1

I am trying to take the average score for each name. I would like to add up the scores values (omitting NAs) and then divide by the number of values. I have gotten the sum, but am struggling with how to divide by the number of values that are not `NA.

Here is some example code:

names <- c("d", "e", "f", "g")
score.a.ALL <- c(1, 2, NA, 3)
score.b.ALL <- c(NA, 4, 5, 6)
score.c.ALL <- c(NA, NA, NA, 1)

dat <- data.frame(names, score.a.ALL, score.b.ALL, score.c.ALL)
dat$goal <- c(1, 3, 5, 3.3)

scoreALL <- dat[, grep("score\\.[^.]+\\.ALL", colnames(dat))]

dat$average <- rowSums(scoreALL, na.rm=T)/ ncol(scoreALL[,is.na==F])

The goal column is what the the average should be. Any help is much appreciated! Thanks!

alistaire
  • 42,459
  • 4
  • 77
  • 117
Emma Kortemeier
  • 57
  • 5

1 Answers1

4

There is a function for row means already implemented: rowMeans. set na.rm=TRUE to omit NA 

> dat$goal <- rowMeans(dat[-1], na.rm = TRUE)
> dat
  names score.a.ALL score.b.ALL score.c.ALL     goal
1     d           1          NA          NA 1.000000
2     e           2           4          NA 3.000000
3     f          NA           5          NA 5.000000
4     g           3           6           1 3.333333
Jilber Urbina
  • 58,147
  • 10
  • 114
  • 138