Styjun

I have one dataframe (df_features) consisting of 32 rows and six columns that relate to potential features of a study and a second dataframe (df_participants) containing 10,000 unique (non-numeric) IDs of my participants. There are no common columns across the two dataframes.

I want to create a dataset that contains each of the 32 rows from df_features for every ID in df_participants (so 320,000 rows and 7 columns in total).

How do I do this? I feel like it should be straightforward but I just can't find anything anywhere!

It sounds like you are looking to do a full outer join that will combine all features with all IDs. This can be done using several packages, and in base-R with the following:

features <- data.frame(f1=c("blue","geeen"),f2=c("young","old"))
participants <- data.frame(ID=c(1:10))

merge(features,participants,all=T)

You can do a full outer join. When you do a full outer join with no common columns across the two dataframes, you get the cartesian product of the two dataframes, which is what you're looking for. You can get this using the merge function. If your only two arguments to merge are the dataframes you want to perform the join on, you will get back the cartesian product of those dataframes.

Example:

df1 <- data.frame(y = 1:4)
df2 <- data.frame(z = 1:3)
df_merged <- merge(df1, df2)
print(df1)
# y
#1 1
#2 2
#3 3
#4 4

print(df2)
# z
#1 1
#2 2
#3 3

print(df_merged)
# y z
#1 1 1
#2 2 1
#3 3 1
#4 4 1
#5 1 2
#6 2 2
#7 3 2
#8 4 2
#9 1 3
#10 2 3
#11 3 3
#12 4 3

I found a fairly convoluted way around it in case anyone is looking to do something similar:

matching_1<- expand.grid(df_participants$ID, df_features$feature_rownumber) %>% arrange(Var1) %>%
 rename("ID"=Var1, "feature_rownumber"=Var2)

matching_2 <- left_join(df_participants, matching_1, by="ID")

final_dataset <- left_join(matching_2, df_features, by="feature_rownumber")

However I'm fairly sure there must be a more concise method out there!