Correlation. What Is Correlation? Why Correlations Are Used

Transcription

1 Correlation 1 What Is Correlation? Correlation is a numerical value that describes and measures three characteristics of the relationship between two variables, X and Y The direction of the relationship Positive or negative The strength or consistency of the relationship The form of the relationship 2 Why Correlations Are Used Prediction Validity Reliability Theory verification 3 1

2 Pearson s Product Moment Correlation Coefficient Pearson s product moment correlation coefficient, or Pearson s r, measures the degree and direction of linear relationship between two variables Pearson s r must lie in the range of -1 to +1 inclusive 4 Interpretation of Pearson s r To interpret Pearson s r, you must consider two parts of it: The sign of r The magnitude, or absolute value of r 5 The Sign of r When r is greater than 0 (i.e. its sign is positive) the variables are said to have a positive or direct relation In a positive relation, as the value of one variable increases, the value of the other variable also tends to increase 6 2

3 The Sign of r When r is less than 0 (i.e., its sign is negative) the variables are said to have a negative or indirect relation In an indirect relation, as the value of one variable increases, the value of the other variable tends to decrease 7 Is the Sign of r + or -? As the number of cigarettes smoked per day increases, GPA tends to decrease As the number of cats in a farm yard increases, the number of mice tends to decrease As the weight of a cat increases, the length of its whiskers tends to increase 8 The Magnitude of r The magnitude refers to the size of the correlation coefficient ignoring the sign of r The magnitude is equivalent to taking the absolute value of r The larger the magnitude of r is, the more perfectly the two variables are related to each other The smaller the magnitude of r is, the less perfectly the two variables are related to 9 each other 3

4 When r equals 1.0, there is a perfect correlation between the variables Knowing the value of one variable exactly predicts the value of the other variable r = 1 10 r = 0 When r equals 0, either the assumptions of correlation have been violated or there is no relation between the two variables The points in a scatter plot with r = 0 will tend to form a circular cluster 11 0 < r < 1 The larger the magnitude or r is, the more the scatter plot s points will tend to cluster tightly about a line 0 < r <

5 Magnitude of r Cohen (1988) recommends the following values of r for small, medium, and large effects Correlation Negative Positive Small -.29 to to.29 Medium -.49 to to.49 Large to to Magnitude of r Which of these have r closest to 0? Which of these has the largest magnitude of r? The number of cats that you own and your IQ Your height in inches and your weight in pounds Your height in inches and your height in centimeters 14 r 15 5

6 r 16 Sum of Products (SP) 17 r 18 6

7 Example Quiz 1 Quiz 2 Quiz 1 Quiz ΣQuiz 1 = 39 ΣQuiz 1 2 = 309 SS Quiz 1 = / 5 = ΣQuiz 2 = 45 ΣQuiz 2 2 = 411 SS Quiz 2 = / 5 = ΣQuiz 1 Quiz 2 = Example SP Quiz 1 Quiz 2 r = 1 SP SS Quiz 1 SS Quiz Quiz 1 Quiz 2 n 20 Pearson s r Pearson s r makes several assumptions about the data When these assumptions are violated, r must be interpreted with extreme caution Assumptions: Linear relation Non-truncated range Sufficiently large sample size 21 7

8 Linear Relation Pearson s r, in its simplest form, only works for variables that are linearly related That is, the equation that allows us to predict the value of one variable from the value of the other is a line: Y = slope * X + intercept Always look at the scatter plot to determine if the two variables are approximately linearly related 22 Linear Relation If the variables are not linearly related, Pearson s r will indicate a smaller relation than actually exists Often, non-linear relations can be transformed into linear ones by taking the appropriate mathematical transformation 23 Square Root of Y Transformation 24 8

9 Non-Truncated Range A truncated range occurs when the range of one of the variables is very small When the range is truncated, Pearson s r will indicate a smaller relation between the variables than what actually exists Once a range truncation occurs, there is little that you can do; be careful not to design studies that will lead to a truncated range 25 Truncated Range A linear relation clearly exists in this data Consider only the data in the square (thereby truncating the range) Is the linear relation as clear as it was? No 26 Sample Size If the size of the sample is too small, relations can appear due to chance These relations disappear when a larger sample is considered Too large of a sample can make near 0 correlations statistically significant, even though they have very little explanatory power 27 9

10 Sample Size The magnitude of r does not depend on sample size The likelihood of finding a statistically significant r does depend on sample size The sample should be large enough to generalize to the population of interest 28 Correlation Causation Outliers can greatly influence r The value r 2 is called the coefficient of determination because it is the proportion of variability in one variable that can be determined from the relationship with the other variable. 29 Hypothesis Tests A sample correlation, r, can be used to decide if the variables are likely to be related in the population, ρ H 0 : ρ = 0 H 1 : ρ 0 H 0 : ρ 0 H 1 : ρ > 0 df = n 2, α =

11 Hypothesis Tests Consult a table like B-6 in the text If the r observed r critical, then reject H 0 or 31 Other Correlation Coefficients Spearman correlation Use when X and Y are both ordinal Use when you want to measure the consistency of a relationship between X and Y even if the relationship is not linear Point-biserial correlation Use when X is numerical, but Y can only take on two values 32 Other Correlation Coefficients Phi-coefficient Use when both X and Y can only take on two values 33 11