https://training-course-material.com/index.php?action=history&feed=atom&title=R_-_Grouping_Data
R - Grouping Data - Revision history
2026-05-14T01:15:28Z
Revision history for this page on the wiki
MediaWiki 1.45.1
https://training-course-material.com/index.php?title=R_-_Grouping_Data&diff=52345&oldid=prev
Daniel Rodriguez at 05:24, 14 February 2017
2017-02-14T05:24:29Z
<p></p>
<p><b>New page</b></p><div>[[Category:Intro to R|030]]<br />
<br />
Find the average salary in the whole company.<br />
> mean(emp$SAL)<br />
[1] 2073.214<br />
<br />
How many employees are in the database?<br />
> length(emp$ID)<br />
[1] 14<br />
<br />
Other salary statistics (minimal, maximal, sum, etc.)<br />
> sd(emp$SAL)<br />
[1] 1182.503<br />
> var(emp$SAL)<br />
[1] 1398314<br />
> min(emp$SAL)<br />
[1] 800<br />
> max(emp$SAL)<br />
[1] 5000<br />
> sum(emp$SAL)<br />
[1] 29025<br />
<br />
> summary(emp)<br />
ID ENAME JOB MGR HIREDATE SAL COMM DEPTNO <br />
Min. :7369 Length:14 ANALYST :2 Min. :7566 Min. :1980-12-17 Min. : 800 Min. : 0 Min. :10.00 <br />
1st Qu.:7588 Class :character CLERK :4 1st Qu.:7698 1st Qu.:1981-04-09 1st Qu.:1250 1st Qu.: 225 1st Qu.:20.00 <br />
Median :7785 Mode :character MANAGER :3 Median :7698 Median :1981-09-18 Median :1550 Median : 400 Median :20.00 <br />
Mean :7727 PRESIDENT:1 Mean :7739 Mean :1981-09-29 Mean :2073 Mean : 550 Mean :22.14 <br />
3rd Qu.:7868 SALESMAN :4 3rd Qu.:7839 3rd Qu.:1981-12-03 3rd Qu.:2944 3rd Qu.: 725 3rd Qu.:30.00 <br />
Max. :7934 Max. :7902 Max. :1983-01-12 Max. :5000 Max. :1400 Max. :30.00 <br />
NA's :1 NA's :10<br />
The median for department number is pretty much useless. It is better to treat DEPTNO as a factor<br />
<br />
<br />
> emp$DEPTNO <- as.factor(DEPTNO)<br />
<br />
> summary(emp)<br />
ID ENAME JOB MGR HIREDATE SAL COMM DEPTNO<br />
Min. :7369 Length:14 ANALYST :2 Min. :7566 Min. :1980-12-17 Min. : 800 Min. : 0 10:3 <br />
1st Qu.:7588 Class :character CLERK :4 1st Qu.:7698 1st Qu.:1981-04-09 1st Qu.:1250 1st Qu.: 225 20:5 <br />
Median :7785 Mode :character MANAGER :3 Median :7698 Median :1981-09-18 Median :1550 Median : 400 30:6 <br />
Mean :7727 PRESIDENT:1 Mean :7739 Mean :1981-09-29 Mean :2073 Mean : 550 <br />
3rd Qu.:7868 SALESMAN :4 3rd Qu.:7839 3rd Qu.:1981-12-03 3rd Qu.:2944 3rd Qu.: 725 <br />
Max. :7934 Max. :7902 Max. :1983-01-12 Max. :5000 Max. :1400 <br />
NA's :1 NA's :10<br />
<br />
=== Grouping ===<br />
Groups contain all rows which have the same values in a column or columns. The grouping columns are called ''factors'' in R.<br />
<br />
Find averages in specific jobs<br />
> tapply(SAL,JOB,mean)<br />
ANALYST CLERK MANAGER PRESIDENT SALESMAN <br />
3000.000 1037.500 2758.333 5000.000 1400.000<br />
<br />
