plot.md.counts: Plot best, worst, and net counts from MaxDiff data

In cnchapman/choicetools: Tools for Choice Modeling, Conjoint Analysis, and MaxDiff analysis of Best-Worst Surveys

Plot best, worst, and net counts from MaxDiff data

Description

This plot shows how often each item was chosen as best and as worst, and shows the difference between those (the net count). This is a simple descriptive alternative to logit model estimation of MaxDiff values and generally corresponds extremely closely with aggregate and upper-level model estimates from multinomial regression and hierarchical Bayes estimation.

Usage

## S3 method for class 'md.counts'
plot(md.define, item.disguise = FALSE)

Arguments

md.define	A MaxDiff study object with $md.block data, as imported by read.md.qualtrics() or read.md.cho().
item.disguise	Remove the item labels and number them generically. This is included to make it easy to share and show results at a conference or other audience when the underlying details are confidential.

Details

In the case that some items were shown more or less often than others, regardless of whether they were chosen as best or worst or neither, the counts here are rescaled to make the relative proportions equivalent. For example, if one item appeared 1000 times but other items appeared only 800 times each, then the counts for those other items would be multiplied by 1.25 to account for the effective "proportion of times shown", compared to the item with 1000 appearances.

Value

A ggplot2 chart with the best, worst, and net counts.

cnchapman/choicetools documentation built on May 28, 2023, 9:14 a.m.