|
|
How does p-value mean for R-square |
|
Message/Author |
|
Soo Owen posted on Monday, October 01, 2018 - 11:43 am
|
|
|
So, I ran a multivariate regression analysis with 7 continuous variables and dummy predictors. I've got some significant coefficients but I don't know how to interpret the pvalues of R-squares. Below is part of my output. While two dummy variables significantly predict the Y1ACADEV, all predictors explain only 7% of Y1ACADEV. But the p-value for the R-square is not significant (0.118). What does it mean? Y1ACADEV ON FIELD -0.025 0.110 -0.223 0.824 PEERSOCIAL -0.314 0.144 -2.184 0.029 BOTH 0.108 0.046 2.380 0.017 R-SQUARE Observed Two-Tailed Variable Estimate S.E. Est./S.E. P-Value Y1ACADEV 0.065 0.042 1.564 0.118 |
|
|
R-square typically has a non-symmetric sampling distribution so you want to use bootstrapping to get non-symmetric CIs and evaluate significance that way instead of p-values which assume symmetric. |
|
Back to top |
|
|