README.md

In antonmalko/ibextor: Easy preprocessing of Ibex data

ibextor

This package is a successor to IBEX.to.R package, which I posted a few years ago here. The current package is a much cleaner re-write of the old code, and has a couple of new feautures.

The goal of the package is to facilitate loading experimental data generated on Ibex into R. Ibex data are stored in a simple .csv file, so loading it into R per se is easy. However, multiple preprocessing steps have to be carried on such raw data. The functions in this package automatically perform many of such steps:

• Columns are automatically assigned appropriate names and data types
• Columns not containing useful info are automatically deleted (e.g. the column indicating whether the answer to a question was correct will contain only NULL if the correct answer was not specified in the Ibex script)
• Controllers adding multiple lines to the data file (e.g. "DashedSentence") are "unrolled", so that all the info from an experimental item is contained in a single data.frame
• Participants are automatically assigned simple numeric ids (1,2, ...), instead of being identified by strings like "1520910272_c575f9a34a249d8f9d34c59469515f14"

Functions overview

Data loading functions

Several functions are available, corresponding to the main default Ibex controllers:

• get_results_ds() - DashedSentence
• get_results_daj() - DashedAcceptabilityJudgment
• get_results_aj() - AcceptabilityJudgment
• get_results_q() - Question

In the most basic case one would just pass the name of the file with Ibex result as the single parameter, and the functions will return pre-processed data in a data.frame. See documentation for additional options (e.g. specifying columns to be deleted)

Subject info function

get_subj_info()

In the most basic case, again, it is enough to pass the name of the file with Ibex results. A couple of parameters are of immediate interest, however:

• form_name - the name of the form used to collect subject info. If there were several forms, you can just pass all of their names as a vector. These names correspond to the names you assigned to the forms in the .js file. This parameter defaults to "intro" (name of the form in the Ibex experiment template).
• info - name of the fields you want to extract. If NULL (default), all available info will be extracted. Otherwise, pass a vector with the names of the fields you want to get (these would be whatever you put in the "name" attribute of the input fields in the HTML code for the forms).

Installation

devtools::install_github("antonmalko/ibextor")

Notice that the package is still being developed and tested.

antonmalko/ibextor documentation built on May 25, 2019, 2:26 p.m.