Measures - Overview
Measure · API Affective Polarization Index affective mass vertical Measure · Behavioral Games Behavioral Measure of Intergroup Polarization affective mass horizontal Measure · Coefficients of Agreement Coefficients of Agreement ideological issue Measure · Dispersion (Weighted) Party System Dispersion ideological elite Measure · Distance (Weighted) mean distance from the most-liked party affective mass vertical Measure · Other Polarization Measurements Other Polarization Measurements ideological issue affective elite horizontal mass vertical Measure · Party Dyads Party Dyads and In-Out-Party Differences ideological affective mass vertical Measure · Party-System Compactness Party-System Compactness ideological elite Measure · Party-System Extremism Party-System Extremism ideological elite Measure · Polarization Index CSES Polarization Index ideological elite mass Measure · Range Range Between Outer-Right and -Left Parties ideological elite Measure · SD Standard Deviation ideological elite mass Measure · Social Distance Scale Closest Acceptable Relationship with Outparty Voters affective mass horizontal Measure · Spread (Weighted) Spread of like-dislike scores affective mass vertical Measure · trait-based Trait-Based Affective Polarization affective mass horizontal Measure · Variance Weighted Variance ideological elite

Polarization measure · Dispersion

(Weighted) Party System Dispersion

Description

Ezrow (2007) proposed the (Weighted) Party System Dispersion measure, which is basically a weighted standard deviation formula. Both Ezrow (2007) and Dreyer and Bauer (2019) use MARPOR data to measure party positions, but in principle it can be used with other datasets that contain the required information, i.e. party positions.

Operationalization

The (weighted) party system dispersion is measured as follows:

$Weighted~party~system~dispersion = \sqrt{\sum_{j=1} VS_{j} (P_{jk} - \bar{P}_k)^2}$

where $P_{jk}$ is the ideological position of party $j$ in country $k$ and $\bar{P_k}$ is the weighted average of the left-right ideological positions of all parties in country year $k$. $VS_j$ is the vote share of party $j$ in the last national election.

polaR

# Import Data
cses_imd <- polaR_import(source = "cses_imd",
						 path = "path/to/dataset.dta")

# Where different issue dimensions are available, 'issue' can be issued to specify it
cses_imd <- dispersion(cses_imd, 
					   issue = "leftright")

​

Visualization

Use cases

Publication that use this measure:

TitleAuthors
Understanding Polarization
  • Bramson et al. (2017)
How do masses react to party polarization?
  • Çakir (2024)
Polarization and ideological congruence between parties and supporters in Europe
  • Caroll and Kubo (2018)
Does voter polarisation induce party extremism?
  • Dreyer and Bauer (2019)
The Variance Matters
  • Ezrow (2007)
A changing climate change climate? A meta-analysis of climate change attitudes and polarization in the Netherlands spanning four decades
  • Peelen and Tolsma (2026)