You can group by more than one factor<br />
> tapply(SAL,list(JOB,DEPTNO),mean)<br />
10 20 30<br />
ANALYST NA 3000 NA<br />
CLERK 1300 950 950<br />
MANAGER 2450 2975 2850<br />
PRESIDENT 5000 NA NA<br />
SALESMAN NA NA 1400<br />
<br />
Example with three factors<br />
> tapply(SAL,list(JOB,DEPTNO,format(emp$HIREDATE,"%Y")),mean)<br />
, , 1980<br />
<br />
10 20 30<br />
ANALYST NA NA NA<br />
CLERK NA 800 NA<br />
MANAGER NA NA NA<br />
PRESIDENT NA NA NA<br />
SALESMAN NA NA NA <br />
<br />
, , 1981<br />
<br />
10 20 30<br />
ANALYST NA 3000 NA<br />
CLERK NA NA 950<br />
MANAGER 2450 2975 2850<br />
PRESIDENT 5000 NA NA<br />
SALESMAN NA NA 1400 <br />
<br />
, , 1982<br />
<br />
10 20 30<br />
ANALYST NA 3000 NA<br />
CLERK 1300 NA NA<br />
MANAGER NA NA NA<br />
PRESIDENT NA NA NA<br />
SALESMAN NA NA NA <br />
<br />
, , 1983<br />
<br />
10 20 30<br />
ANALYST NA NA NA<br />
CLERK NA 1100 NA<br />
MANAGER NA NA NA<br />
PRESIDENT NA NA NA<br />
SALESMAN NA NA NA<br />
<br />
== Grouping with data.table package ==<br />
<source lang="rsplus"><br />
install.packages("data.table")<br />
library("data.table")<br />
# Convert<br />
emp.dt = data.table(emp)<br />
# Group<br />
emp.dt[,sum(SAL),by=JOB]<br />
</source><br />
JOB V1<br />
[1,] PRESIDENT 5000<br />
[2,] MANAGER 8275<br />
[3,] SALESMAN 5600<br />
[4,] CLERK 4150<br />
[5,] ANALYST 6000<br />
<br />
==Exercises==<br />
<br />
'''23. Exercise'''<br />
<br />
Find the minimal, maximal and average salaries in the whole company.<br />
Minimal Maximal Mean<br />
[1,] 800 5000 2073.214<br />
<br />
'''24. Exercise'''<br />
<br />
Find the difference between maximal and minimal salary.<br />
[1] 4200<br />
<br />
<br />
'''25. Exercise'''<br />
<br />
Find the average salary for every post.<br />
ANALYST CLERK MANAGER PRESIDENT SALESMAN <br />
3000.000 1037.500 2758.333 5000.000 1400.000 <br />
<br />
<br />
'''26. Exercise'''<br />
<br />
How many managers work for the company?<br />
MANAGER <br />
3 <br />
<br />
<br />
'''27. Exercise'''<br />
<br />
Find the average annual salaries in departments.<br />
10 20 30 <br />
35000 26100 18800 <br />
<br />
<br />
'''28. Exercise'''<br />
<br />
Find departments with more than 3 workers.<br />
20 30 <br />
5 6 <br />
<br />
'''29. *Exercise'''<br />
<br />
Check that every employees id is unique.<br />
<br />
'''30. *Exercise'''<br />
<br />
For each of the managers find a minimum salary of their subordinates.<br />
<br />
+------+----------+<br />
| mgr | min(sal) |<br />
+------+----------+<br />
| NULL | 5000.00 |<br />
| 7566 | 3000.00 |<br />
| 7698 | 950.00 |<br />
| 7782 | 1300.00 |<br />
| 7788 | 1100.00 |<br />
| 7839 | 2450.00 |<br />
| 7902 | 800.00 |<br />
+------+----------+</div>
Daniel Rodriguez