1.
If you would like to use your way, you can reduce the steps by assigning the
common factor levels to all three variables in step 1).
Then, all the tables will have the same dimension, so you can completely
skip step 3) and 4).
Then, the code would be slightly more general but not general enough,
though.
For example, you need to change your code again if you have 4 or 2 variables
instead of 3.
2.
I guess the method in my previous reply (the main part consists of 3 lines
of code) will do exactly the same job you are doing.
Let me know if it is not the case. :)
For example, the main part of the code can be written as a simple function
as follows.
## Function definition
table2 <- function(x.df, x.vec){
x.df2 <- unlist(x.df)
x.vec2 <- rep(x.vec, ncol(x.df))
table(x.df2, x.vec2)
}
## Use the function to make the table
table2(bamas[,c("P2_A", "P2_B", "P2_C")], bamas$aglomera)
Of course, to be through, you need to put some code to check whether
- x.df is a data.frame (with at least 2 variables)
- the number of rows of x.df is equal to the length of x.vec
and so on.
On Dec 3, 2007 4:55 AM, Alejandro Rodríguez <[EMAIL PROTECTED]>
wrote:
> Hello everyone. I want to thank you for your response.
>
> I tried this on weekend to attack my problem:
>
>
> 1) Turned into factors all my variables to tabulate:
>
> infec1<-bamas$P2_A
> infec2<-bamas$P2_B
> infec3<-bamas$P2_C
> aglomera<-bamas$QCL_1
>
> Where bamas is my data frame, P2_A, P2_B and P2_C the infections
> and QCL_1
> the clusters.
>
> 2) Defined a table for each one of the variables I wanted:
>
> t1<-table(infec1,aglomera)
> t2<-table(infec2,aglomera)
> t3<-table(infec3,aglomera)
>
> 3) Tables generated above are of n1 x 4, n2 x 4, n3 x 4 where n1, n2 and
> n3
> are different and so the dimmensions of my matrices, then if I want to sum
>
> the three tables R showed me the error that they were of
> different dimmensions, so I decided to make them of the same
> dimmensions by "filling the holes" (missing rows) with zeros of each row
> in
> each table in order to transform them into a matrix of N x 4:
>
> For example, comparing my table t1 with t2, I could see that, in table
> 1 there where missing rows 7, 10, 12 and 13, and in the case of t3 rows 7
> and 12 ;
>
> t1
>
> aglomera
> infec1 1 2 3 4
> 1 117 88 76 83
> 2 10 10 9 7
> 3 15 11 14 14
> 4 2 0 1 1
> 5 2 3 1 0
> 6 1 0 1 0
> 8 3 3 0 1
> 9 3 1 1 0
> 11 0 0 1 1
>
> t3
>
> aglomera
> infec2 1 2 3 4
> 1 12 9 6 6
> 2 37 45 33 37
> 3 11 11 8 11
> 4 3 0 0 2
> 5 2 1 0 0
> 6 0 0 0 1
> 8 3 1 2 2
> 9 2 2 0 2
> 10 1 1 2 1
> 11 1 0 0 0
> 13 0 0 1 0
>
> Then I did the following instructions
>
> u<-cbind(0,0,0,0)
> rt1<-rbind(t1[1:6,],u,t1[7:8,],u,t1[9,],u,u)
>
> rt2<-rbind(t2[1:6,],u,t2[7:10,],u,u,t2[11,])
>
> rt3<-rbind(t3[1:6,],u,t3[7:10,],u,u)
>
> 4) Now all my tables are 13 x 4 so I defined as matrix:
>
> mrt1<-matrix(rt1,nrow=13,ncol=4)
> mrt2<-matrix(rt2,nrow=13,ncol=4)
> mrt3<-matrix(rt2,nrow=13,ncol=4)
>
> 5) And then I added them:
>
> total<-mrt1+mrt2+mrt3
> total
>
> [,1] [,2] [,3] [,4]
> [1,] 130 97 83 89
> [2,] 53 55 42 46
> [3,] 33 36 36 39
> [4,] 5 1 2 3
> [5,] 5 4 1 0
> [6,] 1 1 1 3
> [7,] 0 0 0 0
> [8,] 8 6 4 4
> [9,] 6 4 1 2
> [10,] 1 2 2 3
> [11,] 1 1 1 1
> [12,] 0 0 0 0
> [13,] 0 0 1 0
>
> which is the result I wanted.
