In jjchern/meda: Display metadata records for datasets

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

About meda

This package provides R functions that replicates the behavior of four Stata commands: (1) d, (2) lookfor, (3) codebook, compact, and (4) l, sepby(). These commands present metadata records for datasets in a cleaned way.

The package is built upon two R packages: haven, which provides a labelled class that supports labelled data frames, and labelled, which makes it easier to work with labelled data frames.

Installation

Since the labelled package is still not up on CRAN, we might also install it as well.

devtools::install_github("larmarange/labelled")
devtools::install_github("jjchern/meda")

Usage

Describe and Look for Variables

library(meda)
nlsw88 = haven::read_dta("http://www.stata-press.com/data/r13/nlsw88.dta")

cb(nlsw88) # show summary statistics of a data frame

d(nlsw88) # shows variable labels, and whether value label exists for certain variables

# Note that there's a value label for the variable "race", thus we can checkout the values
labelled::val_labels(nlsw88$race)

# Note also that if a variable label is too long, we can find out the whole label with
labelled::var_label(nlsw88$grade)

# Or look at all the variable labels
labelled::var_label(nlsw88)

lookfor(nlsw88, "mar") # search for variables that are related to marriage

List Observations and Separated by Some ID Variable

library(dplyr)
abdata = haven::read_dta("http://www.stata-press.com/data/r13/abdata.dta")
abdata %>% select(id, year, wage) %>% l(by = "id", n = 30)

jjchern/meda documentation built on May 19, 2019, 11:38 a.m.