← SASnRFirst dot concept

Introduction

Many analyses ask for the first record of each subject, first visit of each subject, first event of each term — the idea of picking the first row inside a group.
In SAS, once the data is sorted by the grouping variable, the automatic first. flag becomes TRUE on the first row of each group.
Tidyverse reaches the same idea with group_by() + slice(1) or row_number().
Base R uses !duplicated(group) after sorting to keep the first occurrence of each group.
This lesson shows two small examples on CLASS: a within-group counter, and the lowest-Height subject within each Sex.
Both examples rely on sorting first — the definition of "first" only makes sense after we pick an order.

First dot concept

SAS (Base SAS)+

data CLASS;
 infile datalines dlm='|' dsd missover;
input Name : $8. Sex : $1. Age : best32. Height : best32. Weight : best32.;
label ;
 format ;
 datalines4;
Alfred|M|14|69|112.5
Alice|F|13|56.5|84
Barbara|F|13|65.3|98
Carol|F|14|62.8|102.5
Henry|M|14|63.5|102.5
James|M|12|57.3|83
Jane|F|12|59.8|84.5
Janet|F|15|62.5|112.5
Jeffrey|M|13|62.5|84
John|M|12|59|99.5
Joyce|F|11|51.3|50.5
Judy|F|14|64.3|90
Louise|F|12|56.3|77
Mary|F|15|66.5|112
Philip|M|16|72|150
Robert|M|12|64.8|128
Ronald|M|15|67|133
Thomas|M|11|57.5|85
William|M|15|66.5|112
;;;;
 run;

*------------------------------------------------------------------------------;
*counter;
*------------------------------------------------------------------------------;

proc sort data=class;
     by sex height weight name;
 run;

data counter;
     set class;
     by sex height weight name;
     if first.sex then counter=1;
    else counter+1;
run;

*------------------------------------------------------------------------------;
*subset of lowest height;
*------------------------------------------------------------------------------;

proc sort data=class;
     by sex height;
 run;

data lowestheight;
     set class;
     by sex height;
     if first.sex;
     keep sex height name;
 run;

In SAS, once a dataset is sorted by a variable, the DATA step exposes automatic flags called first.var and last.var when that variable is named in a by statement.
first.sex is TRUE on the first record of each Sex group and FALSE otherwise — exactly the signal we want for a within-group counter.
For the counter example, proc sort data=class; by sex height weight name; sets a stable order, then the DATA step reads CLASS with that same by list.
if first.sex then counter=1; else counter+1; uses the sum statement form — counter+1 automatically retains the counter across rows, so we never have to write an explicit retain.
For the lowest-Height example, sorting by sex height and writing if first.sex; on its own line keeps only the first row of each Sex — the one with the smallest Height.
After running, COUNTER has an incrementing counter within each Sex; LOWESTHEIGHT has one row per Sex with the smallest Height.

R (tidyverse)+

library(tidyverse) 
class<-tribble( 
~Name,~Sex,~Age,~Height,~Weight, 
"Alfred","M",14,69,112.5, 
"Alice","F",13,56.5,84, 
"Barbara","F",13,65.3,98, 
"Carol","F",14,62.8,102.5, 
"Henry","M",14,63.5,102.5, 
"James","M",12,57.3,83, 
"Jane","F",12,59.8,84.5, 
"Janet","F",15,62.5,112.5, 
"Jeffrey","M",13,62.5,84, 
"John","M",12,59,99.5, 
"Joyce","F",11,51.3,50.5, 
"Judy","F",14,64.3,90, 
"Louise","F",12,56.3,77, 
"Mary","F",15,66.5,112, 
"Philip","M",16,72,150, 
"Robert","M",12,64.8,128, 
"Ronald","M",15,67,133, 
"Thomas","M",11,57.5,85, 
"William","M",15,66.5,112, 
) 

counter<-class %>% 
arrange(Sex,Height,Weight,Name) 
%>% group_by(Sex) 
%>% mutate(counter=row_number()) 

lowestheight<-class %>% 
arrange(Sex,Height) %>% 
group_by(Sex) %>% 
slice(1) %>% 
select(Name,Sex,Height)

In tidyverse, the "first row within each group" idea is expressed with group_by() combined with either row_number() or slice(1).
For the counter: arrange(Sex, Height, Weight, Name) %>% group_by(Sex) %>% mutate(counter = row_number()) assigns 1, 2, 3, ... within each Sex group after sorting.
row_number() inside a grouped mutate is the direct tidyverse equivalent of the SAS first.sex then counter=1; else counter+1; pattern.
For the lowest-Height example: arrange(Sex, Height) %>% group_by(Sex) %>% slice(1) %>% select(Name, Sex, Height) returns one row per Sex after sorting by Height ascending.
slice(1) keeps the first row in the current group; combined with the earlier sort, that is the row with the smallest Height.
We run and confirm the counter climbs from 1 within each Sex, and the lowest-Height tibble has one row per Sex with the smallest Height value.

Happy learning!

Search lessons

Introduction

First dot concept