Rsquared for anova

Thanks Dieter,

Even if I use lm(), I get the following output:
summary(fit)
? ? ? ? ? ? ?Df ?Sum Sq Mean Sq F value ? ?Pr(>F)
nh1 ? ? ? ? ? ?1 ? 324.0 ?323.99 ?139.13 < 2.2e-16 ***
nh2 ? ? ? ? ? ?1 ? 723.1 ?723.12 ?310.53 < 2.2e-16 ***
nh3 ? ? ? ? ? ?1 ?1794.2 1794.21 ?770.49 < 2.2e-16 ***
Residuals ? 4604 10721.2 ? ?2.33
---
Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.?

no R squared to be found.

The lm() output gives this:
summary(fit)
? ? ? ? ? ? ?Df ?Sum Sq Mean Sq F value ? ?Pr(>F)
nh1 ? ? ? ? ? ?1 ? 324.0 ?323.99 ?139.13 < 2.2e-16 ***
nh2 ? ? ? ? ? ?1 ? 723.1 ?723.12 ?310.53 < 2.2e-16 ***
nh3 ? ? ? ? ? ?1 ?1794.2 1794.21 ?770.49 < 2.2e-16 ***
Residuals ? 4604 10721.2 ? ?2.33
---
Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.?

I just want to see how wel the model fits...

Another strange thing. I did the same calculations in SPSS and got
different results, among others, higher p values. lm() and aov() just
do linear regressions (anova) right? I tried with the same factorial
variables (without interaction effects to test).

Thanks!
dorien wrote:
I calculate an anova test in the following way:

... aov example

I want to check the fit of the model with Rsquared

Try summary(lm(...)) instead.

Dieter

--
View this message in context: http://r.789695.n4.nabble.com/Rsquared-for-anova-tp3452399p3452434.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

--
Dorien Herremans

Department of Environment, Technology and Technology Management
Faculty of Applied Economics
University of Antwerp

B.513
Prinsstraat 13
2000 Antwerp
Belgium
+32 3 265 41 25
Dorien Herremans

Department of Environment, Technology and Technology Management
Faculty of Applied Economics
University of Antwerp

B.513
Prinsstraat 13
2000 Antwerp
Belgium
+32 3 265 41 25

Even if I use lm(), I get the following output:

summary(fit)
             Df  Sum Sq Mean Sq F value    Pr(>F)
nh1            1   324.0  323.99  139.13 < 2.2e-16 ***
nh2            1   723.1  723.12  310.53 < 2.2e-16 ***
nh3            1  1794.2 1794.21  770.49 < 2.2e-16 ***
Residuals   4604 10721.2    2.33
---
Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.?

You probably used the wrong fit. I get your results, when I use
summary(aov), instead of summary(lm())

Dieter

d = data.frame(nh1 = sample(letters[1:2],4000,TRUE),
               nh2 = sample(letters[1:2],4000,TRUE),
               nh3 = sample(letters[1:2],4000,TRUE),
               x = rnorm(4000))

summary(lm(x~nh1+nh2+nh3,data=d))
Coefficients:
            Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01517    0.03121  -0.486    0.627
nh1b         0.01990    0.03116   0.639    0.523
nh2b        -0.02932    0.03116  -0.941    0.347
nh3b         0.02814    0.03116   0.903    0.367

Residual standard error: 0.9851 on 3996 degrees of freedom
Multiple R-squared: 0.0005315,  Adjusted R-squared: -0.0002189 
F-statistic: 0.7083 on 3 and 3996 DF,  p-value: 0.5469 

summary(aov(x~nh1+nh2+nh3,data=d))
summary(aov(x~nh1+nh2+nh3,data=d))
Df Sum Sq Mean Sq F value Pr(>F)
nh1            1    0.4 0.40556  0.4179 0.5180
nh2            1    0.9 0.86530  0.8916 0.3451
nh3            1    0.8 0.79142  0.8155 0.3666
Residuals   3996 3878.2 0.97052

I suggest you kill your history file and your workspace, and try again from
scratch. I never use these, but always start from scratch with my program
(self-consistent) in the text editor.

Dieter

--
View this message in context: http://r.789695.n4.nabble.com/Rsquared-for-anova-tp3452399p3457451.html
Sent from the R help mailing list archive at Nabble.com.
Dorien Herremans <dorien.herremans <at> ua.ac.be> writes:
Thanks Dieter,

Even if I use lm(), I get the following output:

summary(fit)
? ? ? ? ? ? ?Df ?Sum Sq Mean Sq F value ? ?Pr(>F)
nh1 ? ? ? ? ? ?1 ? 324.0 ?323.99 ?139.13 < 2.2e-16 ***
nh2 ? ? ? ? ? ?1 ? 723.1 ?723.12 ?310.53 < 2.2e-16 ***
nh3 ? ? ? ? ? ?1 ?1794.2 1794.21 ?770.49 < 2.2e-16 ***
Residuals ? 4604 10721.2 ? ?2.33
---
Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.?

no R squared to be found.

