Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: constructing a categorical gini measure

From   Austin Nichols <>
Subject   Re: st: constructing a categorical gini measure
Date   Wed, 30 May 2012 11:13:29 -0400

"Tope, Daniel" <>:
Also note that if you are willing to assume the population
distribution of education is drawn from a known family of
distributions you can estimate parameters that pin down Gini etc.
See e.g.
and references.

On Mon, May 28, 2012 at 4:41 AM,  <> wrote:
> Apparently you would like to measure inequality in "years of education",
> but all that is observed is data grouped into categories. And you have
> data on counties, not individuals.
> On the first issue, you could first convert your categorical variables
> to some "years of education" equivalent (your discipline and country
> conventions may tell you how to do that).   But you're still then stuck
> how to handle the grouping. On this see the discussion last week on the
> list, and track back on previous Statalist discussion using Google.
> A more fundamental question is: why calculate a Gini from these sorts of
> data?
> Stephen
> ------------------
> Professor Stephen P. Jenkins <>
> Department of Social Policy and STICERD
> London School of Economics and Political Science
> Houghton Street, London WC2A 2AE, UK
> Tel: +44(0)20 7955 6527
> Changing Fortunes: Income Mobility and Poverty Dynamics in Britain, OUP
> 2011,
> Survival Analysis Using Stata:
> Downloadable papers and software:
> ------------------------------
> Date: Sun, 27 May 2012 16:13:12 +0000
> From: "Tope, Daniel" <>
> Subject: st: constructing a categorical gini measure
> I would like to construct a county-level educational inequality gini
> measure. I have categorical data on educational attainment for each
> county. That is, I have the number and percentage of people in each
> county that have reached each level of education.
> My 16 educational attainment categories are: NoSchool00; Less than 4
> Years; 5-6 years; 7-8; 9th grade; 10th grade; 11th grade; 12th grade;
> High School Grad; some college - less than one year; some college - more
> than one year; Associates degree; Bachelors degree; Professional degree;
> Masters degree; and Doctorate.  I assume that each category must be
> converted into a years of schooling measure.
> Does anyone have any advice on how to construct such a gini measure for
> each indivdual county?

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index