[R] [R-pkgs] New package: codebook

Ruben Arslan rubenarslan at gmail.com
Fri Mar 23 09:54:52 CET 2018

Dear R users,

codebook (v. 0.5.8) was just released on CRAN.

It uses the metadata/attributes that some other packages add (item labels,
value labels, labelled missings) to automatically generate a codebook that
is both human- and machine-readable. It combines metadata about the
variables with data summaries. Many people will import this kind of
metadata already, when importing data from SPSS or Stata files using the
haven package.

My hope was to encourage people to share data and especially
machine-readable metadata more often by giving them something that is
useful (automating a common, tedious task), so that we can have something
even more useful in the future (searchable, machine-readable metadata).

A small example and more documentation is here:

codebook automates the following tasks:
* computing reliabilities (internal consistencies, retest, multilevel) for
psychological scales
* summarising the distributions of scales and items graphically and using
descriptive statistics
* combining this information with metadata (such as item labels and
labelled values) derived from attributes

To do so, the package relies on 'rmarkdown' partials, so you can generate
HTML, PDF, and Word documents, and embed the codebook in a larger document
providing context.

Codebooks are also generated as tables (CSV, Excel, etc.) and are invisibly
embedded as JSON-LD in generated HTML files. JSON-LD for datasets is not
yet supported by search engines, but hopefully will be.

Critiques, bug reports and enhancement requests are welcome as GitHub
issues. I'm a social scientist and tested this mostly on survey data. I'm
sure that shows in some design decisions. I'd be happy to find
collaborators who can e.g. fill in my knowledge gaps on metadata and



Ruben C. Arslan

Center for Adaptive Rationality
Max Planck Institute for Human Development
Berlin, Germany

	[[alternative HTML version deleted]]

R-packages mailing list
R-packages at r-project.org

More information about the R-help mailing list