Usage
pirateye(
data,
colour_condition = NULL,
x_condition = "variable",
facet_condition = NULL,
facet_scales = "fixed",
facet_row = NULL,
pid_average = F,
norm_pid = "no",
dv,
reorder = "no",
pid = "pid",
dodgewidth = 0.8,
plot_condition = NULL,
cond = NULL,
cond2 = NULL,
facetby = NULL,
ylim = NULL,
xlim = NULL,
w = NULL,
h = 6,
title = NULL,
outp = "analysis",
cols = NULL,
pred_line = F,
error_bar_data = NULL,
ypercent = F,
x_axis = NULL,
y_axis = NULL,
error_dim = F,
error_dim_value = 0,
pred = NULL,
pred_means = NULL,
pred_bar = T,
xlabs = NULL,
xlabpos = 0.7,
error_data = NULL,
cflip = F,
norm = F,
bars = F,
violin = T,
dots = T,
splitV = F,
svw = 1,
dot_h_jitter = 0,
line = F,
error_bars = T,
useall = F,
legend = T,
title_overide = F,
combine_plots = list(),
combine_position = "right",
elementinc = NULL,
type = NULL,
...
)Arguments
- data
data with one person per line, excluding rows with use=0
- colour_condition
colour split
- x_condition
x axis split (if not, specified colour condition used for x axis too)
- facet_condition
for faceting
- pid_average
plot an average of each participant's dv over the named conditions
- norm_pid,
normalise / z score for each participant, either "no","z" or "iq"
- dv
name of single dv column, or multiple columns, in which case they will be split by x_condition unless colour or facet condition set to 'variable'
- pid
whats the name of col that identifies individuals
- plot_condition
instead of passing individually, you can give a vector of up to 3
- cols
specify the colours to use, can be a set of colours, or of condition_level to colour
- error_bar_data
error bar data
- ypercent
Convert y scale to
x_axistitles
y_axistitles
xlabposHow high vertically should they be, as proportion of plot height
error_datadistribution for mean, eg from Bayes analysis, to replace SE
cflipflip to horizontal plot
normnormalise / z-score values for comparison across scales when multiple dvs
useallignore the use column and plot all rows
typeshortcuts: m=just error bars, b=just bars
redordercan be "increasing" or "decreasing", default "no"
xlabDo we have labels to go across x axis, such as post hoc pvalues or MPEs
Defaults to violin plot with error bars and dots, but elements such as dots, bars, violin, error_bars can be turned on or off