The bimsim function covers the informed simulation of the preference
test data, e.g., when an additional item was tested but has ties and/or
missing item combinations according to the ground truth. With the help of
intransitivity calculations a cutoff can be set to exclude any simulation
results that do not show transitivity.
bimsim( rawdat = NULL, GT = GT, simOpt = simOpt, filter.crit = "CE", limitToRun = 5, tcut = 0.9, deviation = 0, minQuantity = 0, seed = TRUE, showPlot = TRUE, ylim = c(0, 0.7) )
| rawdat | rawdata from the bimload function |
|---|---|
| GT | item list with the ground truth (GT; letters are case sensitive!) |
| simOpt | item to be checked (this can be an item from the GT or a new one. Remove the item from GT if specified here!) |
| filter.crit | the filter criterion for the plot (defaults to "CE": consensus error; alternatively the intransitivity ratio (Iratio) can be shown) |
| limitToRun | limit to this number of repetitions for the randomizations (default 5, you'll probably need more) |
| tcut | threshold (cutoff) for the filter.crit variable (default = 1 for 100% of the values); i.e., when filter.crit="CE" and tcut=0.9, the plot will show 90% of the consensus errors. When filter.crit="Iratio" and tcut =0.9, the plot will show items with transitivities > 90%. |
| deviation | how much deviation is allowed for marking ties (default is 0 which translates to 50%; deviation =5 adds 5% to 50% = 55%) |
| minQuantity | minimum quantity to be recognized as difference (default is 0) |
| seed | use constant seeding for stable plot results (default=TRUE) |
| showPlot | show simulation runs with consensus error bubbles |
| ylim | scale y axis of the worth plot (default: c(0,0.7)) |
list with the position frequencies and the plot object