by David Lillis, Ph.D.
In Part 10, let’s look at the aggregate command for creating summary tables using R.
You may have a complex data set that includes categorical variables of several levels, and you may wish to create summary tables for each level of the categorical variable.
For example, your data set may include the variable Gender, a two-level categorical variable with levels Male and Female. Your data set may include other categorical variables such as Ethnicity, Hair Colour, the Treatments received by patients in a medical study, or the number of cylinders in motor vehicles.
In any case, you may wish to produce summary statistics for each level of the categorical variable. This is where the aggregate command is so helpful.
Here is a data set of patients receiving medical treatment (A, B or C). We have data on their gender, their body mass in kg, whether or not they exercise, whether or not they smoke, and whether or not they recovered after treatment. Cut and paste the following data set into R.
attach(patients)
patients
GENDER TREATMENT MASS SMOKE EXERCISE RECOVER
1 M A 79 Y Y Y
2 M A 87 Y Y Y
3 F B 65 N N Y
4 F A 58 Y N N
5 F B 72 N N N
6 M B 95 N N Y
7 M C 76 N Y Y
8 F B 56 N Y N
9 M A 77 Y Y N
10 M A 104 N N Y
11 F B 67 N Y N
12 F B 82 Y Y N
13 F B 59 N N N
14 M A 68 Y N N
15 M C 79 N Y Y
16 M B 125 Y N Y
17 F B 83 N N N
18 F A 63 Y N N
19 M C 57 N N Y
20 M A 84 Y N Y
21 M A 72 Y Y Y
22 F B 68 N N Y
23 F C 65 N N N
24 F B 64 N N Y
25 M B 87 Y N Y
26 M B 92 N Y Y
27 F C 56 Y N Y
Let’s use the aggregate command to obtain a table of mean body mass across the two levels of gender. Note the syntax. The continuous variable becomes the first argument. Then the categorical variable appears inside the list command. Finally, the function you wish to apply (in this case you want the mean) becomes the third argument.
Table1 <- aggregate(MASS, list(GENDER), FUN=mean)
Table1
Group.1 x
1 F 66.00000
2 M 84.42857
Note that the aggregate command does not return the variable names. Anyway – now we use the aggregate command to obtain a table of mean body mass across the two levels of smoker (i.e. people who smoke and people who do not).
Table2 <- aggregate(MASS, list(SMOKE), FUN=mean)
Table2
Group.1 x
1 N 73.46667
2 Y 78.16667
The aggregate command allows us to create more complex tables, across the levels of several categorical variables.
Table3 <- aggregate(MASS, list(GENDER, TREATMENT),FUN=mean)
Table3
Group.1 Group.2 x
1 F A 60.50000
2 M A 81.57143
3 F B 68.44444
4 M B 99.75000
5 F C 60.50000
6 M C 70.66667
Finally, let’s look at maximum body mass across the levels of gender, smoker and treatment.
Table4 <- aggregate(MASS, list(GENDER, SMOKE, TREATMENT),FUN=max)
Table4
Group.1 Group.2 Group.3 x
1 M N A 104
2 F Y A 63
3 M Y A 87
4 F N B 83
5 M N B 95
6 F Y B 82
7 M Y B 125
8 F N C 65
9 M N C 79
10 F Y C 56
That wasn’t so hard!
About the Author: David Lillis has taught R to many researchers and statisticians. His company, Sigma Statistics and Research Limited, provides both on-line instruction and face-to-face workshops on R, and coding services in R. David holds a doctorate in applied statistics.
{ 6 comments… read them below or add one }
In this example, how is the data used by aggregate command called forth (in this example “patients”)? In the aggregate() command there is no argument for data=patients. As expected, this doesn’t work when I try it.
Use the structure dataset$column, for example patients$MASS.
Make sure to use it for each variable, for example:
aggregate(patients$MASS, list(patients$GENDER), FUN=mean)
Hm, reading the table in, so that it shows up, does not add up to functionality.
Here’s cut-and-paste:
> read.table(“patients.txt”)
GENDER TREATMENT MASS SMOKE EXERCISE RECOVER
1 M A 79 Y Y Y
2 M A 87 Y Y Y
3 F B 65 N N Y
4 F A 58 Y N N
5 F B 72 N N N
6 M B 95 N N Y
7 M C 76 N Y Y
8 F B 56 N Y N
9 M A 77 Y Y N
10 M A 104 N N Y
11 F B 67 N Y N
12 F B 82 Y Y N
13 F B 59 N N N
14 M A 68 Y N N
15 M C 79 N Y Y
16 M B 125 Y N Y
17 F B 83 N N N
18 F A 63 Y N N
19 M C 57 N N Y
20 M A 84 Y N Y
21 M A 72 Y Y Y
22 F B 68 N N Y
23 F C 65 N N N
24 F B 64 N N Y
25 M B 87 Y N Y
26 M B 92 N Y Y
27 F C 56 Y N Y
> Table1 <- aggregate(MASS, list(GENDER), FUN=mean)
Error in aggregate(MASS, list(GENDER), FUN = mean) :
object 'MASS' not found
Got it to work, but not trivial on a mac if not using a terminal; here are the steps I used.
1. opened textedit, pasted the table into textedit.
2. saved as RTF (the current most basic option, there is no longer a .txt save option).
3. changed the file extension to .txt manually, clicking the “OK” on the warning.
4. re-opened, highlighted and used the “make plain text” command (doodle-shift-T, or under “format”).
note, if you “make plain” before the first save, you do get many plain text unicode file saving options…but not knowing what to use, I took the long way around, and let the machine decide.
5. made a file under “Users” and [the folder with my username] called “Rfiles” and in that file made “DataFiles” and moved the text file there.
6. at the prompt used
> getwd ()
this told me I was in Users/[the folder with my username]
7. used setwd in two steps (it was being balky when I tried to set the path otherwise) and checked it after each step. namely,
>setwd(“Rfiles”)
>getwd()
[here it gave me output confirming I was in Rfiles]
>setwd(“DataFiles”)
>getwd()
[here it gave me output confirming I was in DataFiles]
8. finally read in the table, remembering to use the quotation marks
>read.table(“patients.txt”)
It then reads in and displays the table as it should.
There are no doubt faster ways, but several that I tried did not work.
You show the content of “patients”, but you don’t shown how that is obtained.
Is this a default set in R?
Save the data in a text file, then use read.table:
E.g. (remembering to use forward slash or double-back slash)
datapath<-"C:\\Users\\…..\\R files\\data files\\patients.txt"
patients<-read.table(datapath, sep="", header=T, as.is=T)