I am working with a data set, which is basically daily usage data (let's just say variable X and Y) by different cities (about 150 cities). I have created a subset of data for only specific cities, choosing just 3 of the 150 cities. Then when I do tapply by cities, I get means for 3 cities but also get NA for all other 147 cities that was in the data set. I am using the below coding df<-read.csv(...) df_sub<-subset(df,df$City==1|df$City==3|df$City==19) X_Breakdown<-tapply(X,df_sub$City, mean, na.rm=TRUE) Print(X_Breakdown) City 1 City 2 15 NA City 3 City 4 12 NA City 5 City 6 NA NA Hope you get the idea. I would like to get a dataset that only contains the 3 cities that I'm interested in. It seems that the set of variables is encoded in R, is there a way to fix this? Kindly advise. Thanks