Professional Documents
Culture Documents
Describes
the strength of the relationship in terms of a number from -1.0 to +1.0. Describes the direction of the relationship as positive or negative.
Types of Correlations
Variable
X increases Variable Y increases Positive Correlation Value ranging from .00 to 1.00 Example: the more you eat, the more weight you will gain
Types of Correlations
Variable
X decreases Variable Y decreases Positive Correlation Value ranging from .00 to 1.00 Example: the less you study, the lower your test score will be
Types of Correlations
Variable
X increases Variable Y decreases Negative Correlation Value ranging from -1.00 to .00 Example: the older you are, the less flexible your body is
Types of Correlations
Variable
X decreases Variable Y increases Negative Correlation Value ranging from -1.00 to .00 Example: the less time you study, the more errors you will make
Correlation Strength
.00
Positive or Negative?
IQ and reading achievement Anxiety and test scores Amount of calories consumed and weight gain. Amount of exercise and weight gain Reading achievement and math achievement Foot size and math ability
Caution!
Correlation
does not indicate causation. Correlation only establishes that a relationship exists; it reflects the amount of variability that is shared between two variables and what they have in common. Examples: Amount of ice sold and number of bee stings. SAT scores and GPA in college.
A Picture of Correlation
A
scattergram or scatter plot visually represents a correlation The X axis is on the horizontal The Y axis is on the vertical.
100
130 90
120
70
IQ
110 100 90 80 70 0 1 2 3 4
GPA
Errors
80 120 100 90 130
110
140 70
14 6 10 12 4 8 2 16
IQ
110 100 90 80 70 0 5 10 15
Errors
140
90 130
80
110
IQ
110 100 90 80 70 110 120 130 140 150 160 170 180 190 200
Weight
Caution
Do
not interpret the coefficient of correlation as a percent! you want to know the percentage of variance in one variable that is accounted for by the variance in the other variable, compute the coefficient of determination
If
Coefficient of Determination
Square r
= .50 r 2 = .25 or 25 % Twenty five percent of the variance in one variable can be accounted for by the variance in the other variable.
r2
Thirty
six percent of reading achievement is related to IQ. Reading achievement and IQ share 36% of the variance.
interpreting the correlation coefficient, always consider the nature of the population in which the two variables were observed. The correlation coefficient will vary from one population to another.
Physical prowess and age are correlated between the ages of 10 and 16. Example: Physical prowess and age are not correlated between the ages of 20 and 26.
In elementary and high school, there is a positive correlation between height and success in basketball. Example: In the pros, there is no such correlation.
may be a correlation between two variables not because there is a relationship between them but because both are related to a third variable.
Example:
Average teacher salary for 20 years and the cost of hard liquor.
is nominal data Y is interval data Correlation Formula: Point biserial Example: Correlation of sex and GPA
is ordinal data Y is ordinal or interval data (interval data must be converted to ordinal)
Correlation
Formula: Spearman rank coefficient Example: Correlation between rank and GPA
is interval Y is interval
Correlation
Formula: Pearson correlation coefficient Example: Age and the number of minutes it takes to solve a problem