Wsrtjtyk

My dataset: df

PID<-c(1,2,3,4,5,6,7,8,9)  

gender<-c(1,1,0,1,0,0,0,1,1)  

smoking<-c(1,1,0,0,0,0,1,0,1)  

disease<-c(1,0,0,1,1,1,0,1,0)  

BMI<-c(24,23,21,28,29,21,18,19,16)    

 df<-data.frame(PID, gender, smoking, disease, BMI)

I want to split this dataset based on gender. Then develop crosstab between smoking and disease. How to do this?
Expected outcome(first question):

Gender:1

crosstab between smoking and disease

Gender:2

Crosstab between smoking and disease.

Expected outcome (second question):

Gender:1

mean of BMI

Gender:2

mean of BMI

No need for external packages:

xtabs(~smoking+disease+gender,data=df)

# , , gender = 0

#        disease

# smoking 0 1

#       0 1 2

#       1 1 0

# , , gender = 1

#        disease

# smoking 0 1

#       0 0 2

#       1 2 1

aggregate(df$BMI, list(gender=df$gender), FUN=mean)

#   gender     x

# 1      0 22.25

# 2      1 22.00

Similarly (thanks thelatemail):

aggregate(BMI ~ gender, data = df, FUN=mean)

Here is a possible way for the first question using dplyr:

library(dplyr)

library(magrittr)





> df %>% split(gender) %>% lapply(function(x) tab=xtabs(gender~smoking+disease, data=x))

$`0`

       disease

smoking 0 1

      0 0 0

      1 0 0



$`1`

       disease

smoking 0 1

      0 0 2

      1 2 1