CODEWITHSUNDEEP
rsubset
Tyler
Tyler

Reputation: 23

Keep multiple observations for individual intact, when discriminating on variable responses

Each person has multiple observations (rows) in my dataset--as many as 5 per person. I need to pull out a subset of people who have met my criteria during any single one of their observations, but without losing any of their observations. Essentially, it would be like this:

ID    x     
1     1
1     2
1     3
1     9
1     9
2     2
2     7
2     8
3     4
3     1
3     2
3     1

I would subset my data so that only people who had ever recorded a 1 for variable x would be kept. Person two would be excluded, but all observations for persons 1 and 3 would be kept in a new data frame.

ID    x     
1     1
1     2
1     3
1     9
1     9
3     4
3     1
3     2
3     1

Upvotes: 0

Views: 112

Answers (1)

Sotos
Sotos

Reputation: 51592

Here is one way,

ind <- unique(df$ID[df$x==1])
df[df$ID %in% ind, ]
#   ID x
#1   1 1
#2   1 2
#3   1 3
#4   1 9
#5   1 9
#9   3 4
#10  3 1
#11  3 2
#12  3 1

Upvotes: 1

Related Questions