Linear Correlation

Linear Correlation

A correlation exists between two variables where one of the variables is related to the

other in some way. Linear correlation is when the relationship that exists between the

two variables is linear.

The degree of linear correlation is found by calculating the Pearson’s Correlation

Coefficient.

  

        2 22 2

n xy x y r

n x x n y y

 

 

  

   

Where x and y are the variables whose relationship is in question. The value of r will

always be between -1 and 1 inclusive.

Remember r is a measure of the strength of a linear association between two variables.

Example

Suppose the age and distance (to the nearest mile) moved was recorded for 25 adults

who moved to Phoenix from outside the state of Arizona. Is there a significant linear

correlation between the age of the adult and the distance moved?

Age Distance (mi.)

25 1852

37 1603

72 450

47 975

72 373

41 1336

59 586

32 282

59 202

37 1400

80 287

63 1801

80 1013

31 356

38 375

31 816

45 749

36 2367

45 2368

22 2721

45 2341

44 2455

48 1001

31 1725

48 1075

Let the ages be X and the distances be Y. To calculate the Correlation Coefficient make

a table.

X=Age Y=Distance (mi.) X Y 2X 2Y

25 1852 46300 625 3429904

37 1603 59311 1369 2569609

72 450 32400 5184 202500

47 975 45825 2209 950625

72 373 26856 5184 139129

41 1336 54776 1681 1784896

59 586 34574 3481 343396

32 282 9024 1024 79524

59 202 11918 3481 40804

37 1400 51800 1369 1960000

80 287 22960 6400 82369

63 1801 113463 3969 3243601

80 1013 81040 6400 1026169

31 356 11036 961 126736

38 375 14250 1444 140625

31 816 25296 961 665856

45 749 33705 2025 561001

36 2367 85212 1296 5602689

45 2368 106560 2025 5607424

22 2721 59862 484 7403841

45 2341 105345 2025 5480281

44 2455 108020 1936 6027025

48 1001 48048 2304 1002001

31 1725 53475 961 2975625

48 1075 51600 2304 1155625

Sum 1168 30509 1292656 61102 52601255

We need a column for the X values, the Y values, X times Y, X squared, and Y squared.

Find the sum of each of these columns. Then these values are plugged into the formula.

  

       

    

       

  

2 22 2

2 2

25 1292656 1168 30509

25 61102 1168 25 52601255 30509

32316400 35634512

1527550 1364224 1315031375 930799081

3318112 33318112

404.1361157 19601.84415163326 384232294

3318112

7

n xy x y r

n x x n y y

r

r

r

r

 

 

 

 

 

 

   

 

  

   

.4188576447 .41886 921813.155

   

This is Pearson’s Correlation Coefficient, but to determine if there is a significant linear

correlation present, we must have something to compare this to. Use the table of critical

values for Pearson Correlation Coefficient.

For a significance level of 0.05, the critical value for a sample size of 25 is .396.

If the absolute value of r is greater than the critical value from the table, then we

conclude there is significant linear correlation.

Since .41886 .41886 .396   = the critical value for 0.05 level of significance at a

sample size of 25, then there is a significant linear correlation between the age and the

distance moved.

Coefficient of Determination

Can the correlation coefficient be used to explain the variation?

Yes, the Coefficient of Determination is the proportion of variation in Y that is explained

by the linear association between x and y.

The Coefficient of Determination is the square of the Pearson Correlation Coefficient.

2Coefficeint of Determination r

Where r is the Pearson Correlation Coefficient.

Example

Suppose the age and distance moved was recorded for 25 adults who moved to

Phoenix from outside the state of Arizona. What proportion of variation in the distance

moved can be explained by the linear relationship between the distance moved and the

ages of those moving?

Pearson’s Correlation Coefficient was calculated above to be .41886r  

The Coefficient of Determination is   22 .41886 .17544r   

We can say that .17544 or 17.54% of the variation in the distance moved can be

explained by the linear relationship between the distance moved and the age of the

adult moving.