Using Bayesian Aldrich-McKelvey Scaling to Study Citizens’ Ideological Preferences and Perceptions

127

Some studies of policy representation test hypotheses about the relationship between citizens' views and elites' positions on multiple issues by proceeding one issue at a time. Others summarize citizens' and elites' preferences with “ideology scores” and test hypotheses with these. I show that approach is flawed. It misinterprets citizens' ideology scores as summaries of policy preferences, but these scores actually measure ideological consistency across areas: how often citizens' ideal policies are liberal or conservative. Examples show how attending to this distinction overturns conventional wisdom: legislators appear similarly moderate as citizens, not more extreme; however, politically engaged citizens appear especially moderate.

Section: Example: Studying Extremism With Ideological Scalesmentioning

confidence: 99%

Approaches to Studying Policy Representation

Broockman¹

2016

127

“…An important concern regarding our data, however, is the well‐known problem of response incomparability (i.e., respondents may interpret identical questions in different ways). To address this issue, we rely on the Bayesian implementation of the Aldrich‐McKelvey method used in Hare et al () to place legislators and voters on the same scale.…”

Section: Individual Estimates Of Ideologymentioning

confidence: 99%

“…The Aldrich‐McKelvey (A‐M) model assumes that given a set of respondents

I = ({}|, 1, \dots, n)

and a set of stimuli

J = ({}|, 1, \dots, m)

, the perceived location of stimulus j by individual i , denoted by

z_{i j}

, is given by

z_{i j} = λ_{i} + φ_{i} + Z_{j} + e_{i j},

where

Z_{j}

, is the true location of j ;

λ

is an intercept capturing a respondent's systematic bias in reported placements;

φ

captures any expansions/contractions of the reported placements on the scale; and

e_{i j}

is a random variable with zero expectation, positive variance that is independent of i and j , and zero covariance across the i s and j s (Aldrich and McKelvey ; Hare et al ). Using the

z_{i j}

matrix of reported positions, the A‐M scaling procedure recovers the location of the stimuli using singular value decomposition (SVD).…”

Section: Individual Estimates Of Ideologymentioning

confidence: 99%

See 1 more Smart Citation

Disentangling the Role of Ideology and Partisanship in Legislative Voting: Evidence from Argentina

Alemán

Micozzi

Pinto

et al. 2017

We present a novel approach to disentangle the effects of ideology, partisanship, and constituency pressures on roll‐call voting. First, we place voters and legislators on a common ideological space. Next, we use roll‐call data to identify the partisan influence on legislators' behavior. Finally, we use a structural equation model to account for these separate effects on legislative voting. We rely on public opinion data and a survey of Argentine legislators conducted in 2007–08. Our findings indicate that partisanship is the most important determinant of legislative voting, leaving little room for personal ideological position to affect legislators' behavior.

“… In reality, survey respondents often interpret scales differently, a problem known as differential item functioning (DIF) (see Brady ; King et al ). Typically, for a large public opinion survey, we could address this problem using Aldrich and McKelvey's () seminal method for correcting the perceptual biases of respondents in surveys (see also Hare et al ). It is not feasible for us to perform a similar rescaling exercise here because our data do not provide us with a perceptual fixed point with which to anchor the A‐M transformation.…”

mentioning

confidence: 99%

Measuring Legislative Power: An Expert Reweighting of the Fish‐Kroenig Parliamentary Powers Index

Chernykh

Doyle

Power

2016

The Parliamentary Powers Index (PPI) developed by Fish and Kroenig (2009) is the most important effort to date to measure legislative power in cross‐national perspective, but it has been criticized on both theoretical and methodological grounds. We build on the 32‐item PPI to develop an alternative indicator of legislative strength that is based on an expert survey of 296 political scientists in 2014. We reweight each of the powers by expert opinion, creating a new Weighted Legislative Powers Score (WLPS) for the 158 national legislatures in the Fish and Kroenig data set. In addition, the article reports the expert‐assigned weight factors for the entire set of 32 powers contained in the original PPI, thus allowing researchers to innovate alternative, disaggregated indicators of legislative power.