How to Calculate the Coefficient of a Correlation on a Calculator

Save

The correlation coefficient is a way to determine the strength of the linear association between two variables, such as "X" and "Y." The correlation will always be between the numbers "-1" and "1." If the correlation is negative, then there is a negative relationship. If the correlation is "0," then there is a neutral relationship. If the correlation is positive, then there is a positive relationship.

  • Write out the numbers of the "X" and "Y" values you want the correlation coefficient of:

    X Values Y Values

    20 4.1

    21 4.6

  • Count the total number of "X" and "Y" values; in this case, there are four, so "N"=4.

  • Multiply "X" by "Y," "X" by "X" and "Y" by "Y" for all four values with a calculator:

    XY=204.1=82 XY=214.6=96.6

    XX=2020=400 XX=2121=441

    YY=4.14.1=16.81 YY=4.64.6=21.16

  • Add up the "X" values, the "Y" values and all of the multiplied "X" and "Y" values; find the square roots of the "X" and "Y" values and add them together:

    X+X=20+21=41

    Y+Y=4.1+4.6=8.7

    XY=204.1=82 and XY=214.6=96.6 and 82+96.6=178.6

    XX=2020=400 and XX=2121=441 and 400+441=841

    YY=4.14.1=16.81 and YY=4.64.6=21.16 and 16.81+21.16=37.97

  • Plug the numbers into the formula (r) =[ NΣXY - (ΣX)(ΣY) / Sqrt([NΣX^2 - (ΣX)^2][NΣY^2 - (ΣY)^2])] and make the calculations with the calculator:

    Correlation(r)=((4)(178.6)-(41)(8.7)) / square-root ([4)(841)-(4141)][(4)(37.97)-(8.7*8.7)])

    Correlation(r)=(714.4-356.7) / square-root ([3364-1681)*[151.88-75.69)

    Correlation(r)=357.7 / square-root (1683*76.19)

    Correlation(r)=357.7 / square-root (128227.77)

    Correlation(r)=357.7 / 358.089

    Correlation(r)=0.9989

Tips & Warnings

  • Use a square root calculator, such as Math.com's, to get the square roots of large numbers; simply enter the number and press "Calculate," and the square root will appear.
  • Remember to plug the numbers you get from the calculator into their correct places within the formula; you won't get the proper correlation coefficient otherwise, and any number you get that is either below -1 or above 1 is an incorrect coefficient.

References

Promoted By Zergnet

Comments

Resources

You May Also Like

Related Searches

Check It Out

Can You Take Advantage Of Student Loan Forgiveness?

M
Is DIY in your DNA? Become part of our maker community.
Submit Your Work!