Textbook Notes (290,000)

CA (170,000)

UTSC (20,000)

Sociology (1,000)

SOCC31H3 (6)

Shirin Montazer (3)

Chapter 10

Department

SociologyCourse Code

SOCC31H3Professor

Shirin MontazerChapter

10This

**preview**shows half of the first page. to view the full**2 pages of the document.**SOCB06 chapter 10:

Correlation: age, intelligence and education attainment vary from one person to another and therefore

referred to as variables.

Many relationships are statistically significant- stronger than you would expect to obtain just as

a result of sampling error alone.

Correlations vary with respect to their strength, we visualize differences in strengths of

correlations by means of a scatter plot or scatter diagram, a graph that shows the way of scores

on any two variables, X and Y, are scatter throughout the range of possible score values.

Scatter plot: set up as the X is arranged horizontally, Y is measured across the vertical line.

- Directions of Correlation:

It can be either positive or negative in terms of direction.

Positive correlation: respondents getting high scores on the X variables also tend to get high

scores on the Y variable.

Negative correlation: respondents have high scores on the X variable and low scores on the Y

variable. Such an example is education and prejudice.

- Curvilinear Correlation:

One variable can increase while the other increase, until the other reverses itself so that one

variable can increase while the other decrease.

Correlation Coefficient: expresses both strengths and weakness and direction of straight-line

correlation. You have -1.00 and +1.00.

-1.00, -.60,-.30 and -.10 signify a negative relationship and +1.00, +.60, +.30 and +.10 indicate

positive correlation.

- Pearsons Correlation Coefficient: we can determine the strengths of X and Y variables, at the interval

level. Pearsons r gives us a measure of the strength and direction of the correlation in the sample being

studied. If we take a random sample from a specified population, we may still seek to determine

whether the obtained association between X and Y exist in the population and is not merely due to

sampling error.

- To test a measure of correlation, we must set up a null hypothesis that no relationship exist in the

population. The null hypothesis states that the population correlation p(rho) is zero .. That is p=0.

- Requirements for Use of Pearsons r Correlation Coefficent: finding out an association between X and Y

variables:

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

Unlock to view full version