CODEWITHSUNDEEP
r
bulba
bulba

Reputation: 57

Find number of occurrences in different variables

I have a data frame with some data about different people. They look like this:

Year Item ID
2005   a   1234
2005   b   1234
2005   a   4567
2005   b   4567
2006   a   4567
2006   a   7894

My data has 45000 observations and about 1000 different ID's and 10 different years. I want to find the ID's of people that appear in more than 1 year, how do i do this? I thought of separating the data by ID and seeing if the resulting data has different years, but that doesn´t seem like the smartest way to do that

Upvotes: 2

Views: 37

Answers (3)

NelsonGon
NelsonGon

Reputation: 13319

We can get the duplicated IDs and then get those duplicated within these:

Dups<-df[duplicated(df$ID),]
 Dups[duplicated(Dups$ID),]["ID"]
 # ID
 # 5 4567

Upvotes: 1

d.b
d.b

Reputation: 32548

split the Year by ID and then keep only those sub-groups that have more than one unique Year

list1 = lapply(split(df1$Year, df1$ID), unique)
list1 = list1[lengths(list1) > 1]
data.frame(ID = names(list1), count = lengths(list1))
#       ID count
#4567 4567     2

#DATA
df1 = structure(list(Year = c(2005L, 2005L, 2005L, 2005L, 2006L, 2006L),
                     Item = c("a", "b", "a", "b", "a", "a"), ID = c(1234L, 1234L, 4567L, 4567L, 4567L, 7894L)), 
                class = "data.frame", 
                row.names = c(NA,  -6L))

Upvotes: 1

Ronak Shah
Ronak Shah

Reputation: 388982

With dplyr we can use n_distinct and get only those ID's which have more than 1 year. 

library(dplyr)

df %>%
  group_by(ID) %>%
  filter(n_distinct(Year) > 1) %>%
  pull(ID) %>%
  unique

#[1] 4567

A base R alternative with table

unique(df$ID)[rowSums(table(df$ID, df$Year) > 0) > 1]
#[1] 4567

Upvotes: 1

Related Questions