(R) functions for tables/ table(), ftable(),addmargins(), prop.table(), margin.table()

<table()>

table function is generally used for the frequency of each elements in variables. 

*table of a variable (a variable: gender)

> x = c(rep("M", 3), rep("F", 2), rep(NA, 3))

> x

[1] "M" "M" "M" "F" "F" NA  NA  NA 

> table(x)

x

F M 

2 3 

Without 'exclude=NULL' we reject missing values automatically. If you want to add NA in the table, you can add 'exclude=NULL'

Include NA in table	Not include NA in table
>table(x, exclude = NULL)  #To include the missing values.     x        F    M <NA>         2    3    3	> table(x) x F M  2 3
> as.data.frame(table(x,exclude=NULL))      x Freq 1    F    2 2    M    3 3 <NA>    3	> as.data.frame(table(x))   x Freq 1 F    2 2 M    3

When you only focus on the frequency of male in the data, 

> table(x)["M"] 

M 

3 

*table of multiple variables

> school=c(rep("A", 2), rep("B", 4), rep("C", 4))

> school

 [1] "A" "A" "B" "B" "B" "B" "C" "C" "C" "C"

> gender= c(rep("M", 2), rep("F", 6), "M","F")

> gender

 [1] "M" "M" "F" "F" "F" "F" "F" "F" "M" "F"

> major=c(rep(c("AMS","CS","BUS"),2),"TSM","AMS","AMS","CS")

> major

 [1] "AMS" "CS"  "BUS" "AMS" "CS"  "BUS" "TSM" "AMS" "AMS" "CS" 

If you have different lengths of variables, you cannot combine those data in table. Example I give below consists of three variables which has the equal length 10. 

> table(school,gender,major)

, , major = AMS

      gender

school F M

     A 0 1

     B 1 0

     C 1 1

, , major = BUS

      gender

school F M

     A 0 0

     B 2 0

     C 0 0

, , major = CS

      gender

school F M

     A 0 1

     B 1 0

     C 1 0

, , major = TSM

      gender

school F M

     A 0 0

     B 0 0

     C 1 0

data.frame(school,gender,major)

as.data.frame(table(school,gender,major))

> data.frame(school,gender,major)    school gender major 1       A      M   AMS 2       A      M    CS 3       B      F   BUS 4       B      F   AMS 5       B      F    CS 6       B      F   BUS 7       C      F   TSM 8       C      F   AMS 9       C      M   AMS 10      C      F    CS

You can make a data frame which includes frequency.  > as.data.frame(table(school,gender,major))    school gender major Freq 1       A      F   AMS    0 2       B      F   AMS    1 3       C      F   AMS    1 4       A      M   AMS    1 5       B      M   AMS    0 6       C      M   AMS    1 7       A      F   BUS    0 8       B      F   BUS    2 9       C      F   BUS    0 10      A      M   BUS    0 11      B      M   BUS    0 12      C      M   BUS    0 13      A      F    CS    0 14      B      F    CS    1 15      C      F    CS    1 16      A      M    CS    1 17      B      M    CS    0 18      C      M    CS    0 19      A      F   TSM    0 20      B      F   TSM    0 21      C      F   TSM    1 22      A      M   TSM    0 23      B      M   TSM    0 24      C      M   TSM    0

<ftable()>

ftable() function display the table consists of multiple variables in a flat form. 

table(school,gender,m	ftable(school, gender,major)
> table(school,gender,major) , , major = AMS       gender school F M      A 0 1      B 1 0      C 1 1 , , major = BUS       gender school F M      A 0 0      B 2 0      C 0 0 , , major = CS       gender school F M      A 0 1      B 1 0      C 1 0 , , major = TSM       gender school F M      A 0 0      B 0 0      C 1 0	> ftable(school,gender,major)               major AMS BUS CS TSM school gender                      A      F              0   0  0   0        M              1   0  1   0 B      F              1   2  1   0        M              0   0  0   0 C      F              1   0  1   1        M              1   0  0   0

<addmargins- add tendencies to table>

> gender= c(rep("M", 2), rep("F", 6), "M","F")

> gender

 [1] "M" "M" "F" "F" "F" "F" "F" "F" "M" "F"

> major=c(rep(c("AMS","CS","BUS"),2),"TSM","AMS","AMS","CS")

> major

 [1] "AMS" "CS"  "BUS" "AMS" "CS"  "BUS" "TSM" "AMS" "AMS" "CS" 

>a=table(gender, major)

> addmargins(a)  -----------add sum of frequencies for each row and column. 

