CODEWITHSUNDEEP
r
user2716568
user2716568

Reputation: 1946

Ifelse statement for multiple, alternating conditions

My data involves an experiment with a dependent variable (DV) collected over several conditions and is structured as follows:

AnalysedData <- data.frame(Participant = c("Bill", "Bill", "Bill", "Harry", "Harry", "Harry", "Paul", "Paul", "Paul"),
                           Code = c(1, 1, 1, 2, 2, 2, 1, 1, 1),
                           Condition = c("Con", "Expr", "Plac", "Con", "Expr", "Plac", "Con", "Expr", "Plac"),
                           DV = c(26.07, 26.06, 26.05, 26.09, 26.04, 26.65, 26.64, 26.62, 26.63))

All the participants received the same conditions but completed two of the conditions in a randomised order - the control condition was always conducted first. Those who completed the experimental (Expr) condition as their second trial have a 1 for Code whilst those who completed the placebo (Plac) as their second trial have a 2 for Code.

I wish to include a new column AnalysedData$Trial whereby the order of trials, for each participant, are included. My preferred output would be:

AnalysedData <- data.frame(Participant = c("Bill", "Bill", "Bill", "Harry", "Harry", "Harry", "Paul", "Paul", "Paul"),
                           Code = c(1, 1, 1, 2, 2, 2, 1, 1, 1),
                           Condition = c("Con", "Expr", "Plac", "Con", "Expr", "Plac", "Con", "Expr", "Plac"),
                           DV = c(26.07, 26.06, 26.05, 26.09, 26.04, 26.65, 26.64, 26.62, 26.63),
                           Trial = c(1, 2, 3, 1, 3, 2, 1, 2, 3))

The preferred output address that Harry completed the placebo condition second and the experimental condition third, giving him c(1, 3, 2). In contrast, Bill and Paul both complete the experimental condition second and the placebo condition third, giving c(1, 2, 3). As per above, this alternate sequence is indicated by Code equalling 1 for Bill and Paul, 2 for Harry.

I know that I can include a 1 to designate the first trial completed (Con for all participants) using the following:

AnalysedData$Trial <- ifelse(AnalysedData$Condition == "Con", 1, 0)

How do I complete the above for multiple conditions?

Upvotes: 1

Views: 134

Answers (2)

Otto_K
Otto_K

Reputation: 341

This should work:

AnalysedData$Trial <- ifelse(AnalysedData$Condition == "Con", 1, 0)

index<-AnalysedData$Condition == "Plac"

AnalysedData$Trial[index]<-rep(2,sum(index))

Upvotes: 0

Jaap
Jaap

Reputation: 83235

With a combination of ifelse, as.integer (or as.numeric) and factor you can make use of the fact that factor variables are stored as integers with labels:

AnalysedData$Trial <- ifelse(AnalysedData$Code==1, 
                             as.integer(factor(AnalysedData$Condition, 
                                               levels = c("Con", "Expr", "Plac"))),
                             as.integer(factor(AnalysedData$Condition, 
                                               levels = c("Con", "Plac", "Expr"))))

you get:

> AnalysedData
  Participant Code Condition    DV Trial
1        Bill    1       Con 26.07     1
2        Bill    1      Expr 26.06     2
3        Bill    1      Plac 26.05     3
4       Harry    2       Con 26.09     1
5       Harry    2      Expr 26.04     3
6       Harry    2      Plac 26.65     2
7        Paul    1       Con 26.64     1
8        Paul    1      Expr 26.62     2
9        Paul    1      Plac 26.63     3

Upvotes: 1

Related Questions