Using R, how can I make columns become rows and correlate with values of another column?-CodePudding

Basically I need a new column called "census" containing columns dbh1, dbh3, and dbh4 as sequential rows, correlating with their tree number and species of interest.

The data looks like this (with 10k rows):

Tree	Spec	dbh1	dbh3	dbh4
1	PICO	6	8.8	9
2	ABLA	21	24.1	25.4
3	PICO	12	14.3	15.2
4	PIEN	24	25.5	25.8

I need it to look like this:

Tree	Spec	Census
1	PICO	6
1	PICO	8.8
1	PICO	9
2	ABLA	21
2	ABLA	24.1
2	ABLA	25.4
3	PICO	12
3	PICO	14.3
3	PICO	15.2
4	PIEN	24
4	PIEN	25.5
4	PIEN	25.8

I've tried looking around the tidyverse and I'm sure I'm missing what could be a very simple answer.

CodePudding user response：

You can use pivot_longer() from tidyr to pivot your dataframe, and then remove the new column with the information of which original variable the values from Census came from.

Data

df <-
  structure(list(Tree = 1:4, Spec = c("PICO", "ABLA", "PICO", "PIEN"
  ), dbh1 = c(6L, 21L, 12L, 24L), dbh3 = c(8.8, 24.1, 14.3, 25.5
  ), dbh4 = c(9, 25.4, 15.2, 25.8)), class = "data.frame", row.names = c(NA,-4L))

Code

library(dplyr)
library(tidyr)

df %>% 
  pivot_longer(cols = starts_with("dbh"),values_to = "Census") %>% 
  select(-name)

Output

# A tibble: 12 x 3
    Tree Spec  Census
   <int> <chr>  <dbl>
 1     1 PICO     6  
 2     1 PICO     8.8
 3     1 PICO     9  
 4     2 ABLA    21  
 5     2 ABLA    24.1
 6     2 ABLA    25.4
 7     3 PICO    12  
 8     3 PICO    14.3
 9     3 PICO    15.2
10     4 PIEN    24  
11     4 PIEN    25.5
12     4 PIEN    25.8