scplot 0.7.0

Bug fixes

• Fixed bug that did not allow to set pvar to something else than “phase” and add a statline at the same time.

New function

• split_dataline(): This function allows to split the dataline of a treatment phase into two parts based on a treatment variable in a dataset. This is mainly used to visualize alternating treatments in a phase (e.g. alternating treatments in a B phase).

scplot(scan::example_atd) |> 
  split_dataline("treatment")

New features / arguments

• set_dataline(): new argument “show_gaps”. Logical. If TRUE, missing values in the data will result in gaps in the line. If FALSE, missing values will be ignored and the line will be drawn continuously.
• Improved the automatic naming of statlines when the phase argument is set. (e.g. “mean”, phase = “A” now creates a statline with the name “mean values A”).
• Themes that have been created with scplot prior version 0.6.2 (including 0.6.1) must be recreated by executing the respective functions due to the adoption to S7 classes.
• add_statline(): trends can now be plotted for any phase or phase combination.
• add_statline(): Four regression methods now apply to all trend settings.
• add_stalline(): phase = “all” calculates a statistic for all values.
• New argument segmented for add_statline to indicate if there is a gap in the statline between phases.
• New argument case specifying for which cases a statline should be added.
• New experimental option to include staircase style phase separators that go across plots

scplot(exampleA1B1A2B2) |> 
  scplot::set_theme("basic", "tiny", "staircase") |> 
  set_yaxis(c(NA,50))
  
scplot(exampleABC) |> 
  set_separator(staircase = TRUE, linewidth = 2, linetype = "solid")  

Multiple new themes:

‘clarity’, ‘journal’, ‘poster’, ‘handout’, ‘sparse’,‘midnight’, ‘playful’, ‘atelier’, ‘bauhaus’, ‘sienna’

scplot 0.6.1

• Compatible with the new ggplot2 version which changed to S7 classes.
• Themes that have been created with scplot prior version 0.6.1 must be recreated by executing the respective functions due to the adoption to S7 classes.
  
• Extended the roxygen documentation for several functions.

scplot 0.6.0

• New function scplot_rand(): Takes an scdf and a statistic from the rand-test() function and returns a plot with the repective statistical results for varying phase B start positions.

scplot_rand(byHeart2011[1:5], "Slope B-A")

• Fixed bug that did did not allow to set the dvar and add a statline.

scplot 0.5.1

• Fixed bug that did not allow to draw plots for randomization test results.

scplot 0.5.0

• Deprecated: add_legend() argument label.
• New label argument for the set_dataline() and add_statline() arguments. This sets the characte string that is used in a legend.
• New generic for scplot().
• New function scplot.sc_rand(): Create histogram or xy plots for rand_tests.
• New function scplot.sc_tauu(): Create a forestplot for Tau-U analyses. scplot(tau_u(Leidig2018), effect = "A vs. B - Trend A")
• New function scplot.sc_hplm(): Create a forestplot for hplm analyses. scplot(hplm(Leidig2018), effect = "intercept")

scplot 0.4.0

• New function add_line(): Allows to add ab lines, horizontal or vertical lines.
• Solved: set_phasenames now works when phases are not coded as factors
• Solved: Removed connecting lines that occurred with identical phase names
• “trendA theil-sen” now represents theil-sen regresssion line (not Siegel)
• New theme “illustration”
• New default for scplot(): When an scdf includes an info and/or author attribute this is printed as a footnote/ caption (to remove it: scplot(example_A24) |> add_caption(NULL))

scplot 0.3.3

• moved scan dependency to scan import.

scplot 0.3.1

• solved bug: “trendA” and “trendA theil-sen” now correctly represent the extrapolated trend line of phase A.

scplot 0.3

new functions:

• new_theme(): creates a new scplot-theme.

new_theme <- new_theme() |> 
  set_xaxis(color = "brown") |>
  set_yaxis(color = "sienna3") |>
  set_ylabel(color = "sienna3", angle = 0) |>
  set_xlabel(size = 1, color = "brown") |>
  add_title(color = "sienna4", face = "italic") |>
  set_casenames(color = "steelblue4", size = 0.7) |>
  set_phasenames(size = 0) |> 
  set_panel(fill = c("grey80", "grey95"), color = "sienna4") |>
  add_grid(color = "grey85", size = 0.5) |>
  set_dataline(
    size = 0.5, linetype = "solid", 
    point = list(colour = "sienna4", size = 0.5, shape = "circle")
  ) |>
  set_separator(size = 0.5, linetype = "solid", color = "sienna") |>
  set_background(fill = "grey99") 

scplot(exampleABC) |> set_theme(new_theme)

