I have a data frame in which each individual (row) has two data points per variable.
Example data:
df1 <- read.table(text = "IID L1.1 L1.2 L2.1 L2.2
1 1 38V1 38V1 48V1 52V1
2 2 36V1 38V2 50V1 48Y1
3 3 37Y1 36V1 50V2 48V1
4 4 38V2 36V2 52V1 50V2",
stringsAsFactor = FALSE, header = TRUE)
I have many more columns than this in the full dataset and would like to recode these values to label unique identifiers across the two columns. I know how to get identifiers and relabel a single column from previous questions (Creating a unique ID and How to assign a unique ID number to each group of identical values in a column) but I don't know how to include the information for two columns, as R identifies and labels factors per column.
Ultimately I want something that would look like this for the above data:
(df2)
IID L1.1 L1.2 L2.1 L2.2
1 1 1 1 1 4
2 2 2 4 2 5
3 3 3 2 3 1
4 4 1 5 4 3
It doesn't really matter what the numbers are, as long as they indicate unique values across both columns. I've tried creating a function based on the output from:
unique(df1[,1:2])
but am struggling as this still looks at unique entries per column, not across the two.
Something like this would work...
pairs <- (ncol(df1)-1)/2
for(i in 1:pairs){
refs <- unique(c(df1[,2*i],df1[,2*i+1]))
df1[,2*i] <- match(df1[,2*i],refs)
df1[,2*i+1] <- match(df1[,2*i+1],refs)
}
df1
IID L1.1 L1.2 L2.1 L2.2
1 1 1 1 1 4
2 2 2 4 2 5
3 3 3 2 3 1
4 4 4 5 4 3
You could reshape it to long format, assign the groups and then recast it to wide:
library(data.table)
df_m <- melt(df, id.vars = "IID")
setDT(df_m)[, id := .GRP, by = .(gsub("(.*).","\\1", df_m$variable), value)]
dcast(df_m, IID ~ variable, value.var = "id")
# IID L1.1 L1.2 L2.1 L2.2
#1 1 1 1 6 9
#2 2 2 4 7 10
#3 3 3 2 8 6
#4 4 1 5 9 8
This should also be easily expandable to multiple groups of columns. I.e. if you have L3. it should work with that as well.
Related
I'm pretty new to R and hope i'll make myself clear enough.
I have a table of several columns which are factors. I want to make a score for each of these columns. Then I want to calculate the mean of each score, and display the list of columns ranked by their mean scores, is that possible ?
Table would be:
head(musico[,69:73])
AVIS1 AVIS2 AVIS3 AVIS4 AVIS5
1 2 1 2 3 2
2 2 5 2 3 2
3 3 2 5 5 1
4 1 2 5 5 5
5 1 5 1 3 1
6 4 1 4 5 4
I want to make a score for each:
musico$score1<-0
musico$score1[musico$AVIS1==1]<-1
musico$score1[musico$AVIS1==2]<-0.5
then do the mean of each column score: mean of score1, mean of score2, ...:
mean(musico$score1), mean(musico$score2), ...
My goal is to have a list of titles (avis1, avis2,...) ranked by their mean score.
Any advice appreciated !
Here's one way using base although it is somewhat unclear what you want. What does score1 have to do with AVIS1? I think you may be missing some of the data from musico.
Based on the example provided, here's a base R solution. vapply loops through the data.frame and produces the mean for each column. Then the stack and order are only there to make the output a dataframe that looks nice.
music <- read.table(text = "
AVIS1 AVIS2 AVIS3 AVIS4 AVIS5
1 2 1 2 3 2
2 2 5 2 3 2
3 3 2 5 5 1
4 1 2 5 5 5
5 1 5 1 3 1
6 4 1 4 5 4", header = TRUE)
means <- vapply(music, mean, 1)
stack(means[order(means, decreasing = TRUE)])
values ind
4 4.000000 AVIS4
3 3.166667 AVIS3
2 2.666667 AVIS2
5 2.500000 AVIS5
1 2.166667 AVIS1
This is how I would do it by first introducing a scores vector to be used as a lookup. I assume that scores are decreasing by 0.5 and that the number of scores needed are according to the maximum number of levels found in your columns (i.e. 6 seen in AVIS1).
