Ifelse statement by participant R-CodePudding

I have a dataset like this, where Count is the sequence of 1s for each participant:

Participant TrialNumber Correct    Count
     118       1          1           1
     118       2          1           2
     118       3          1           3
     118       4          1           4
     118       5          1           5
     120       1          1           1
     120       2          0           0
     120       3          0           0
     120       4          1           1
     120       5          1           2
     121       1          1           1
     121       2          1           2
     121       3          1           3
     121       4          1           4
     121       5          0           0

I need to find all trial numbers where the count is 4, and if a participant doesn't have a count of 4 (i.e. participant 120) to write 0 instead.

This is the code I have so far:

df<-df[with(df, !(Count> 4)),] 
df$Performance<- ifelse(df$Count==4, df$TrialNumber, NA)
df_Performance<-aggregate(Performance~Participant, data=df, first)

The only problem with this, is that I lose any participants who don't have a count of 4.

I'm thinking I need to add another ifelse statement to the second line of code saying if there are no 4s for that participant then write 0 for the first trial, then for any other trials write NA.

I'm stuck on how to write this next part of the statement, any suggestions on how to do this would be much appreciated, thank you

CodePudding user response：

library(tidyverse)
data <-
  tibble::tribble(
    ~Participant, ~TrialNumber, ~Correct, ~Count,
    118, 1, 1, 1,
    118, 2, 1, 2,
    118, 3, 1, 3,
    118, 4, 1, 4,
    118, 5, 1, 5,
    120, 1, 1, 1,
    120, 2, 0, 0,
    120, 3, 0, 0,
    120, 4, 1, 1,
    120, 5, 1, 2,
    121, 1, 1, 1,
    121, 2, 1, 2,
    121, 3, 1, 3,
    121, 4, 1, 4,
    121, 5, 0, 0
  )

data %>%
  group_by(Participant) %>%
  mutate(
    has_four = 4 %in% Count
  ) %>%
  distinct(has_four) %>%
  transmute(
    Participant,
    performance = ifelse(has_four, 4, 0)
  )
#> # A tibble: 3 × 2
#> # Groups:   Participant [3]
#>   Participant performance
#>         <dbl>       <dbl>
#> 1         118           4
#> 2         120           0
#> 3         121           4

^{Created on 2021-09-16 by the reprex package (v2.0.1)}

CodePudding user response：

Here is another tidyverse approach. After grouping by Participant, use summarise and designate the new column to include the TrialNumber where the Count is 4. If none of the trials have a 4 (using any), then include 0. This would assume you have only one trial of Count equaling 4 for any participant.

library(tidyverse)

df %>%
  group_by(Participant) %>%
  summarise(Performance = ifelse(any(Count == 4), TrialNumber[Count == 4], 0))

Output

  Participant Performance
        <int>       <dbl>
1         118           4
2         120           0
3         121           4

Data

df <- structure(list(Participant = c(118L, 118L, 118L, 118L, 118L, 
120L, 120L, 120L, 120L, 120L, 121L, 121L, 121L, 121L, 121L), 
    TrialNumber = c(1L, 2L, 3L, 4L, 5L, 1L, 2L, 3L, 4L, 5L, 1L, 
    2L, 3L, 4L, 5L), Correct = c(1L, 1L, 1L, 1L, 1L, 1L, 0L, 
    0L, 1L, 1L, 1L, 1L, 1L, 1L, 0L), Count = c(1L, 2L, 3L, 4L, 
    5L, 1L, 0L, 0L, 1L, 2L, 1L, 2L, 3L, 4L, 0L)), class = "data.frame", row.names = c(NA, 
-15L))

CodePudding user response：

Here's a base R approach.

data$Performance <- ifelse(data$Count == 4, data$TrialNumber, 0)
aggregate(x = data$Performance, by = list(Participant = data$Participant), FUN = sum)

#>   Participant x
#> 1         118 4
#> 2         120 0
#> 3         121 4

CodePudding user response：

Using base R aggregate with if/else.

aggregate(Count~Participant, df, function(x) if(any(x == 4)) 4 else 0)

#  Participant Count
#1         118     4
#2         120     0
#3         121     4