Compute the means and standard deviations of var1 within each level of var2

Question

I have a data set where the variable Tier can be 1, 2, 3 , or 4. I need to find the mean of Full.time for each type of Tier. Here is my code:

mean(my_data$Full.time ~ my_data$Tier)

and here is my error message:

[1] NA
Warning message:
In mean.default(my_data$Full.time ~ my_data$Tier) :
  argument is not numeric or logical: returning NA

(I also have to do the same thing with standard deviation but I assume the correct answer will be the same with sd() instead of mean())

Here is the file:

The answers to [this question](https://stackoverflow.com/questions/21982987/mean-per-group-in-a-data-frame) are a terrific summary of the options available for getting means by group. — A. S. K., Dec 11 '19 at 19:50
Does this answer your question? [Mean per group in a data.frame](https://stackoverflow.com/questions/21982987/mean-per-group-in-a-data-frame) — A. S. K., Dec 11 '19 at 19:50
[See here](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) on making an R question that folks can help with. That includes a sample of data, all necessary code, and a clear explanation of what you're trying to do and what hasn't worked. That said, I don't think I've seen `mean` called on a formula...where did you get that? — camille, Dec 11 '19 at 19:51
@A.S.K. I don't think so? This is what my textfile looks like — Hannah, Dec 11 '19 at 20:13
Try `aggregate(my_data$Full.time, list(my_data$Tier), mean)`; that should create the summary you're looking for. — A. S. K., Dec 11 '19 at 20:19
Probably because Tier 4 contains NAs somewhere; try `aggregate(my_data$Full.time, list(my_data$Tier), mean, na.rm = T)`. — A. S. K., Dec 11 '19 at 20:32

score 0 · Answer 1 · answered Dec 11 '19 at 20:30

Maybe the following code can make things you want:

if you want compact result of mean and standard deviation, you can use aggreate(), i.e.,

Mean <- aggregate(Full.time ~ Tier, my_data, mean)

and

SD <- aggregate(Full.time ~ Tier, my_data, sd)

if you want results of mean and standard deviation with the same rows of my_data, you can use ave(), i.e.,

Mean <- with(my_data, ave(Full.time, Tier, mean))

and

SD <- with(my_data, ave(Full.time, Tier, sd))

smci · Answer 2 · 2019-12-12T20:05:53.023

Here's the dplyr approach:

library(dplyr)

mtcars %>%
    group_by(Tier) %>%
    summarize(mean=mean(Full.time), sd=sd(Full.time))

Note that dplyr allows us to name each aggregation result column. And here's an illustration on mtcars, because we can't access your dataset:

data(mtcars)

mtcars %>% group_by(cyl) %>% summarize(mean_hp=mean(hp), sd_hp=sd(hp))

    cyl mean_hp sd_hp
  <dbl>   <dbl> <dbl>
1     4    82.6  20.9
2     6   122.   24.3
3     8   209.   51.0

Compute the means and standard deviations of var1 within each level of var2

2 Answers2