How to desconsider row for statistical operation that contains NA in specific column in R?

Asked May 16 '23 at 15:17

Active May 16 '23 at 17:06

Viewed 16 times

I want to execute Shapiro test for the following data frame (example) but I think the NA value in row 5 could change the result. How can I disregard the row for the analysis without excluding the row containing the NA value?

id	subject	measurement
1	control	0.151708
2	control	0.239930
3	patient	0.104289
4	patient	0.329139
5	patient	NA

Would be nice if the Shapiro function accepted 'na.rm' or 'na.omit' as argument. However, it doesn't work.

byf.shapiro(dados_novo$measurement ~ dados_novo$subject, dados_novo, na.rm=TRUE)

edited May 16 '23 at 17:06

asked May 16 '23 at 15:17

Ítalo César Porto Xavier

1

Try something like `byf.shapiro(dados_novo$measurement ~ dados_novo$subject, dados_novo[!is.na(dados_novo$measurement), ])`? – jpsmith May 16 '23 at 15:26
Same result as ```byf.shapiro(dados_novo$measurement ~ dados_novo$subject, dados_novo)``` – Ítalo César Porto Xavier May 16 '23 at 17:02
Then what is the problem? – jpsmith May 16 '23 at 17:03
The result considers the NA value, or exclude from the analysis the row 5 automatically? – Ítalo César Porto Xavier May 16 '23 at 17:08
The results do not consider `NA` values when implementing the code from my comment, and if those results are the same as your attempt, your attempt also exclude the `NA` values – jpsmith May 16 '23 at 17:09

How to desconsider row for statistical operation that contains NA in specific column in R?

0 Answers0