression model best used? |
|
Definition
| In analyzing the relationship between independent variables and a continuous dependent variable that is normally distributed |
|
|
Term
| When is the logistic regression model appropriate to use? |
|
Definition
| When the dependent variable we have is a binary or dichotomous one |
|
|
Term
| What is a linear probability model? |
|
Definition
| One which assumes a linear relationship between an IV and the probability of a DV |
|
|
Term
| If the logistic regression coefficient is positive, the value of the odds multiplier will be _____ than 1, indicating an increase in the odds of the DV occurring. |
|
Definition
|
|
Term
| What is the likelihood ratio statistic? |
|
Definition
| A goodness of fit test for logistic regression which is -2 times the natural logarithm of the likelihood |
|
|
