amscale: Aldrich-McKlevey Scaling

In philswatton/polimetrics: Miscellaneous functions useful for political science

Aldrich-McKlevey Scaling

Description

Performs Aldrich-McKlevey's (1977) scaling method for perceptual data using the QR decomposition method (Swatton 2021) for computation. The function will automatically filter missing data. Inspired by the aldmck function from the basicspace R package (Poole et al 2016).

Usage

amscale(x, resp = NULL, polarity = NULL)

Arguments

x	A dataframe or matrix containing numeric values of respondent placements of stimuli, with stimuli on columns and respondents on rows.
resp	An optional numeric vector containing respondent self-placements.
polarity	An optional integer giving the column index of a stimulus that you wish to have a negative value. All stimuli and respondent self-placements will also be coded accordingly.

Value

An object of class "amscale". The function summary prints a summary of the model and its results. An object of class "amscale" contain the following elements:

stimuli	Double vector containing scaled stimuli estimates.
respondent	Dataframe of the same length as the input containing respondent intercepts, weights; and if resp was not NULL respondent ideal points.
fit	The adjusted fit statistic proposed in Aldrich and McKelvey's paper (1977) for the model.
ninput	The number of respondents in the initial input.
nresp	The number of respondents used for scaling the stimuli in the model after filtering for missing data.
nstim	The number of stimuli scaled in the model.

References

\insertRef

aldrich1977psmisc

\insertRef

poole2016psmisc

\insertRef

swatton2021psmisc

philswatton/polimetrics documentation built on Jan. 30, 2023, 3:21 p.m.