Addresses tasks along the pipeline from raw
data to analysis and visualization for eye-tracking data. Offers several
popular types of analyses, including linear and growth curve time analyses,
onset-contingent reaction time analyses, as well as several non-parametric
bootstrapping approaches. For references to the approach see Mirman,
Dixon & Magnuson (2008)
EyetrackingR is now compatible with dplyr > 0.5.0.
This package is designed to make dealing with eye-tracking data easier. It addresses tasks along the pipeline from raw data to analysis and visualization. It offers several popular types of analyses, including growth-curve analysis, onset-contingent reaction time analyses, as well as several non-parametric bootstrapping approaches.
To install from CRAN:
For the development version (make sure you have run
install.packages("devtools") to get devtools first):
EyetrackingR only requires that your data is in an R dataframe and has a few necessary columns. For that reason, eyetrackingR is compatible with any eyetracker, so long as you can export your data to a table and import it into R. See the preparing your data vignette.
Once your data is in R, you can prepare it for eyetrackingR by running the
make_eyetrackingr_data function, e.g.:
data <- make_eyetrackingr_data(your_original_data, participant_column = "ParticipantName", trial_column = "Trial", time_column = "Timestamp", trackloss_column = "TrackLoss", treat_non_aoi_looks_as_missing = TRUE )
From here, all of eyetrackingR's functionality becomes available for this data. Check out the eyetrackingR workflow to get an accesible overview of this functionality, or check out the vignettes for guides on how to clean your data, visualize it, and perform analyses.
Copyright (c) 2015, Jacob Dink and Brock Ferguson
Released under the MIT License (see LICENSE for details)
Changes in 0.1.8:
Changes in 0.1.7:
Changes in 0.1.6:
lmertime-bin or cluster analysis, via the "treatment_level" argument.
Changes in 0.1.5:
Changes in 0.1.4:
Changes in 0.1.3:
analyze_time_binsand therefore cluster-analyses have been re-written internally. Full support for (g)lm, (g)lmer, wilcox. Support for interaction terms/predictors. Experimental support for using boot-splines within cluster analysis.
analyze_time_binsand cluster analyses
analyze_boot_splinesare now deprecated. To perform this type of analysis, use
analyze_time_clustersfunction now checks that the extra arguments passed to it are the same as the arguments passed
simulate_eyetrackingr_datafunction to generate fake data for simulations.
Changes in 0.1.1:
clean_by_trackloss. Previously did not work for certain column names.
make_eyetrackingr_data. Previously did not work correctly with
treat_non_aoi_as_missing = TRUE.
analyze_time_clusters: previously did not compute permutation-distribution correctly.
make_time_sequence_datato summarize. This DV can then be plotted and used in downstream functions (like
analyze_time_binsand functions that call this (e.g
analyze_time_clusters, allowing the user to take advantage of multiple cores to speed up this relatively slow function.
get_time_clustersfor getting information about clusters in a data.frame (rather than a printed summary-- better for programming).