[R] Missing rows anova

Justin Thong justinthong93 at gmail.com
Wed Jul 20 13:34:34 CEST 2016


Hi Michael,

Thank you for the reply.

I am sorry I forgot to print out the anova table to make my question clear.

              Df            Sum Sq      Mean Sq     F value   Pr(>F)
S             2            0.000019    9.630e-06    0.818    0.444
x1            1            0.000256    2.560e-04   21.751   9.44e-06 ***
ID           47           0.003524    7.498e-05    6.370     3.35e-15 ***
Resid    102           0.001201    1.177e-05
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

There is *no* unique value for ID for each combination of S and x1. For
example, S=1 and x1=0 can equal to either  B or C or  D or  E or  F .
Perhaps you mean that for each combination of S and x1 have different
values. If that is the case, I think maybe it makes sense.

*What I think? *
Anova has this thing where it fits the terms of 1st order first ( a formula
term including no interactions) before it fits a 2nd order term ( a formula
term including 1 interaction) and so on.

First Order--> Second Order--> Third Order--> etc

Therefore, it is known that the true fitting formula is not S+x1+S:x1+ID
but it is S+x1+ID+S:x1. Hence, it appears that ID is fitted before S:x1 but
since ID is a more refined factor than S:x1, it can be said that S:x1 is
already included in the fit of ID so R recognizes the linear dependance and
excludes the term S:x1.
In other words, S:x1 is linearly dependant to ID. And so, the row S:x1
disappears because it is considered within ID.

Does this makes sense?








On 19 July 2016 at 16:19, Michael Dewey <lists at dewey.myzen.co.uk> wrote:

> Presumably it disappears because there is a unique value of ID for eac
> combination of S*x1 so they are indistinguishable.
>
>
> On 19/07/2016 12:53, Justin Thong wrote:
>
>> Why does the S:x1 column disappear (presumably S:x1 goes into ID but I
>> dont
>> know why)? S is a factor, x1 is a covariate and ID is a factor.
>>
>>     rich.side<-aov(y~S*x1+ID)
>>     summary(rich.side)
>>
>> Below is the model frame
>>
>>     model.frame(~S*x1+ID)
>>
>>         S x1  ID
>>     1   1 12   A
>>     2   1 12   A
>>     3   1 12   A
>>     4   1 12   A
>>     5   1  0   B
>>     6   1  0   B
>>     7   1  0   B
>>     8   1  0   B
>>     9   1  0   C
>>     10  1  0   C
>>     11  1  0   C
>>     12  1  0   C
>>     13  1  0   D
>>     14  1  0   D
>>     15  1  0   D
>>     16  1  0   D
>>     17  1  0   E
>>     18  1  0   E
>>     19  1  0   E
>>     20  1  0   E
>>     21  1  0   F
>>     22  1  0   F
>>     23  1  0   F
>>     24  1  0   F
>>     25  2  6  AB
>>     26  2  6  AB
>>     27  2  6  AB
>>     28  2  6  AB
>>     29  2  6  AC
>>     30  2  6  AC
>>     31  2  6  AC
>>     32  2  6  AC
>>     33  2  6  AD
>>     34  2  6  AD
>>     35  2  6  AD
>>     36  2  6  AD
>>     37  2  6  AE
>>     38  2  6  AE
>>     39  2  6  AE
>>     40  2  6  AE
>>     41  2  6  AF
>>     42  2  6  AF
>>     43  2  6  AF
>>     44  2  6  AF
>>     45  2  0  BC
>>     46  2  0  BC
>>     47  2  0  BC
>>     48  2  0  BC
>>     49  2  0  BD
>>     50  2  0  BD
>>     51  2  0  BD
>>     52  2  0  BD
>>     53  2  0  BE
>>     54  2  0  BE
>>     55  2  0  BE
>>     56  2  0  BE
>>     57  2  0  BF
>>     58  2  0  BF
>>     59  2  0  BF
>>     60  2  0  BF
>>     61  2  0  CD
>>     62  2  0  CD
>>     63  2  0  CD
>>     64  2  0  CD
>>     65  2  0  CE
>>     66  2  0  CE
>>     67  2  0  CE
>>     68  2  0  CE
>>     69  2  0  CF
>>     70  2  0  CF
>>     71  2  0  CF
>>     72  2  0  CF
>>     73  2  0  DE
>>     74  2  0  DE
>>     75  2  0  DE
>>     76  2  0  DE
>>     77  2  0  DF
>>     78  2  0  DF
>>     79  2  0  DF
>>     80  2  0  DF
>>     81  2  0  EF
>>     82  2  0  EF
>>     83  2  0  EF
>>     84  2  0  EF
>>     85  3  4 ABC
>>     86  3  4 ABC
>>     87  3  4 ABC
>>     88  3  4 ABC
>>     89  3  4 ABD
>>     90  3  4 ABD
>>     91  3  4 ABD
>>     92  3  4 ABD
>>     93  3  4 ABE
>>     94  3  4 ABE
>>     95  3  4 ABE
>>     96  3  4 ABE
>>     97  3  4 ABF
>>     98  3  4 ABF
>>     99  3  4 ABF
>>     100 3  4 ABF
>>     101 3  4 ACD
>>     102 3  4 ACD
>>     103 3  4 ACD
>>     104 3  4 ACD
>>     105 3  4 ACE
>>     106 3  4 ACE
>>     107 3  4 ACE
>>     108 3  4 ACE
>>     109 3  4 ACF
>>     110 3  4 ACF
>>     111 3  4 ACF
>>     112 3  4 ACF
>>     113 3  4 ADE
>>     114 3  4 ADE
>>     115 3  4 ADE
>>     116 3  4 ADE
>>     117 3  4 ADF
>>     118 3  4 ADF
>>     119 3  4 ADF
>>     120 3  4 ADF
>>     121 3  4 AEF
>>     122 3  4 AEF
>>     123 3  4 AEF
>>     124 3  4 AEF
>>     125 3  0 BCD
>>     126 3  0 BCD
>>     127 3  0 BCD
>>     128 3  0 BCD
>>     129 3  0 BCE
>>     130 3  0 BCE
>>     131 3  0 BCE
>>     132 3  0 BCE
>>     133 3  0 BCF
>>     134 3  0 BCF
>>     135 3  0 BCF
>>     136 3  0 BCF
>>     137 3  0 BDE
>>     138 3  0 BDE
>>     139 3  0 BDE
>>     140 3  0 BDE
>>     141 3  0 BDF
>>     142 3  0 BDF
>>     143 3  0 BDF
>>     144 3  0 BDF
>>     145 3  0 BEF
>>     146 3  0 BEF
>>     147 3  0 BEF
>>     148 3  0 BEF
>>     149 3  0 CDE
>>     150 3  0 CDE
>>     151 3  0 CDE
>>     152 3  0 CDE
>>     153 3  0 CDF
>>     154 3  0 CDF
>>     155 3  0 CDF
>>     156 3  0 CDF
>>     157 3  0 CEF
>>     158 3  0 CEF
>>     159 3  0 CEF
>>     160 3  0 CEF
>>     161 3  0 DEF
>>     162 3  0 DEF
>>     163 3  0 DEF
>>     164 3  0 DEF
>>
>>
> --
> Michael
> http://www.dewey.myzen.co.uk/home.html
>



-- 
Yours sincerely,
Justin

*I check my email at 9AM and 4PM everyday*
*If you have an EMERGENCY, contact me at +447938674419(UK) or
+60125056192(Malaysia)*

	[[alternative HTML version deleted]]



More information about the R-help mailing list