0
$\begingroup$

I have two different p-values here, do I use the Pr( t ) or the regular p-value?

Call:
lm(formula = ddd ~ CON, data = data)

Residuals:
    Min      1Q  Median      3Q     Max 
-0.5726 -0.4859 -0.4570  0.5141  0.5430 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  1.60144    0.09117  17.565   <2e-16 ***
CON         -0.02889    0.02360  -1.224    0.222    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.5004 on 315 degrees of freedom
Multiple R-squared:  0.004735,  Adjusted R-squared:  0.001575 
F-statistic: 1.499 on 1 and 315 DF,  p-value: 0.2218
$\endgroup$
  • 4
    $\begingroup$ What model are you using? There are p-values associated with regression parameters and for the overall fit F test. So it depends on which one you are talking about. $\endgroup$ – Michael Chernick May 1 '17 at 20:46
4
$\begingroup$

Because you fit a simple regression model (one covariate), so the p-value for slop (0.222) and overall F-test p-value (0.2218) are the same thing. If you ask the program give you one more digit on slop p-value, you will get 0.2218, instead of 0.222.

About p-value for intercept (<2e-16), I think you are not interested in it.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.