      major

gender AMS BUS CS TSM Sum

   F          2     2     2    1   7

   M         2     0     1   0   3

   Sum     4     2     3   1   10

table(gender, major)	addmargins(table(gender, major))
> table(gender,major)       major gender AMS BUS CS TSM      F   2   2  2   1      M   2   0  1   0	> addmargins(table(gender,major))       major gender AMS BUS CS TSM Sum    F     2   2  2   1   7    M     2   0  1   0   3    Sum   4   2  3   1  10

addmargins(table(gender, major))	addmargins(table(gender, major),margin=1)	addmargins(table(gender, major),margin=2)
> addmargins(table(gender,major))       major gender AMS BUS CS TSM Sum    F          2       2     2    1     7    M         2       0     1    0    3    Sum     4       2      3   1     10	#table +sums for eacn row  > addmargins(table(gender, major),margin=1)        major gender AMS BUS CS TSM    F        2        2     2   1    M       2        0     1   0    Sum   4        2     3    1	#table +sums for each columns         major gender AMS BUS CS TSM Sum      F        2      2      2   1   7      M        2     0      1   0   3

> addmargins(a,FUN=mean)  -----------add mean of frequencies for each row and column. 

Margins computed over dimensions

in the following order:

1: gender

2: major

      major

gender  AMS  BUS   CS  TSM mean

  F         2.00 2.00 2.00 1.00 1.75

  M        2.00 0.00 1.00 0.00 0.75

  mean  2.00 1.00 1.50 0.50 1.25

addmargins(table(gender, major),margin=1,FUN=mean)	addmargins(table(gender, major),margin=2,FUN=mean)
> addmargins(table(gender, major),margin=1,FUN=mean)        major gender AMS BUS  CS TSM   F         2.0 2.0 2.0 1.0   M        2.0 0.0 1.0 0.0   mean  2.0 1.0 1.5 0.5	> addmargins(table(gender, major),margin=2,FUN=mean)        major gender  AMS  BUS   CS  TSM mean      F      2.00 2.00 2.00 1.00 1.75      M     2.00 0.00 1.00 0.00 0.75

addmargins(table(gender, major),margin=c(1,2),FUN=c(mean,sum))  #means for rows, sums for columns	addmargins(table(gender, major),margin=c(2,1),FUN=c(mean,sum)) #means for column, sums for rows
> addmargins(table(gender, major),margin=c(1,2),FUN=c(mean,sum))  Margins computed over dimensions in the following order: 1: gender 2: major       major gender AMS BUS  CS TSM sum   F    2.0 2.0 2.0 1.0 7.0   M    2.0 0.0 1.0 0.0 3.0   mean 2.0 1.0 1.5 0.5 5.0	> addmargins(table(gender, major),margin=c(2,1),FUN=c(mean,sum)) Margins computed over dimensions in the following order: 1: major 2: gender       major gender  AMS  BUS   CS  TSM mean    F   2.00 2.00 2.00 1.00 1.75    M   2.00 0.00 1.00 0.00 0.75    sum 4.00 2.00 3.00 1.00 2.50

<tables for conditional proportions for margin> 

proportion of each categories, you can get 1 by adding up all proportions.

> prop.table(table(gender, major)) 

      major

gender AMS BUS  CS TSM

     F     0.2   0.2   0.2  0.1

     M    0.2   0.0   0.1 0.0

conditional proportion given 'gender' marginal distribution, you can get 1 by adding up all proportions of rows.

#sum of each rows of the table=1

> prop.table(table(gender, major),margin=1)

      major

gender       AMS       BUS        CS       TSM

     F 0.2857143 0.2857143 0.2857143 0.1428571

     M 0.6666667 0.0000000 0.3333333 0.0000000

conditional proportion given 'major' marginal distribution, you can get 1 by adding up all proportions of columns.

sum of each column of the table =1

> prop.table(table(gender, major),margin=2) 

      major

gender       AMS       BUS        CS       TSM

     F 0.5000000 1.0000000 0.6666667 1.0000000

     M 0.5000000 0.0000000 0.3333333 0.0000000

<table for marginal distribution>

 > table(gender, major)

      major

gender AMS BUS CS TSM

     F       2      2     2   1

     M      2      0     1   0

> margin.table(table(gender, major),margin=1) -----marginal distribution of variable in the row ' gender'

gender

F M 

7 3 

> margin.table(table(gender, major),margin=2) -----marginal distribution of variable in the column ' major'

major

AMS BUS  CS TSM 

  4   2   3   1