Finding Coefficient of Determination

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
1 replies · 2K views
Tchao
Messages
7
Reaction score
0

Homework Statement



If the slope of the regression line is negative and the correlation coefficient is -0.5, coefficient of determination is:

Homework Equations



coefficient of determination: r2 = 1 - SSE/SST

correlation coefficient: r = Sxy / (√Sxx * √Syy )

The Attempt at a Solution


r = -0.5
r2 = 0.25 but because slope of regression line is negative, then coefficient of determination would be also negative.
r2 = -0.25

I am not sure if my though process in determining the correct answer is right or not.
 

Attachments

  • upload_2017-4-23_22-43-16.png
    upload_2017-4-23_22-43-16.png
    8.6 KB · Views: 516
  • upload_2017-4-23_22-43-35.png
    upload_2017-4-23_22-43-35.png
    16.6 KB · Views: 504
on Phys.org
Tchao said:

Homework Statement



If the slope of the regression line is negative and the correlation coefficient is -0.5, coefficient of determination is:

Homework Equations



coefficient of determination: r2 = 1 - SSE/SST

correlation coefficient: r = Sxy / (√Sxx * √Syy )

The Attempt at a Solution


r = -0.5
r2 = 0.25 but because slope of regression line is negative, then coefficient of determination would be also negative.
r2 = -0.25

I am not sure if my though process in determining the correct answer is right or not.

Hi Tchao,

Indeed, the coefficient of determination is ##r^2=0.25##.
And it is positive, since a square is always positive.
It means that the coefficient of determination doesn't tell us whether the slope is upward or downward, only that there is a correlation.
 
  • Like
Likes   Reactions: Greg Bernhardt