[R] TukeyHSD and 0.0000000

0 views
Skip to first unread message

Fredrik Karlsson

unread,
Nov 5, 2008, 6:42:46 AM11/5/08
to r-help
Dear list,

Sorry to ask you this, but I just ran a TukeyHSD on an model with a
two thee level factors as independent variables and a numeric score
dependent variable.
The aov gives a significant interaction effect, and using the
TukeyHSD, I get almost every row to be exactly 0.0000000. Should I be
worried??

/Fredrik

--
"Life is like a trumpet - if you don't put anything into it, you don't
get anything out of it."

______________________________________________
R-h...@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.

hadley wickham

unread,
Nov 5, 2008, 8:15:57 AM11/5/08
to Fredrik Karlsson, r-help
On Wed, Nov 5, 2008 at 5:42 AM, Fredrik Karlsson <darg...@gmail.com> wrote:
> Dear list,
>
> Sorry to ask you this, but I just ran a TukeyHSD on an model with a
> two thee level factors as independent variables and a numeric score
> dependent variable.
> The aov gives a significant interaction effect, and using the
> TukeyHSD, I get almost every row to be exactly 0.0000000. Should I be
> worried??

You should be worried that you did not provide any information to
reproduce your problem. How can you expect anyone else to give you
advice?

Hadley


--
http://had.co.nz/

Fredrik Karlsson

unread,
Nov 5, 2008, 9:07:20 AM11/5/08
to hadley wickham, r-help
Hi,

Sorry about that. Sure, here is some further information:


> summary(InekeReduc)
segorg intorg l2group mani score
Belfast:3782 Belfast:3782 1: 2604 1 :2480 Min. : 1.000
L2 :7502 L2 :7440 2:12462 2 :2542 1st Qu.: 1.000
German :3782 German :3844 3 :2480 Median : 2.000
4 :2480 Mean : 3.011
5 :2480 3rd Qu.: 5.000
NA's:2604 Max. : 6.000
NA's :200.000
> summary( fit <- aov( score ~ segorg * intorg ,data=InekeReduc) )
Df Sum Sq Mean Sq F value Pr(>F)
segorg 2 15025 7512 2219.063 < 2.2e-16 ***
intorg 2 1019 509 150.441 < 2.2e-16 ***
segorg:intorg 4 164 41 12.086 8.299e-10 ***
Residuals 14857 50297 3
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
200 observations deleted due to missingness
> TukeyHSD(fit,"segorg:intorg")
Tukey multiple comparisons of means
95% family-wise confidence level

Fit: aov(formula = score ~ segorg * intorg, data = InekeReduc)

$`segorg:intorg`
diff lwr upr p adj
L2:Belfast-Belfast:Belfast 1.59590485 1.33909758 1.8527121 0.0000000
German:Belfast-Belfast:Belfast 3.04430961 2.71827718 3.3703420 0.0000000
Belfast:L2-Belfast:Belfast 0.24375305 -0.01369296 0.5011991 0.0801617
L2:L2-Belfast:Belfast 1.55958957 1.30211188 1.8170673 0.0000000
German:L2-Belfast:Belfast 2.96873225 2.71126512 3.2261994 0.0000000
Belfast:German-Belfast:Belfast 0.44195992 0.12355001 0.7603698 0.0005667
L2:German-Belfast:Belfast 2.36192845 2.10440839 2.6194485 0.0000000
German:German-Belfast:Belfast 3.42962398 3.11110098 3.7481470 0.0000000
German:Belfast-L2:Belfast 1.44840476 1.19092315 1.7058864 0.0000000
Belfast:L2-L2:Belfast -1.35215180 -1.51425695 -1.1900466 0.0000000
L2:L2-L2:Belfast -0.03631527 -0.19847075 0.1258402 0.9988643
German:L2-L2:Belfast 1.37282740 1.21068871 1.5349661 0.0000000
Belfast:German-L2:Belfast -1.15394493 -1.40170389 -0.9061860 0.0000000
L2:German-L2:Belfast 0.76602360 0.60380087 0.9282463 0.0000000
German:German-L2:Belfast 1.83371914 1.58581484 2.0816234 0.0000000
Belfast:L2-German:Belfast -2.80055656 -3.05867524 -2.5424379 0.0000000
L2:L2-German:Belfast -1.48472003 -1.74287033 -1.2265697 0.0000000
German:L2-German:Belfast -0.07557736 -0.33371711 0.1825624 0.9925827
Belfast:German-German:Belfast -2.60234969 -2.92130373 -2.2833957 0.0000000
L2:German-German:Belfast -0.68238116 -0.94057371 -0.4241886 0.0000000
German:German-German:Belfast 0.38531438 0.06624743 0.7043813 0.0056382
L2:L2-Belfast:L2 1.31583652 1.15267135 1.4790017 0.0000000
German:L2-Belfast:L2 2.72497920 2.56183071 2.8881277 0.0000000
Belfast:German-Belfast:L2 0.19820687 -0.05021410 0.4466278 0.2441352
L2:German-Belfast:L2 2.11817540 1.95494338 2.2814074 0.0000000
German:German-Belfast:L2 3.18587093 2.93730501 3.4344369 0.0000000
German:L2-L2:L2 1.40914267 1.24594419 1.5723412 0.0000000
Belfast:German-L2:L2 -1.11762966 -1.36608347 -0.8691758 0.0000000
L2:German-L2:L2 0.80233887 0.63905689 0.9656209 0.0000000
German:German-L2:L2 1.87003441 1.62143567 2.1186331 0.0000000
Belfast:German-German:L2 -2.52677233 -2.77521518 -2.2783295 0.0000000
L2:German-German:L2 -0.60680380 -0.77006911 -0.4435385 0.0000000
German:German-German:L2 0.46089174 0.21230395 0.7094795 0.0000003
L2:German-Belfast:German 1.91996853 1.67147082 2.1684662 0.0000000
German:German-Belfast:German 2.98766407 2.67639017 3.2989380 0.0000000
German:German-L2:German 1.06769554 0.81905292 1.3163382 0.0000000

/Fredrik

On Wed, Nov 5, 2008 at 1:15 PM, hadley wickham <h.wi...@gmail.com> wrote:
> On Wed, Nov 5, 2008 at 5:42 AM, Fredrik Karlsson <darg...@gmail.com>
wrote:
>> Dear list,
>>
>> Sorry to ask you this, but I just ran a TukeyHSD on an model with a
>> two thee level factors as independent variables and a numeric score
>> dependent variable.
>> The aov gives a significant interaction effect, and using the
>> TukeyHSD, I get almost every row to be exactly 0.0000000. Should I be
>> worried??
>
> You should be worried that you did not provide any information to
> reproduce your problem. How can you expect anyone else to give you
> advice?
>
> Hadley
>
>
> --
> http://had.co.nz/
>

--

"Life is like a trumpet - if you don't put anything into it, you don't get
anything out of it."

[[alternative HTML version deleted]]

Reply all
Reply to author
Forward
0 new messages