R for BRFSS: ch 4-- subset your data file

Описание: Welcome back! In this chapter I'll teach you a few tricks, we'll subset our data file, and write a hypothesis.

Bad news guys! I can't use greater than or less than signs in the description. So I'm going to indicate the assignment arrow by typing the word ~arrow~!!

Set a CRAN mirror:
Packages -- Set CRAN mirror -- 0-Cloud [https]

Change your directory:
File -- Change dir… -- Desktop -- R fof BRFSS

To look up a function from the R console:
?as.integer

Load the libraries:
library(rio)
library(tidyverse)

Import the data file using a base R function, set it as an object:
llcp2017 ~arrow~ read.csv("llcp2017.csv", header = TRUE)

For the CDC BRFSS 2017 website:
Search for “cdc brfss 2017” and select the first option “… Survey Data and Documentation”
Open the “2017 BRFSS Codebook” and “BRFSS Modules Used by State”

Subset your data file to California:
ca2017 ~arrow~ filter(llcp2017, X_STATE == 6)

Inspect your data file:
str(ca2017)
glimpse(ca2017)

Return column sums for missing data:
colSums(is.na(ca2017))

Return back TRUE/ FALSE for columns with 100% missing data:
colSums(is.na(ca2017)) == nrow(ca2017))

Return back only columns with 100% missing data:
which(colSums(is.na(ca2017)) == nrow(ca2017)))

Set the above command to an object:
drop_var ~arrow~ which(colSums(is.na(ca2017)) == nrow(ca2017)))

Drop the drop_var columns from your data file and rename the data file:
ca2017drop_var ~arrow~ ca2017[, -drop_var]

Count the number of respondents for each SEX category:
count(ca2017drop_var, vars = SEX)

Return back only the people who refused to report their sex:
which(ca2017drop_var$SEX == 9)

Set that command to an object:
sexR ~arrow~ which(ca2017drop_var$SEX == 9)

Drop those rows from the data files, rename the data file:
ca2017drop_varMF ~arrow~ filter(ca2017drop_var[-sexR, ])

Double check the number of rows:
nrow(ca2017drop_varMF)

Rename your data file:
ca_girls17 ~arrow~ ca2017drop_varMF

Export your data file:
export(ca_girls17, "ca_girls17.csv ")