>
> I don't know if this is the best way to make this kind of tables, but at
> least it worked well on weekend late at night :P
>
> If someone have a better (faster) way to make them please help me.
>
> Warm regards from Mexico.
>
>
>
> -----Mensaje original-----
> De: Petr PIKAL [mailto:[EMAIL PROTECTED]
> Enviado el: Lunes, 03 de Diciembre de 2007 02:30 a.m.
> Para: [EMAIL PROTECTED]
> CC: r-help@r-project.org
> Asunto: Re: [R] Help with tables
>
>
> Hi
>
> I am not sure but if I remember correctly I had seen similar output in
> Frank Harrel's book. So you could check some function from Hmisc package,
> probably summarize.
>
> Regards
>
> Petr
> [EMAIL PROTECTED]
>
> [EMAIL PROTECTED] napsal dne 02.12.2007 03:16:09:
>
> > I guess you can get the result by
> > 1) concatenating all the variables (P2_A, P2_B, P2_C) into one variable
> ,
> > 2) replicating segment membership properly,
> > 3) make the table of 1) and 2)
> >
> > For example, the following may do the job.
> >
> > > ## (1) Generate data set
> > > # Set random seed
> > > set.seed(0)
> > >
> > > n.obs <- 15 # Number of patients
> > > disease.lv <- 1:10 ## different types of disease
> > > data1 <- as.data.frame(t(replicate(n.obs, sample(disease.lv, 3)))) #
> > Create data set
> > > ## Create segment membership
> > > segment <- sample(LETTERS[1:3], n.obs, replace=TRUE)
> > >
> > > cbind(data1, segment)
> > V1 V2 V3 segment
> > 1 9 3 10 B
> > 2 6 9 2 C
> > 3 9 10 6 A
> > 4 7 1 2 B
> > 5 2 7 4 C
> > 6 8 5 6 C
> > 7 10 4 7 B
> > 8 10 2 6 C
> > 9 2 3 4 B
> > 10 1 4 7 A
> > 11 4 5 9 A
> > 12 5 2 7 A
> > 13 7 8 1 A
> > 14 8 4 7 B
> > 15 7 8 5 B
> > >
> > >
> > > ## (2) Table
> > > infec.all <- unlist(data1) # Concatenate all variables into one
> variable
> > > segment.all <- rep(segment, ncol(data1)) # Replicate the segment
> > membership as necessary
> > > table(infec.all, segment.all)
> > segment.all
> > infec.all A B C
> > 1 2 1 0
> > 2 1 2 3
> > 3 0 2 0
> > 4 2 3 1
> > 5 2 1 1
> > 6 1 0 3
> > 7 3 4 1
> > 8 1 2 1
> > 9 2 1 1
> > 10 1 2 1
> >
> >
> >
> > On Nov 30, 2007 10:10 AM, Alejandro Rodríguez
> <[EMAIL PROTECTED]>
> > wrote:
> >
> > > Hello, I'm new using R and developing tables.
> > >
> > > I have a problem in developing a table. In a questionaire I made I
> ask
> > > this
> > > question "Please tell me the first three sympthoms caused by
> Respiratory
> > > tract infection you've caught this year", then the people answer three
>
> > > sympthoms, the first mention (Top of mind) is saved in a variable
> called
> > > "P2_A", the second mention in a variable called "P2_B" and the third
> > > mention
> > > in "P2_C". Each answer is coded with numbers like this:
> > >
> > > 1 = Flu
> > > 2 = Cough
> > > 3 = Asthma
> > > ....
> > >
> > > 13 = Fever
> > >
> > > I've already done a K-cluster analysis and segmented my data base. So
>
> my
> > > first task is to develop tables to develop tables of frequencies
> crossing
> > > Cluster vs. "P2_A" in order to know which are the top of mind products
> and
> > > their frequencies by cluster, then the second mention and third
> mention.
> > > I've used this instruction which worked well:
> > >
> > > > table(infec1,aglomera)
> > > aglomera
> > > infec1 1 2 3 4
> > > 1 117 88 76 83
> > > 2 10 10 9 7
> > > 3 15 11 14 14
> > > 4 2 0 1 1
> > > 5 2 3 1 0
> > > 6 1 0 1 0
> > > 8 3 3 0 1
> > > 9 3 1 1 0
> > > 11 0 0 1 1
> > >
> > > Where "infec1" is a factor of "P2_A" and "aglomera" is a factor of the
> > > variable "Cluster" I made. It worked well when I study them
> > > separately...however I would like to know the TOTAL mentions of each
> > > sympthom by cluster. I've done this exercise in SPSS using the
> "Multiple
> > > Response" instruction first grouping my three variables ( i.e. "P2_A",
> > > "P2_B"
> > > and "P2_C") into a variable called "sick" and cross tabulating it vs.
