Chi-square Goodness of Fit for a Test of Normality in r

Question

I want to test for normality of a set of data using Chi-Square Goodness of Fit Test in R just the way I tested for Shapiro- wilk test. I have my sample sizes to be 10, 20,50 and 100 while my replicate is 1000

## Shapiro- wilk test [sw]

x <- rnorm(x, 0, 1)
out <- t(sapply(c(10, 20, 50, 100), function(x) 
  table(replicate(1000,shapiro.test(rnorm(x)))["p.value",] < 0.05)))
row.names(out) <- c(10, 20, 50, 100)
out

#     FALSE TRUE
# 10    947   53
# 20    945   55
# 50    943   57
# 100   943   57

Answer 1

You need to make sure the chi test get non negative values. Try replacing the table calculation with this code:

table(replicate(1000,chisq.test(x=abs(rnorm(x)),p=c(rep(1/x,x))))["p.value",] < 0.05)