Possible Duplicate:
How to interpret F- and p-value in ANOVA?

I found that I can use ANOVA also for ONE Model, doing something like:

> anova(lm(a~b))
Analysis of Variance Table

Response: a
           Df   Sum Sq   Mean Sq F value    Pr(>F)    
b           1 0.002679 0.0026791  11.191 0.0009001 ***
Residuals 398 0.095282 0.0002394                      
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 

I know that ANOVA check the means BUT what test is that if I use only ONE model? If the p.value is above 0.05 it means that the regression fit good?

  • $\begingroup$ Please undertake some research before asking questions. The meaning of a p-value is extensively documented and heavily discussed even on this site. $\endgroup$
    – whuber
    Nov 21, 2011 at 20:42
  • $\begingroup$ @whuber I KNOW what is a "p-value" but I don't know why (IN THIS CASE) Is usefull because i'm not comparing groups, I only used one linear model...so what "means" ANOVA is checking ?? $\endgroup$
    – Dail
    Nov 21, 2011 at 20:46
  • $\begingroup$ A prior answer I believe should be satisfactory for this question, Logic behind the ANOVA F-test in simple linear regression. $\endgroup$
    – Andy W
    Nov 21, 2011 at 20:47
  • $\begingroup$ Thanks, @Andy. There are many questions on this site that could serve as a duplicate of this one (the search brings up 43 of them). $\endgroup$
    – whuber
    Nov 21, 2011 at 20:50
  • 1
    $\begingroup$ Yes @whuber, I agree. To Dail, it is good practice to comb through the related questions on the site. In this instance, just perusing the questions tagged anova would turn up a wealth of potential resources and explanation for similar questions. $\endgroup$
    – Andy W
    Nov 21, 2011 at 21:24


Browse other questions tagged or ask your own question.