Class Notes (1,100,000)

CA (630,000)

York (40,000)

PSYC (5,000)

PSYC 2030 (200)

Rebecca Jubis (30)

Lecture 6

This

**preview**shows half of the first page. to view the full**3 pages of the document.**1

PSYC 2030

February 17,2017

The Correlation Coefficient

A positive relationship states that if one variable increases then the corresponding variable

also increases and vice versa for the negative variable. The typically used coefficient is

Pearson r unless otherwise indicated it is appropriate to use this if there I an interval or

ratio scale. It is a descriptive statistic that describe the characteristics of the data coefficient

describes the relationship of the two variables.

Restriction of Range is when only a narrow range of scores are used for both variables

and this misrepresents the correlation and reduces the ability to make predictions

An outlier is a score that is extremely different form the others in the data set and can

distort the correlation coefficient.

Regression Analysis (line of the best fit)

- This involves making predictions on the basis of correlational data.

- Knowing the size of the correlation ad a value for xpredictor variable, it is

possible to predict the value from ycriterion variable

- Minimizes the vertical distances in all of the dots

Multiple Regression- 2 predictor variables

Coefficient of Determination

- to calculate all you do is take the coefficient r and square it

- It indicates the proportion of variants in one variable that can be accounted for by

another variable

r= +.25 is interpreted as one increases so does the other but the relationship is weak and

cannot be predicted

r squared = 0.625 = 6.25% very weak

second example

r= +.90

r squared = 81%

find more resources at oneclass.com

find more resources at oneclass.com

###### You're Reading a Preview

Unlock to view full version