Then using tidyr you can organise your data set such that you have to variables (i.e. AVIS and Value) containing the respective levels. Then add a score variable with the mutate function from dplyr in which the position of the score in the score vector matches the value in the Value variable. From here you can find the mean scores corresponding to the AVIS levels, arrange them accordingly and put them in a list.
music <- read.table(text = "
AVIS1 AVIS2 AVIS3 AVIS4 AVIS5
1 2 1 2 3 2
2 2 5 2 3 2
3 3 2 5 5 1
4 1 2 5 5 5
5 1 5 1 3 1
6 4 1 4 5 4", header = TRUE) # your data
scores <- seq(1, by = -0.5, length.out = 6) # vector of scores
library(tidyr)
library(dplyr)
music2 <- music %>%
gather(AVIS, Value) %>% # here you tidy the data
mutate(score = scores[Value]) %>% # match score to value
group_by(AVIS) %>% # group AVIS levels
summarise(score.mean = mean(score)) %>% # find mean scores for AVIS levels
arrange(desc(score.mean))
list <- list(AVIS = music2$AVIS) # here is the list
> list$AVIS
[1] "AVIS1" "AVIS5" "AVIS2" "AVIS3" "AVIS4"
I have a long data frame with players' decisions who worked in groups.
I need to convert the data in such a way that each row (individual observation) would contain all group members decisions (so we basically can see whether they are interdependent).
Let's say the generating code is:
group_id <- c(rep(1, 3), rep(2, 3))
player_id <- c(rep(seq(1, 3), 2))
player_decision <- seq(10,60,10)
player_contribution <- seq(6,1,-1)
df <-
data.frame(group_id, player_id, player_decision, player_contribution)
So the initial data looks like:
group_id player_id player_decision player_contribution
1 1 1 10 6
2 1 2 20 5
3 1 3 30 4
4 2 1 40 3
5 2 2 50 2
6 2 3 60 1
But I need to convert it to wide per each group, but only for some of these variables, (in this example specifically for player_contribution, but in such a way that the rest of the data remains. So the head of the converted data would be:
data.frame(group_id=c(1,1),
player_id=c(1,2),
player_decision=c(10,20),
player_1_contribution=c(6,6),
player_2_contribution=c(5,5),
player_3_contribution=c(4,6)
)
group_id player_id player_decision player_1_contribution player_2_contribution player_3_contribution
1 1 1 10 6 5 4
2 1 2 20 6 5 6
I suspect I need to group_by in dplyr and then somehow gather per group but only for player_contribution (or a vector of variables). But I really have no clue how to approach it. Any hints would be welcome!
Here is solution using tidyr and dplyr.
Make a dataframe with the columns for the players contributions. Then join this dataframe back onto the columns of interest from the original Dataframe.
library(tidyr)
library(dplyr)
wide<-pivot_wider(df, id_cols= - player_decision,
names_from = player_id,
values_from = player_contribution,
names_prefix = "player_contribution_")
answer<-left_join(df[, c("group_id", "player_id", "player_decision") ], wide)
answer
group_id player_id player_decision player_contribution_1 player_contribution_2 player_contribution_3
1 1 1 10 6 5 4
2 1 2 20 6 5 4
3 1 3 30 6 5 4
4 2 1 40 3 2 1
5 2 2 50 3 2 1
6 2 3 60 3 2 1
I want to create a new variable in an R data frame by using an existing column as a lookup value on another column in the same table. For example, in the following data frame:
df = data.frame(
pet = c("smalldog", "mediumdog", "largedog",
"smallcat", "mediumcat", "largecat"),
numPets = c(1, 2, 3, 4, 5, 6)
)
> df
pet numPets
1 smalldog 1
2 mediumdog 2
3 largedog 3
4 smallcat 4
5 mediumcat 5
6 largecat 6
I want to to create a new column called numEnemies which is equal to zero for small animals but equal to the number of animals of the same size but the different species for medium and large animals. I want to end up with this:
pet numPets numEnemies
1 smalldog 1 0
2 mediumdog 2 5
3 largedog 3 6
4 smallcat 4 0
5 mediumcat 5 2
6 largecat 6 3
The way I was attempting to do this was by using conditional logic to generate a character variable which I could then use to look up the final value I want from the same data frame, which got me to here:
calculateEnemies <- function(df) {
ifelse(grepl('small', df$pet), 0,
ifelse(grepl('dog', df$pet), gsub('dog', 'cat', df$pet),
ifelse(grepl('cat', df$pet),
gsub('cat', 'dog', df$pet), NA)))
}
df$numEnemies <- calculateEnemies(df)
> df
pet numPets numEnemies
1 smalldog 1 0
2 mediumdog 2 mediumcat
3 largedog 3 largecat
4 smallcat 4 0
5 mediumcat 5 mediumdog
6 largecat 6 largedog
I want to modify this function to use the newly generated string to lookup the values from df$numPets based on the corresponding value in df$pet. I'm also open to a better approach that also generalizes.
