CODEWITHSUNDEEP
r
Ja Mah
Ja Mah

Reputation: 29

replace one column vector based on another column condition

Hi I am a beginner in R, I have a df named di_time as follow:

A tibble: 3,037 x 4
Groups:   id, trial [1,519]
       id trial first_block  time
    <int> <int> <lgl>       <dbl>
 1 866093     1 FALSE       145. 
 2 866093     1 TRUE        114. 
 3 866093     2 FALSE       125. 
 4 866093     2 TRUE         60.4
 5 866093     3 FALSE       107. 
 6 866093     3 TRUE         19.1
 7 866093     4 FALSE        86.9
 8 866093     4 TRUE         10.2
 9 866093     5 FALSE        66.2
10 866093     5 TRUE         33.9
with 3,027 more rows

I would like to replace all trial x to trial (x+6) if first_block== FALSE and first_block== TRUE remain the same.

Upvotes: 0

Views: 56

Answers (3)

ThomasIsCoding
ThomasIsCoding

Reputation: 102880

Here is a solution with base R, where within() and ifelse() are used:

df <- within(df,trial <- ifelse(!first_block, trial+6, trial))

or 

df <- within(df,trial <- trial + ifelse(!first_block, 6, 0))

such that

> df
       id trial first_block  time
1  866093     7       FALSE 145.0
2  866093     1        TRUE 114.0
3  866093     8       FALSE 125.0
4  866093     2        TRUE  60.4
5  866093     9       FALSE 107.0
6  866093     3        TRUE  19.1
7  866093    10       FALSE  86.9
8  866093     4        TRUE  10.2
9  866093    11       FALSE  66.2
10 866093     5        TRUE  33.9

DATA

df <- structure(list(id = c(866093L, 866093L, 866093L, 866093L, 866093L, 
                            866093L, 866093L, 866093L, 866093L, 866093L), trial = c(1L, 1L, 
                                                                                    2L, 2L, 3L, 3L, 4L, 4L, 5L, 5L), first_block = c(FALSE, TRUE, 
                                                                                                                                     FALSE, TRUE, FALSE, TRUE, FALSE, TRUE, FALSE, TRUE), time = c(145, 
                                                                                                                                                                                                   114, 125, 60.4, 107, 19.1, 86.9, 10.2, 66.2, 33.9)), row.names = c(NA, 
                                                                                                                                                                                                                                                                      -10L), class = c("data.frame"))

Upvotes: 0

Sotos
Sotos

Reputation: 51612

A base R approach can be,

with(df, ave(trial, id, FUN = function(i)replace(i, first_block[1] == FALSE, i+6)))
#[1]  7  7  8  8  9  9 10 10 11 11

Upvotes: 0

Ronak Shah
Ronak Shah

Reputation: 389325

You could do

df$trial1 <- df$trial + ((!df$first_block) * 6)
df

#       id trial first_block  time trial1
#1  866093     1       FALSE 145.0      7
#2  866093     1        TRUE 114.0      1
#3  866093     2       FALSE 125.0      8
#4  866093     2        TRUE  60.4      2
#5  866093     3       FALSE 107.0      9
#6  866093     3        TRUE  19.1      3
#7  866093     4       FALSE  86.9     10
#8  866093     4        TRUE  10.2      4
#9  866093     5       FALSE  66.2     11
#10 866093     5        TRUE  33.9      5

This adds 6 when first_block == FALSE and keeps the value as it is for TRUE value.

data

df <- structure(list(id = c(866093L, 866093L, 866093L, 866093L, 866093L, 
866093L, 866093L, 866093L, 866093L, 866093L), trial = c(1L, 1L, 
2L, 2L, 3L, 3L, 4L, 4L, 5L, 5L), first_block = c(FALSE, TRUE, 
FALSE, TRUE, FALSE, TRUE, FALSE, TRUE, FALSE, TRUE), time = c(145, 
114, 125, 60.4, 107, 19.1, 86.9, 10.2, 66.2, 33.9)), class = 
"data.frame", row.names = c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10"))

Upvotes: 2

Related Questions