rENA draws plots with plotly (the default) or qeviz, the library behind webENA’s network plots. Existing plotting code works on either: add backend = "qeviz" to ena.plot(), or set it for the session with
options(rENA.plot.backend = "qeviz")qeviz (>= 0.5.0) installs from qe-libs.org:
install.packages("qeviz", repos = c("https://cran.qe-libs.org", "https://cran.rstudio.org"))Below, each view is drawn three ways from RS.data: the rENA code with each backend (only backend differs), and the same view written with qeviz’s own functions, which take the ENA set directly. The full list of differences is in the ?rENA-plot-backends help page.
All three use this setup:
library(rENA)
data(RS.data)
codeNames <- c("Data", "Technical.Constraints", "Performance.Parameters",
"Client.and.Consultant.Requests", "Design.Reasoning", "Collaboration")
accum <- ena.accumulate.data(
units = RS.data[, c("Condition", "UserName")],
conversation = RS.data[, c("Condition", "GroupName")],
codes = RS.data[, codeNames],
window.size.back = 4
)
set <- ena.make.set(accum)
first <- set$meta.data$Condition == "FirstGame"
second <- set$meta.data$Condition == "SecondGame"
lw <- as.matrix(set$line.weights)
pts <- as.matrix(set$points)A group’s network and its mean
Click the figure for full size.
On qeviz, line width is absolute per weight and each node’s size is the sum of its edge widths, so separate plots of one model are directly comparable.
rENA (backend is "plotly" or "qeviz"):
p <- ena.plot(set, title = "FirstGame", backend = backend)
p <- ena.plot.network(p, network = colMeans(lw[first, ]))
p <- ena.plot.group(p, pts[first, ], labels = "FirstGame", colors = "red",
confidence.interval = "box")qeviz directly:
p <- qe_plot(set, title = "FirstGame", group_col = "Condition",
scale_points = FALSE, range = "network") |>
qe_axes() |>
qe_colors(FirstGame = "#FF0000") |>
qe_edges(group = "FirstGame") |>
qe_group("FirstGame")A subtraction with both means
Click the figure for full size.
Colour and width are on the model-wide scale, so the small differences in a subtraction draw lighter than a group’s network; multiplier (rENA) or magnify (qeviz) enlarges them.
rENA (backend is "plotly" or "qeviz"):
p <- ena.plot(set, title = "FirstGame - SecondGame", backend = backend)
p <- ena.plot.network(p, network = colMeans(lw[first, ]) - colMeans(lw[second, ]),
colors = c("red", "blue"))
p <- ena.plot.group(p, pts[first, ], labels = "FirstGame", colors = "red",
confidence.interval = "box")
p <- ena.plot.group(p, pts[second, ], labels = "SecondGame", colors = "blue",
confidence.interval = "box")qeviz directly:
p <- qe_plot(set, title = "FirstGame - SecondGame", group_col = "Condition",
scale_points = FALSE, range = "network") |>
qe_axes() |>
qe_colors(FirstGame = "#FF0000", SecondGame = "#0000FF") |>
qe_edges(group = "FirstGame", compare = "SecondGame") |>
qe_group("FirstGame", "SecondGame")Units and means
Click the figure for full size.
Each mean keeps its own confidence interval.
rENA (backend is "plotly" or "qeviz"):
p <- ena.plot(set, title = "Units and means", scale.to = "points", backend = backend)
p <- ena.plot.points(p, pts[first, ], colors = "red")
p <- ena.plot.points(p, pts[second, ], colors = "blue")
p <- ena.plot.group(p, pts[first, ], labels = "FirstGame", colors = "red",
confidence.interval = "box")
p <- ena.plot.group(p, pts[second, ], labels = "SecondGame", colors = "blue",
confidence.interval = "box")qeviz directly:
p <- qe_plot(set, title = "Units and means", group_col = "Condition",
scale_points = FALSE, range = "points") |>
qe_axes() |>
qe_colors(FirstGame = "#FF0000", SecondGame = "#0000FF") |>
qe_edges(show = FALSE) |>
qe_points() |>
qe_group("FirstGame", "SecondGame")Generated by data-raw/plot-backends-figures.R; re-run it after changing either backend.