Sample 50 rows from 27,000 entries?

Question

Novice coder here!

I am trying to sample 50 rows from a data frame with about 27,000 rows. I have tried this:

sample <- df %>%
  sample_n(50)

but I get this error message:

Error: "size" must be less or equal than 1 (size of data), set "replace" = TRUE to use sampling with replacement.

It seems like the error is that the data frame is smaller than the sample size, which is obviously not the case. How can I fix this?

[See here](https://stackoverflow.com/q/5963269/5325862) on making a reproducible example that is easier for folks to help with. You say it's obviously not the case that the data frame is smaller than the sample size, but we don't know that for sure without access to a representative sample of data — camille, May 27 '21 at 01:11

score 0 · Answer 1 · edited May 27 '21 at 00:58

0

Check dim(df) to see if it’s actually 27k rows. This isn’t tidyverse but it will work:

index <-sample.int(nrow(df),50)
sample <- df[index, ]

edited May 27 '21 at 00:58

Ronak Shah

answered May 26 '21 at 23:04

zeejay

score 0 · Answer 2 · answered May 27 '21 at 00:57

That must be because your data is grouped. Let's take mtcars as an example -

library(dplyr)
mtcars %>% group_by(cyl) %>% sample_n(30)

Error: size must be less or equal than 11 (size of data), set replace = TRUE to use sampling with replacement.

Try ungrouping your data before using sample_n -

sample <- df %>% ungroup %>% sample_n(50)

If you have new dplyr (>= 1.0.0) you can use slice_sample instead of sample_n -

sample <- df %>% ungroup %>% slice_sample(n = 50)

2 Answers2