Ethnic Residential Segregation: A Multilevel, Multigroup, Multiscale Approach Exemplified by London in 2011

Abstract

We develop and apply a multilevel modeling approach that is simultaneously capable of assessing multigroup and multiscale segregation in the presence of substantial stochastic variation that accompanies ethnicity rates based on small absolute counts. Bayesian MCMC estimation of a log-normal Poisson model allows the calculation of the variance estimates of the degree of segregation in a single overall model, and credible intervals are obtained to provide a measure of uncertainty around those estimates. The procedure partitions the variance at different levels and implicitly models the dependency (or autocorrelation) at each spatial scale below the topmost one. Substantively, we apply the model to 2011 census data for London, one of the world’s most ethnically diverse cities. We find that the degree of segregation depends both on scale and group.

Spatial Polarization of Presidential Voting in the United States, 1992–2012: The “Big Sort” Revisited

“…A method for doing this, based in multilevel modelling procedures with Bayesian statistical properties, has recently been developed (Jones et al, 2015;Manley et al, 2015aManley et al, , 2015bJohnston et al, 2016). Its index of segregation -the Median Rate Ratio (MRR) -shows its level at each scale net of the larger scales within which the smallest observation units are nested.…”

“…Those MRRs have associated asymmetric Bayesian credible intervals (CIs), with which assessments of the robustness of differences between pairs of values can be made. Studies of ethnic residential segregation employing the MRR to date have reported analyses of multi-group situations (Jones et al, 2015, Manley et al, 2015a, 2015b; they compare the distributions of more than two ethnic or occupational groups in the cities analysed and the formal modelling treats their distribution as the outcome of a Poisson process. For the current study of polarisation we are dealing with binary categorisationsDemocrat and Republican voters -and so the modelling assumes that their distributions are the outcome of a Binomial process.…”

“…We apply this method (full details of which are in Jones et al, 2015) to a bespoke data set giving the percentage of the votes for the Democrat and Republican candidates that were cast for the former at each presidential election from 1992 to 2012, inclusive, for 3,077 Counties (or the equivalent unit), nested within the forty-nine States and the nine Census Divisions.…”

Multiscale Measures of Population: Within- and between-City Variation in Exposure to the Sociospatial Context

“…The relevance of scale has been well established for the segregation literature (see, e.g., White 1983;Wong 2004;W. A. V. Clark et al 2015;Jones et al 2015), the neighborhood effects literature (Galster 2001; R. Andersson and Musterd 2010;Vall ee et al 2015), and, more broadly, research on sociospatial inequalities (Suttles 1972;Manley, Flowerdew, and Steel 2006;Prouse et al 2014), where scale is often addressed as one aspect of the modifiable areal unit problem (MAUP; see Openshaw and Taylor 1979;Manley 2014). Sociospatial inequalities can be more fully understood by exploring variation in geographic contexts across multiple scales, within the so-called spatial opportunity structure, rather than by confining to a single geographic context (Galster and Sharkey 2017).…”

The Growing Spatial Polarization of Presidential Voting in the United States, 1992–2012: Myth or Reality?

“…A full evaluation of any evolving spatial pattern thus requires exploration of trends at a variety of scales. However, as pointed out some decades ago (Duncan, Cuzzort, and Duncan 1961 ) but rarely taken into consideration since, any measure of segregation at one spatial scale necessarily incorporates its measure at any larger scales: if there is growing polarization at the state scale, for example, this is bound to be incorporated-to an unknown extent-at the county scale too and, as Jones et al ( 2015 ) have argued, any measure of segregation at a micro-scale is likely to be over-stated if it does not "hold constant" that measure at a macro-scale within which the micro-scale units are nested (e.g. counties within states).…”