Here's how I would approach this using the data.table packages
library(data.table)
setDT(df)[, numEnemies := rev(numPets), by = sub(".*(large|medium).*", "\\1", pet)]
df[grep("^small", pet), numEnemies := 0L]
# pet numPets numEnemies
# 1: smalldog 1 0
# 2: mediumdog 2 5
# 3: largedog 3 6
# 4: smallcat 4 0
# 5: mediumcat 5 2
# 6: largecat 6 3
What I basically did, is to first create groups of medium and large over the whole data set and just reverse the values within each group.
Then, I've assigned 0 to all the values in numPets when grep("^small", pet).
This should be both very efficient and robust, as it will work on any number of animals and you don't actually need to know the animals names apriori.
I have a few large data sets with many variables. There is a "key" variable that is the ID for the research participant. In these data sets, there are some IDs that are duplicated. I have written code to extract all data for duplicated IDs, but I would like a way to check if the remainder of the variables for those IDs are equal or not. Below is a simplistic example:
ID X Y Z
1 2 3 4
1 2 3 5
2 5 5 4
2 5 5 4
3 1 2 3
3 2 2 3
3 1 2 3
In this example, I would like to be able to identify that the rows for ID 1 and ID 3 are NOT all equal. Is there any way to do this in R?
You can use duplicated for this:
d <- read.table(text='ID X Y Z
1 2 3 4
1 2 3 5
2 5 5 4
2 5 5 4
3 1 2 3
3 2 2 3
3 1 2 3
4 1 1 1', header=TRUE)
tapply(duplicated(d), d[, 1], function(x) all(x[-1]))
## 1 2 3 4
## FALSE TRUE FALSE TRUE
Duplicated returns a vector indicating, for each row of a dataframe, whether it has been encountered earlier in the dataframe. We use tapply over this logical vector, splitting it in to groups based on ID and applying a function to each of these groups. The function we apply is all(x[-1]), i.e. we ask whether all rows for the group, other than the initial row, are duplicated?
Note that I added a group with a single record to ensure that the solution works in these cases as well.
Alternatively, you can reduce the dataframe to unique records with unique, and then split by ID and check whether each split has only a single row:
sapply(split(unique(d), unique(d)[, 1]), nrow) == 1
## 1 2 3 4
## FALSE TRUE FALSE TRUE
(If it's a big dataframe it's worth calculating unique(d) in advance rather than calling it twice.)
I have columns 1 and 2 (ID and value). Next I would like a count column that lists the # of times that the same value occurs per id. If it occurs more than once, it will obviously repeat the value. There are other variables in this data set, but the new count variable needs to be conditional only on 2 of them. I have scoured this blog, but I can't find a way to make the new variable conditional on more than one variable.
ID Value Count
1 a 2
1 a 2
1 b 1
2 a 2
2 a 2
3 a 1
3 b 3
3 b 3
3 b 3
Thank you in advance!
You can use ave:
df <- within(df, Count <- ave(ID, list(ID, Value), FUN=length))
You can use ddply from plyr package:
library(plyr)
df1<-ddply(df,.(ID,Value), transform, count1=length(ID))
>df1
ID Value Count count1
1 1 a 2 2
2 1 a 2 2
3 1 b 1 1
4 2 a 2 2
5 2 a 2 2
6 3 a 1 1
7 3 b 3 3
8 3 b 3 3
9 3 b 3 3
> identical(df1$Count,df1$count1)
[1] TRUE
Update: As suggested by #Arun, you can replace transform with mutate if you are working with large data.frame
Of course, data.table also has a solution!
data[, Count := .N, by = list(ID, Value)
The built-in constant, ".N", is a length 1 vector reporting the number of observations in each group.
The downside to this approach would be joining this result with your initial data.frame (assuming you wish to retain the original dimensions).