Really ??

  When I run the example in ?lm I get:
summary(lm.D9)
Call:
lm(formula = weight ~ group)

Residuals:
    Min      1Q  Median      3Q     Max 
-1.0710 -0.4938  0.0685  0.2462  1.3690 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)   5.0320     0.2202  22.850 9.55e-15 ***
groupTrt     -0.3710     0.3114  -1.191    0.249    
---
Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 

Residual standard error: 0.6964 on 18 degrees of freedom
Multiple R-squared: 0.07308,	Adjusted R-squared: 0.02158 
F-statistic: 1.419 on 1 and 18 DF,  p-value: 0.249 

    Note the last three lines.  I would be shocked if there
weren't a similar set of information in your output ...

  Your output looks like the result of summary(aov(...))
Another strange thing. I did the same calculations in SPSS and got
different results, among others, higher p values. lm() and aov() just
do linear regressions (anova) right? I tried with the same factorial
variables (without interaction effects to test).
There are a lot of things that could be different. Are
you treating factors (categorical predictors) as numeric by
accident?
Thanks!

On 15 April 2011 18:07, Dieter Menne <dieter.menne <at> menne-biomed.de> wrote:
dorien wrote:
I calculate an anova test in the following way:

... aov example

I want to check the fit of the model with Rsquared

Try summary(lm(...)) instead.

Dieter

--
View this message in context:
http://r.789695.n4.nabble.com/Rsquared-for-anova-tp3452399p3452434.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

--
Dorien Herremans

Department of Environment, Technology and Technology Management
Faculty of Applied Economics
University of Antwerp

B.513
Prinsstraat 13
2000 Antwerp
Belgium
+32 3 265 41 25

Omg Dieter, thanks so much... I just had to q() and R... that solved
it :-) excellent! finally got the R2 :-)
Dorien Herremans <dorien.herremans <at> ua.ac.be> writes:

Thanks Dieter,

Even if I use lm(), I get the following output:

summary(fit)
? ? ? ? ? ? ?Df ?Sum Sq Mean Sq F value ? ?Pr(>F)
nh1 ? ? ? ? ? ?1 ? 324.0 ?323.99 ?139.13 < 2.2e-16 ***
nh2 ? ? ? ? ? ?1 ? 723.1 ?723.12 ?310.53 < 2.2e-16 ***
nh3 ? ? ? ? ? ?1 ?1794.2 1794.21 ?770.49 < 2.2e-16 ***
Residuals ? 4604 10721.2 ? ?2.33
---
Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.?

no R squared to be found.

?Really ??

?When I run the example in ?lm I get:

summary(lm.D9)
Call:
lm(formula = weight ~ group)

Residuals:
? ?Min ? ? ?1Q ?Median ? ? ?3Q ? ? Max
-1.0710 -0.4938 ?0.0685 ?0.2462 ?1.3690

Coefficients:
? ? ? ? ? ?Estimate Std. Error t value Pr(>|t|)
(Intercept) ? 5.0320 ? ? 0.2202 ?22.850 9.55e-15 ***
groupTrt ? ? -0.3710 ? ? 0.3114 ?-1.191 ? ?0.249
---
Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1

Residual standard error: 0.6964 on 18 degrees of freedom
Multiple R-squared: 0.07308, ? ?Adjusted R-squared: 0.02158
F-statistic: 1.419 on 1 and 18 DF, ?p-value: 0.249

? ?Note the last three lines. ?I would be shocked if there
weren't a similar set of information in your output ...

?Your output looks like the result of summary(aov(...))

Another strange thing. I did the same calculations in SPSS and got
different results, among others, higher p values. lm() and aov() just
do linear regressions (anova) right? I tried with the same factorial
variables (without interaction effects to test).
? There are a lot of things that could be different. Are
you treating factors (categorical predictors) as numeric by
accident?

Thanks!

On 15 April 2011 18:07, Dieter Menne <dieter.menne <at> menne-biomed.de> wrote:
dorien wrote:
I calculate an anova test in the following way:

... aov example

I want to check the fit of the model with Rsquared

Try summary(lm(...)) instead.

Dieter

--
View this message in context:
http://r.789695.n4.nabble.com/Rsquared-for-anova-tp3452399p3452434.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help <at> r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

--
Dorien Herremans

Department of Environment, Technology and Technology Management
Faculty of Applied Economics
University of Antwerp

B.513
Prinsstraat 13
2000 Antwerp
Belgium
+32 3 265 41 25

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Dorien Herremans

Department of Environment, Technology and Technology Management
Faculty of Applied Economics
University of Antwerp

B.513
Prinsstraat 13
2000 Antwerp
Belgium
+32 3 265 41 25