> > > QCL_1
> > > (my cluster variable) and it gave me the table I need in this way
> (showed
> > > at
> > > the bottom of this mail):
> > >
> > > How can I made a table like this in R???. I've tried combining my
> > > variables
> > > in a matrix and using xtabs, ftable, table, structable and a lot of
> > > combination of them but I haven't had succed with any of them.
> > >
> > > Please help me with this issue, I don't want to keep using SPSS any
> more.
> > >
> > > Thanx in advance.
> > >
> > > P.D. Result from SPSS is shown below.
> > >
> > >
> > >
> > > * * * C R O S S T A B U L A T I O N * * *
> > >
> > > $SICK (group) mr sick
> > > by QCL_1 Cluster Number of Case
> > >
> > >
> > > QCL_1
> > >
> > > Count I
> > > I Row
> > > I Total
> > > I 1 I 2 I 3 I 4 I
> > > $SICK --------+--------+--------+--------+--------+
> > > 1 I 130 I 97 I 83 I 89 I 399
> > > Gripe, influenza, ca I I I I I 83.1
> > > +--------+--------+--------+--------+
> > > 2 I 53 I 55 I 42 I 46 I 196
> > > Tos de cualquier tip I I I I I 40.8
> > > +--------+--------+--------+--------+
> > > 3 I 33 I 36 I 36 I 39 I 144
> > > Dolor irritación I I I I I 30.0
> > > +--------+--------+--------+--------+
> > > 4 I 5 I 1 I 2 I 3 I 11
> > > Bronquitis I I I I I 2.3
> > > +--------+--------+--------+--------+
> > > 5 I 5 I 4 I 1 I 0 I 10
> > > Sinusitis I I I I I 2.1
> > > +--------+--------+--------+--------+
> > > 6 I 1 I 1 I 1 I 3 I 6
> > > Rinitis I I I I I 1.3
> > > +--------+--------+--------+--------+
> > > 8 I 8 I 6 I 4 I 4 I 22
> > > Amigdalitis I I I I I 4.6
> > > +--------+--------+--------+--------+
> > > 9 I 6 I 4 I 1 I 2 I 13
> > > Faringitis I I I I I 2.7
> > > +--------+--------+--------+--------+
> > > 10 I 1 I 2 I 2 I 3 I 8
> > > Laringitis I I I I I 1.7
> > > +--------+--------+--------+--------+
> > > 11 I 1 I 1 I 1 I 1 I 4
> > > Neumonia I I I I I .8
> > > +--------+--------+--------+--------+
> > > 13 I 0 I 0 I 1 I 0 I 1
> > > Asma I I I I I .2
> > > +--------+--------+--------+--------+
> > > Column 153 116 104 107 480
> > > Total 31.9 24.2 21.7 22.3 100.0
> > >
> > > Percents and totals based on respondents
> > >
> > > 480 valid cases; 0 missing cases
> > >
> > >
> > >
> > > Act. Calef Alejandro Rodríguez Cuevas
> > > Analista de mercado
> > >
> > > Laboratorios Farmasa S.A. de C.V.
> > > Schwabe Mexico, S.A. de C.V.
> > >
> > > Bufalo Nr. 27
> > > Col. del Valle 03100
> > > Mexico, D.F.
> > > Mexico
> > >
> > > Tel. 52 00 26 80
> > > email: [EMAIL PROTECTED]
> > >
> > > www.schwabe.com.mx
> > > www.umckaloabo.com.mx
> > >
> > > ______________________________________________
> > > R-help@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.
> > >
> >
> >
> >
> > --
> > ======================================
> > T.K. (Tae-kyun) Kim
> > Ph.D . student
> > Department of Marketing
> > Marshall School of Business
> > University of Southern California
> > ======================================
> >
> > [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help@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.
>
> ______________________________________________
> R-help@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.
>
--
======================================
T.K. (Tae-kyun) Kim
Ph.D. student
Department of Marketing
Marshall School of Business
University of Southern California
======================================
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
