franzmann_kaiser {manifestoR}

Left-Right Scores based on Franzmann & Kaiser Method

Description

Computes left-right scores based on the Franzmann & Kaiser Method (see reference below). The issue structures are not calculated from scratch but taken as given from Franzmann 2009 (or later updates). Note that they are not available for the entire Manifesto Project Dataset, but only for a subset of countries and elections.

Usage

franzmann_kaiser(
  data,
  basevalues = TRUE,
  smoothing = TRUE,
  vars = grep("per\\d{3}$", names(data), value = TRUE),
  issue_structure = read_fk_issue_structure(mean_presplit = mean_presplit),
  party_system_split = split_belgium,
  mean_presplit = TRUE,
  ...
)

read_fk_issue_structure(
  path = system.file("extdata", "fk_issue_structure_2019.csv", package = "manifestoR"),
  mean_presplit = TRUE,
  format_version = 2
)

fk_smoothing(data, score_name, use_period_length = TRUE, ...)

Arguments

data

A data.frame with cases to be scaled, variables named "per..."

basevalues

flag for transforming data to be relative to the minimum

smoothing

flag for using smoothing

vars

Variables/Categories to use for computation of score. Defaults to all available handbook version 4 categories.

issue_structure

issue structure to use for Franzmann & Kaiser method, default to most recent bundled version (for details see read_fk_issue_structure)

party_system_split

function to recode the country variable to re-partition party systems. Defaults to splitting Belgium into two halfs as done in Franzmann 2009

mean_presplit

if TRUE, for Belgium as a whole (before the split into two party systems) the mean of the issue weights is used (which is equal to taking the mean of the output values, since all subsequent transformations are linear). This step is required to replicate the Franzmann 2009 dataset. If the issue structures already contain values for Belgium as a whole they are overwritten by the newly generated ones.

...

passed on to fk_smoothing and party_system_split

path

path from were to read issue structures (as csv data file). Defaults to the most recent file bundled in the manifestoR package.

format_version

can be 1 or 2 to switch between different structural versions of the issue structures file (1 for files containing "structure"-columns, 2 for files containing "per"-columns)

score_name

name of variable with LR Score values to be smoothed

use_period_length

whether to use electoral period length in weighting

References

Franzmann, Simon/Kaiser, Andre (2006): Locating Political Parties in Policy Space. A Reanalysis of Party Manifesto Data, Party Politics, 12:2, 163-188

Franzmann, Simon (2009): The Change of Ideology: How the Left-Right Cleavage transforms into Issue Competition. An Analysis of Party Systems using Party Manifesto Data. PhD Thesis. Cologne.