In wang114/foofactorsran: Make factors less annoying

This documents contains detailed information of the functions in the foofactorsran package. fbind() and freq_out() were left unchanged.

Define an age group

age_group() takes a numeric vector (ages) with each element in the vector bigger than 0 and smaller than 120 to be a valid age. It returns a factor of the same length as the input and replaces each age by its age group. The age groups are defined as follows:

• 0<x<=14: "Children"
• 14<x<=24: "Youth"
• 24<x<=65: "Adult"
• 65<x : "Senior"

If droplevel = FALSE, the levels for x will be c("Children","Youth","Adult","Senior"). If droplevel = TRUE, which is also the default setting, the unused levels of age groups will be dropped, but the order of levels stay the same.

Example:

library(foofactorsran)

x <- c(2,45,22,27)
age_group(x)
age_group(x,droplevel = T)
age_group(x,droplevel = FALSE)

The following input would produce an error.

oops1 <- age_group(c(12,56),droplevel = 2)
oops2 <- age_group(c(-1,56))

Reorder factor levels

factor_level() is a function that converts the input x to a factor (if it is not a factor ) and defines the order of factor levels at the same time. The level argument is set to "alp" by default which means the levels will be ordered alphabetically. If level = "asit", the factor levels will be set in the order they appear in x. User can also define the level order the way they define in the function factor(), other arguments of factor() can be passed to factor_level() as well.

To avoid turning a value of x to "NA" by defining levels (like factor() sometimes does), only level with the same length as original number of levels (or number of levels of as.factor(x) if x is not originally a factor), and the level must include all values of x. Note that factor_level() would not drop unused levels if the input x is a factor.

Examples:

x <- factor(c("b","b","a","c","d","a"),levels = c("c","a","b","d","e"))

factor_level(x, level = c("e","d","c","b","a"))
factor_level(x)
factor_level(x,"asit")

The following will produce an error.

y <- c("aa","aa","cc","bb","cc")
z <- as.factor(y)
factor_level(y, level = c("aa","cc"))
factor_level(z, level = c("aa","cc"))

factor_level(y, level = c("red","green","blue"))
factor_level(z, level = c("red","green","blue"))

Factor_level() can pass arguments of the function factor() only if the levelargument is user defined. To see an example:

v <- c("apple","apple","orange","pear")
factor_level(v,level = c("apple","pear","orange"), labels = c("red","orange","yellow"))

order_asit() is a function that takes a vector of characters (or a factor) and returns a factor with order of levels defined in the way they appear in x. order_asit(x) is equivalent to factor_level(x,"asit"). Note that a warning would show up if the input has many unique values (same thing applies to factor_level()).

w <- as.factor(c("a","b","d","c","e"))
order_asit(w)

wang114/foofactorsran documentation built on May 4, 2019, 12:57 a.m.