• extract_theme(): returns the theme of an scplot object (instead of creating a plot). This theme can be reused in other plots with the add_theme() function.

new_theme <- scplot() |> 
  set_xaxis(color = "brown") |>
  set_yaxis(color = "sienna3") |>
  set_ylabel(color = "sienna3", angle = 0) |>
  set_xlabel(size = 1, color = "brown") |>
  add_title(color = "sienna4", face = "italic") |>
  set_casenames(color = "steelblue4", size = 0.7) |>
  set_phasenames(size = 0) |> 
  set_panel(fill = c("grey80", "grey95"), color = "sienna4") |>
  add_grid(color = "grey85", size = 0.5) |>
  set_dataline(
    size = 0.5, linetype = "solid", 
    point = list(colour = "sienna4", size = 0.5, shape = "circle")
  ) |>
  set_separator(size = 0.5, linetype = "solid", color = "sienna") |>
  set_background(fill = "grey99") |>
  extract_theme()

scplot(exampleABC) |> set_theme(new_theme)

fixed bugs

• solved: scdf with dulicated casenames crash

existing functions:

• ggplot2 deprecated the size argument for lines and replaced it with linewidth. We adapted several functions to follow this change which now use the linewidth argument instead of size: set_dataline(), add_dataline(), add_statline(), set_grid().

• add_statline(): New type “bar”
  

• set_xaxis() set_yaxis(): New argument expand (see ggplot scale_continuous).

• set_xaxis(): Extended Argument position with strip-right and strip-top

scplot 0.2

• replaced all dplyr functions with base R functions.
• each scplot function is now in a separate file.

scplot 0.1.0

New experimental function for ploting

The new scplot() function is here! It allows for a more tidy coding and the use of |> (or |>) operators. scplot is in an experimental state and code with current syntax might not work in a later version due to changes in function and argument names. Still, scplot() works in many cases. We plan to add new graphical features primarily to scplot which is already capable of doing more than plot.scdf().

Here is an example that implicitly also introduces several of the new graphical functions:

scplot(exampleABAB) |> 
  add_statline(stat = "trendA", colour = "tomato2") |>
  add_statline(stat = "maxA", colour = "lightblue") |>
  add_marks(case = 1:2, positions = 14, colour = "red3", size = 3, shape = 4) |>
  add_marks(case = "all", positions = 'points < quantile(points, 0.1)', colour = "blue3", size = 2) |>
  add_marks(positions = outlier(exampleABAB), colour = "brown", size = 2) |>
  add_labels(colour = "sienna") |>
  set_xaxis(increment = 4, size = 0.7, colour = "brown") |>
  set_yaxis(limits = c(0, 50), colour = "sienna3", size = 0.7) |>
  set_ylabel("Points", colour = "sienna3", size = 0.7, orientation = 0) |>
  set_xlabel("Weeks", size = 0.7, colour = "brown") |>
  add_title("Points by week", colour = "sienna4", size = 1.5, font = 3) |>
  set_phasenames("Baseline", "Intervention", "Fall-Back", "Intervention", cex = 1, colour = "darkgreen") |>
  add_theme("tiny") |>
  set_background(c("grey94", "grey99")) |>
  add_grid(colour = "grey85", width = 0.5) |>
  add_frame("sienna4") |>
  set_dots("sienna4", size = 1, shape = 18) |>
  set_line("black", width = 1, type = "dotted") |>
  add_text(case = 1, x = 5, y = 35, "Wow!!", colour = "darkgreen", angle = 20) |>
  add_text(case = 1, 1, 22, "PND", colour = "darkblue", size = 1.3) |>
  add_text(case = 1, 4, 8, "Trend", colour = "tomato", size = 1.3) |>
  add_arrow(case = 1, 5, 30, 5, 22, colour = "steelblue") |>
  add_ridge("white") |>
  set_casenames("MY", "FUNNY", "VALENTINE", colour = "steelblue4", size = 0.6) |>
  add_box("sienna1", width = 2) |>
  set_separator(extent = 0.9, width = 0.5, type = "solid", colour = "sienna")