chapter 5 correlation

Question 1

correlation and regression are both

Answer 1

procedures used for examining the relationship between two variables. X and Y

Question 2

similarities and differences between correlation and regression

Answer 2

correlation= strength of relationship of 2 variables 
regression= predictability of relationship between 2 variables

Question 3

characteristics of regression situations

important to know in order to to distinguish a correlation from a regression rather than confuse the two

Answer 3

• one dependent variable (Y) and one OR more independent variables (X)
• levels of the independent variable are clearly defined and selected in advance
• the value of the dependent variable for a given level of the independent variable is free to vary

-the researcher is primarily interested in predicting Y from a knowledge of X
(which makes since if you think about the fact that we have a pre-existing/selected knowledge of X already. and Y is free to vary because it is what we are interested in predicting)

Question 4

characteristics of correlation situations (how correlation differs in comparison to regression)

Answer 4

• neither variable is considered independent (there is no obvious independent variable)
• both X and Y are free to vary because neither of their values are preselected
• the researcher is primarily interested in assessing the strength of the relationship between X and Y.

(and possibly in predicting either variable from a knowledge of the other)

Question 5

similarities of correlation and regration

Answer 5

both are procedures for assessing the relationship between two variables

Question 6

why is it important to distinguish between situations that use correlation procedures and regression procedures?

Answer 6

because the assumptions underlying the use of the two procedures differ.

Question 7

4 ways correlation and regression differ

Answer 7

1. nature of the variables (presence or absence of an independent variable)
2. use of random assignment of participants to the experimental conditions
3. researchers primary interest (prediction/strength between variabeles)
4. to some extent… the kind of conclusions that can be drawn

Question 8

a bivariate frequency distribution (scatterplot) is a…

Answer 8

representation of the joint frequency of two variables

Question 9

pearson product-moment correlation coefficient

Answer 9

-a measure of the linear relationship between 2 quantitative variables
(this is the numerical index of correlation)

Question 10

possibly range for the value of a correlation coefficient

Answer 10

can range from -1 to +1

Question 11

kind of relationship when coefficient has a value of +1 (r = +1)

Answer 11

positive relationship

• data points fall in a straight line
• high scores in one variable mean high scores in the other variable
• low scores in one variable mean low scores in the other variable

Question 12

kind of relationship when coefficient has a value of -1 (r= -1)

Answer 12

negative or inverse relationship

• data points fall in a straight line
• high scores in one variable mean low scores in the other variable
• this results in a line that slopes down instead of up

Question 13

characteristics if there is no linear association between the variables (r = 0)

Answer 13

-data points fall in a circle

Question 14

intermediate degrees of association

Answer 14

coefficients less than 0
-1 < r < 0
coefficients greater than 0
0 < r < 1

-data points tend to form an ellipse

Question 15

two things a coefficient tells you

Answer 15

1. strength of the relationship (extent to which r differs from 0)
2. direction of the relationship (positive or negative)

Question 16

cross product (numerator of formula)

Answer 16

the product of the scores of two variables that are expressed as deviations from their respective means

Question 17

data falls in quadrants 1 and 3

Answer 17

positive relationship exists between variables

-data is above or below the mean for both variables

Question 18

data falls in quadrants 2 and 4

Answer 18

negative or inverse relationship exists between variable

-above the mean on one variable but below the mean on the other

Question 19

the size of the absolute value of the cross product indicates what?

Answer 19

the strength of the association

Question 20

covariance

Answer 20

mean of cross product sum

Question 21

how to obtain a measure of strength of association that is independent of the number of pairs of scores.

Answer 21

compute the mean of the cross product sum (covariance)
and divide the covariance by the standard deviations of x and y.

-the resulting measure of strength of association is independent of the size of the dispersions of x and y variables.