mp_nicheness {manifestoR}

Party nicheness measures

Description

Computes party nicheness measures suggested by Bischof 2015 and Meyer and Miller 2013.

Usage

mp_nicheness(data, method = "bischof", ...)

nicheness_meyer_miller(
  data,
  groups = meyer_miller_2013_policy_dimensions(),
  transform = NULL,
  smooth = FALSE,
  weights = "pervote",
  party_system_normalization = TRUE,
  only_non_zero = TRUE
)

nicheness_bischof(
  data,
  out_variables = c("party", "date", "specialization", "nicheness", "nicheness_two"),
  groups = bischof_issue_groups(),
  diversification_bounds = c(0, rep(1/length(groups), length(groups)) %>% {
     -(. *
    log(.))
 } %>% sum()),
  smooth = function(x) {
     (x + lag(x, default = first(first(x))))/2
 }
)

Arguments

data

a dataframe or matrix in format of Manifesto Project Main Dataset

method

choose between bischof and meyermiller

...

parmaeters passed on to specialized functions for differnet methods

groups

groups of issues to determine niches/policy dimensions; formatted as named lists variable names. For Meyer & Miller: Defaults to adapted version of Baeck et. al 2010 Policy dimensions (without industry, as used in the original paper by Meyer & Miller). For Bischof: defaults to issue groups used in the Bischof 2015 paper

transform

transform to apply to each of the group indicators. Can be a function, character "bischof" to apply log(x + 1), or NULL for no transformation.

smooth

Smoothing of policy dimension values before nicheness computation, as suggested and used by Bischof 2015

weights

vector of the length nrow(data) or the name of a variable in data; is used to weight mean party system position and nicheness; defaults to "pervote" as in Meyer & Miller 2013

party_system_normalization

normalize nicheness result within election (substract weighted mean nicheness)

only_non_zero

When dividing by the number of policy dimensions used for nicheness estimation, ignore dimensions that are zero for all parties (election-wise)

out_variables

names of variables to return in data.frame. Can be any from the input or that are generated during the computation of Bischof's nicheness measure. See details for a list.

diversification_bounds

Bounds of the range of the diversification measure (Shannon's entropy $s_p$ in Bischof 2015), used for inversion and normalization; default to the theoretical bounds of the entropy of a distribution on 5 discrete elements. If "empirical", the empirical max and min of the diversification measure are used

Details

List of possible outputs of nicheness_bischof:

diversification: Shannon's entropy $s_p$ in Bischof 2015

max_divers: used maximum for diversification

min_divers: used minimum for diversification

specialization: inverted diversification

specialization_stand: standardized specialization

nicheness: nicheness according to Meyer & Miller 2013 without vote share weighting

nicheness_stand: standardized nicheness

nicheness_two: sum of nicheness_stand and specialization_stand as proposed by Bischof 2015

References

Bischof, D. (2015). Towards a Renewal of the Niche Party Concept Parties, Market Shares and Condensed Offers. Party Politics.

Meyer, T.M., & Miller, B. (2013). The Niche Party Concept and Its Measurement. Party Politics 21(2): 259-271.

Baeck, H., Debus, M., & Dumont, P. (2010). Who gets what in coalition governments? Predictors of portfolio allocation in parliamentary democracies. European Journal of Political Research 50(4): 441-478.