
Returns local segregation indices for each category defined
by unit
.
mutual_local(
data,
group,
unit,
weight = NULL,
se = FALSE,
CI = 0.95,
n_bootstrap = 100,
base = exp(1),
wide = FALSE
)
Returns a data.table with two rows for each category defined by unit
,
for a total of 2*(number of units)
rows.
The column est
contains two statistics that
are provided for each unit: ls
, the local segregation score, and
p
, the proportion of the unit from the total number of cases.
If se
is set to TRUE
, an additional column se
contains
the associated bootstrapped standard errors, an additional column CI
contains
the estimate confidence interval as a list column, an additional column bias
contains
the estimated bias, and the column est
contains the bias-corrected estimates.
If wide
is set to TRUE
, returns instead a wide dataframe, with one
row for each unit
, and the associated statistics in separate columns.
A data frame.
A categorical variable or a vector of variables
contained in data
. Defines the dimension
over which segregation is computed.
A categorical variable or a vector of variables
contained in data
. Defines the group for which local
segregation indices are calculated.
Numeric. (Default NULL
)
If TRUE
, the segregation estimates are bootstrapped to provide
standard errors and to apply bias correction. The bias that is reported
has already been applied to the estimates (i.e. the reported estimates are "debiased")
(Default FALSE
)
If se = TRUE
, compute the confidence (CI*100)
in addition to the bootstrap standard error.
This is based on percentiles of the bootstrap distribution, and a valid interpretation
relies on a larger number of bootstrap iterations. (Default 0.95
)
Number of bootstrap iterations. (Default 100
)
Base of the logarithm that is used in the calculation. Defaults to the natural logarithm.
Returns a wide dataframe instead of a long dataframe.
(Default FALSE
)
Henri Theil. 1971. Principles of Econometrics. New York: Wiley.
Ricardo Mora and Javier Ruiz-Castillo. 2011. "Entropy-based Segregation Indices". Sociological Methodology 41(1): 159–194.
# which schools are most segregated?
(localseg <- mutual_local(schools00, "race", "school",
weight = "n", wide = TRUE
))
sum(localseg$p) # => 1
# the sum of the weighted local segregation scores equals
# total segregation
sum(localseg$ls * localseg$p) # => .425
mutual_total(schools00, "school", "race", weight = "n") # M => .425
Run the code above in your browser using DataLab