CODEWITHSUNDEEP
rdataframemean
evian_bottle
evian_bottle

Reputation: 77

Calculate mean of data frame by row

Is there a way to calculate the mean of a row from a data frame?

So for example:

df <- data.frame(X = c(1,1,1,1,2,2), Y = c(1,4,4,4,4,3), Z = c(2,5,6,8,3,1))

df
  X Y Z
1 1 1 2
2 1 4 5
3 1 4 6
4 1 4 8
5 2 4 3
6 2 3 1

So how can I calculate the mean per row?

mean(df[,1:3])

I really don't understand why this won't work, I mean the same code works for example for min() or max() but not for mean().

min(df[,1:3])

[1] 1

Upvotes: 1

Views: 2703

Answers (2)

Macosso
Macosso

Reputation: 1439

To do this you need to use apply function you can compute the mean of all the rows by using the following syntax

apply(df,1, mean)
[1] 1.333333 3.333333 3.666667 4.333333 3.000000 2.000000

#when the second argument is 1, you are computing mean for each row, if it is set to 2 then you are computing for each column

To compute the mean of a specific row you need to subset the row use the following code

> apply(df[4,],1, mean) #here you compute the mean of the 4th row
#output
       4 
4.333333

Upvotes: 1

Trusky
Trusky

Reputation: 503

When you say df[,1:3] you are choosing all rows of df and columns 1:3. When you apply min or max to that, it simply looks for the min/max among all numbers. It is not doing it by row.

So when yo try to apply the same logic to mean, it again finds the mean value among all numbers in all three columns. Again, not by row.

You need to apply a function to a dimension of df. For this, use apply(df, 1, mean) as PKumar suggested. If you need the mean for each column, you say apply(df, 2, mean). To learn more about apply type ?apply on the R console.

rowMeans and colMeans are shortcuts for apply.

Upvotes: 1

Related Questions