Party-System Compactness

Description

Party-system compactness, proposed by Alvarez and Nagler (2004), measures the dispersion of voters’ ideological positions relative to the dispersion of parties’ positions. Alvarez and Nagler apply the measure to data from the American National Election Study (ANES), the British Election Study (BES), and the Canadian Election Study (CES). In principle, it can also be applied to other survey data, such as the CSES.

Operationalization

Alvarez and Nagler calculated an unweighted and a weighted version of the measure, the latter using the parties’ vote shares as weights. Their measures of party-system compactness are calculated as follows: $$CP_k = \frac{\sigma_k}{max|(P_{jk}-P_{lk})|} \forall j,l$$ where $CP_k$ is the party-system compactness on issue $k$, $P_{jk}$ is the position of party $j$ on issue $k$, $P_{lk}$ is the position of party $l$ and issue $k$, and $\sigma_k$ is the standard deviation of respondents’ position on issue $k$. The weighted version is calculated as follows: $VWCP_k = \frac{\sigma_k}{\sum_{j=1}^N V_j|(P_{jk}-\bar{P_{k}})|}$ where this formula uses the same concepts and adds $V_j$, which is the vote share of the $j$th party.

Use cases

Publication that use this measure: