In tca2/datavyu: Prepare and Create Output from datavyu Coding Files

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%",
  message = FALSE
)

devtools::load_all(".")

birdseyevyu

The goal of {birdseyevyu} is to to to facilitate the use of the open-source datavyu software for the analysis of qualitative audiovisual data. This package extensively uses {datavyur} for preparing the data for summary statistics and plotting functions in this package. We do not intend to duplicate the functionality of that excellent package; instead we focus on summarizing datavyu output and preparing the output for use in other analyses.

Installation

First, install the datavyu qualitative coding software; see here

You can install the development version of this package from GitHub with; for now, this is how you'll have to install the package. First, uncomment the line for # install.packages("remotes") by removing the # symbol. Then run both that and the next line of code to install the package.

We'll also install a second package, datavyur, which can assist with our work with datavyu output.

# install.packages("remotes")
remotes::install_github("tca2/birdseyevyu")
remotes::install_github("iamamutt/datavyur")

library(birdseyevyu)

Preparing data for analysis

See the Preparing Data for Analysis vignette.

Exploring the columns and files

Using the other R package that works with datavyu, {datavyur} you can find information on all of the files (and their columns and codes) in a given directory (not run here):

library(datavyur)

datavyu_search("inst/extdata/datavyu_output_11-16-2020_13-26")

Summarizing a column

{birdseyevyu} can help to summarize a column. It defaults to summarizing the frequency of codes for a specified column.

Note that, by default, the code name combines the column and code names; so, the hand code in the above summary would be combined with its column childhands (or parenthands) to be specified as childhands.hand or parenthands_hand:

summarize_column(column = "childhands",
                 code = "hand",
                 directory = "inst/extdata/datavyu_output_11-16-2020_13-26",
                 summary = "duration")

---

We can also explore the frequencies by file by changing the by_file argument to TRUE.

We'll be typing that folder file path a number of times. You can set an option that let the folder file path you set be used by default, though you can over-ride it any time you like.

options(directory = "inst/extdata/datavyu_output_11-16-2020_13-26")

Then, use this default by changing the the by_file argument to TRUE:

summarize_column(column = "childhands",
                 code = "hand",
                 by_file = TRUE)

---

To summarize durations (instead of frequencies) change the summary argument(defaults as "frequency") to "duration":

my_summary <- summarize_column(column = "childhands",
                               code = "hand",
                               summary = "duration")

my_summary

Columns of durations can also be summarized by file:

duration_output <- summarize_column(column = "childhands",
                                    code = "hand",
                                    by_file = TRUE,
                                    summary = "duration")

Ploting the results of a summary of a column

{birdseyevyu} can also help to plot the summary of a column. Here, we save the output from summarize_column() to an object (we call this freq_summary, but it can be named whatever we like).

Then, we use this output in the function plot_column_summary():

freq_summary <- summarize_column(column = "childhands", code = "hand")

plot_column_summary(freq_summary)

This also works by file---so long as the column is summarized by file:

freq_summary <- summarize_column(column = "childhands", code = "hand",
                                 by_file = TRUE, summary = "duration")

plot_column_summary(freq_summary)

Similarly, if the output is for the duration, rather than the frequency, the durations are plotted:

duration_summary <- summarize_column(column = "childhands", code = "hand",
                                     summary = "duration")

plot_column_summary(duration_summary)

Like for frequency, these can be ploted by file:

duration_summary_by_file <- summarize_column(column = "childhands", 
                                             code = "hand",
                                             summary = "duration",
                                             by_file = TRUE)

plot_column_summary(duration_summary_by_file)

Using the pipe operator

Finally, output can be passed between functions with the pipe operator:

summarize_column(column = "childhands", 
                 code = "hand",
                 summary = "duration",
                 by_file = TRUE) %>% 
  plot_column_summary()

Time series preparation and plot (in-development)

To visualize frequency of a certain coded datavyu column accounting for the duration of the video, we can use the function plot_time_series(). First we must prepare the data via saving the prep_time_series() function as an object (here we use prepared_time_series).

Note that the available files can be found using datavyur::datavyu_col_search():

prepared_time_series <- prep_time_series(column = "childhands",
                                         code = "hand",
                                         specified_file = "dyad1")

prepared_time_series

The units argument defaults to "s" (seconds), but can be changed to "m" (rounds the data to minutes) or "ms" (retains datavyu's default units, milliseconds).

We can see how using milliseconds increases the number of data points:

prepared_time_series_ms <- prep_time_series(column = "childhands",
                                            code = "hand",
                                            specified_file = "dyad1",
                                            units = "ms")

prepared_time_series_ms

Your prepared_time_series_ms object can now be plotted usinf the plot_time_series() function. Here is an example of how time series data can be represented, utilizing the functions defaulted unit value of seconds:

plot_time_series(prepared_time_series)

Features in-development

• inter-rater agreement
• creating a tidy data frame for ad hoc and custom analyses
• Plotting code co-occurrences with plot_cooccurence()
• Summarizing an entire file (not just a single column in a file) with summarize_file()
• Addressing a number of issues, including some that relate to the {datavyur} package

Contributing

Please note that the datavyu project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Acknowledgment

This material is based upon work supported by the National Science Foundation under Grant No. 1920796. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not reflect the views of the National Science Foundation.

tca2/datavyu documentation built on Dec. 23, 2021, 8:40 a.m.