Local and overall Segregation indices with optional local segregation curves (For multigroup cases see dicseg)
localseg unitvar groupvar [weights] [if exp] [in range] [, format(string) save sc nograph x(string) y(string) xtitle(string) ytitle(string) graph_options(string) ]
fweights, aweights and aweights are allowed; see weights.
Important
It requires to have matsort, written by Paul Millar, previously installed; if not, install it writing in the command line:
. net install matsort, all from(http://fmwww.bc.edu/RePEc/bocode/m)
Description
localseg computes local and overall segregation indices across units in a multigroup context, as proposed by Alonso-Villar and Del Río, 2010.
Micro data are required (data should not be collapsed by units).
The first variable unitvar indicates units (ex. occupations, census tracts, schools, ....).
The second variable groupvar identifies groups (gender, race, ...).
If option sc is specified it also draws the local segregation curves, and creates new variables using x(string) and y(string) options.
Reporting
Overall Segregation indices:
. Ip, multigroup index of dissimilarity.
. M, the mutual information index.
Note that M = GE(1); Mutual Information Index M [ with natural log instead of log(2) ].
. Gini, the unbounded version of the multigroup Gini index.
. Duncan, the Duncan and Duncan index of dissimilarity (only reported in the case of two groups).
Local Segregation indices:
. Ip, multigroup index of dissimilarity.
. GE(c), c=0, .10, .25, .50, .75, .90, 1, family of indices related to the Generalized Entropy family, for different values of the segregation sensitivity parameter.
Note that GE(0) is only defined for groups with members in all units.
Note that M = GE(1); Mutual Information Index M [ with natural log instead of log(2) ].
. Gini, variation of the Gini index.
The relative contribution of each group to overall segregation is also reported.
Segregation Curves (if option sc is specified).
Options
format(string) to change numeric format
save to save results
sc to compute segregation curves
nograph to do not graph segregation curves (if option sc was specified)
x(string) y(string) to create variables for segregation curves (if option sc was specified)
xtitle(string),ytitle(string) title for segregation curves (if option sc was specified)
graph_options(string) to change default graph options for segregation curves (if sc was specified)
Saved results
Matrices:
r(oseg) : overall segregation and relative contributions
r(lseg) : local segregation
Examples
. use segdata.dta, clear
. localseg occupation race [aw=pwgtp], save
. localseg occupation sex [aw=pwgtp], save sc
. localseg occupation race [aw=pwgtp], save sc x(cumem) y(cumtgt) xt(Cumulative employment) yt(Cumulative target workers)
Author
Carlos Gradín <cgradin@uvigo.es> Facultade de CC. Económicas Universidade de Vigo 36310 Vigo, Galicia, Spain.
References
For theory:
Alonso-Villar, O. and C. Del Río (2010), Local versus Overall Segregation Measures, Mathematical Social Sciences, vol. 60(1), pp. 30-38.
For applications:
Alonso-Villar, O., C. Del Río, C., and C. Gradín, C. (2010), The extent of occupational segregation in the US: Differences by race, ethnicity, and gender, Ecineq Working Paper Series 2010-180.
Del Rio, C. and Alonso-Villar, O. (2011), "Occupational Segregation of Immigrant Women in Spain", Feminist Economics (forthcoming).
Del Río, C. and Alonso-Villar, O. (2010), "Gender segregation in the Spanish labor market: An alternative approach", Social Indicators Research, vol. 98 (2), September, pp. 337-362
Gradín, C., O. Alonso-Villar, and C. Del Río (2010), Occupational segregation by race and ethnicity in the US: Differences across states, Ecineq Working Paper Series 2011-190.
Also see
dicseg if installed; duncan if installed; hutchens if installed; seg if installed