Hi so I'm taking a stats class, and we were given a dataset "NHANES" that we filtered down to get adult smokers --> "NHANES_adult".
library(NHANES)
# create a NHANES dataset without duplicated IDs
NHANES <-
NHANES %>%
distinct(ID, .keep_all = TRUE)
NHANES_adult <- NHANES %>%
filter(Age >= 18) %>% # only include individuals 18 or older
filter(SmokeNow != 'NA') # drop any observations with NA for SmokeNow
My prof asked the following:
1b. Now let's take a single sample of 100 individuals from the NHANES_adult dataframe, and compute the proportion of smokers, saving it to a variable called p_smokers.
set.seed(12345) # PROVIDED CODE - this will cause it to create the same
# random sample each time
sample_size = 100 # size of each sample
p_smokers <- NHANES_adult %>%
sample(sample_size) %>% # take a sample from the data frame [I think this is okay]
____(____ = ____(____)) %>% # compute the probability of smoking [This is the point at which I'm struggling to understand what one-line function fits these blank parameters.
____() # extract the variable from the data frame [I believe this is the mutate() function?]
p_smokers
Maybe this is what you are looking for. It seems that you should use sample_n() rather than sample(). To find the proportion in one line, you would use mean().
sample_size <- 100
NHANES_adult %>%
sample_n(sample_size) %>%
summarize(p_smok = mean(SmokeNow == "Yes")) %>%
pull(p_smok)