devPSI: Deviation Action Scoring of PSI matrix

In Dustin-Wood/funfield: Functional Field Functions

Deviation Action Scoring of PSI matrix

Description

Will estimate how p's rating of outcomes of each action deviate within situation s deviated from their average rating of outcomes of different actions in that situation

Usage

devPSI(PSIdata, locX = c(4:ncol(PSIdata)))

Arguments

PSIdata	PSI-structured dataframe (should have 'p','s','i' as their first three features)
locX	location of expected outcome ratings (note: default generously assumes the prescribed 'c(p,s,i,...,Likelihood)' data structure) ratings provided at different portion of survey)

Details

This essentially forces 'deviation' scoring.

Note that when nI = 2, the deviation scores will be reflections of one another. In which case, keeping only the maximum i (which should be i=1) can be useful to help avoid artificial inflation of sample size (although other strategies, like multilevel modeling in lavaan, may be preferable to address this issue).

If we want the difference of the two actions (rather than deviation scoring), we can pair the devPSI function with diffPSI.

Value

deviated ratings of effects of action i as rated by p within situation s from the person's average rating of actions in that situation.

Examples

(develop some example code)

Dustin-Wood/funfield documentation built on July 20, 2023, 7:10 a.m.