Category Archives: Teaching

Multidimensional Poverty Measurement Methodologies


Sabina Alkire

  • Approaches to multidimensional poverty measurement
  • Dashboard approach; composite indices; venn diagrams; dominance approach;
    statistical approaches; fuzzy sets approach; axiomatic approach


Video (with guide)

Guide to the video

00.00: Introduction

04.15: Overview of approches to multidimensional poverty measurement

05.00: Dashboard approach

09.30: Dashboard approach – advantages and disadvantages

20.14: Composite indices

23.15: Composite indices –  advantages and disadvantages

26.21:  Joint distribution

27.16:  Venn diagrams

31.12:  Venn diagrams – advantages and disadvantages

32.40: Dominance approach

37.50: Dominance approach – advantages and disadvantages

41.50: Statistical approaches

54.00: Statistical approaches – advantages and disadvantages

01.00.45: Fuzzy sets approach

01.08.20: Fuzzy sets approach – advantages and disadvantages

01.14.25: Axiomatic approach

01.20.10: Axiomatic approaches – advantages and disadvantages

01.20.30: Concluding remarks

Further resources

Listen to a seminar on multidimensional poverty measurement methodologies by Suman Seth, November 2013.

Watch a presentation on this subject by Maria Emma Santos and Paola Ballon Fernandez at OPHI’s 2013 summer school in Washington, D. C.

Inequality among the Poor

Suman Seth

  • How to capture distributional issues in multidimensional poverty measures.
  • Approaches to computing inequality between and within groups.


Video (with guide)

Guide to the video


00:45 Main sources for lecture

1:25 Importance of distributional issues in multidimensional poverty

3:19 Relevant properties of multidimensional poverty

4:30 Examples of MD poverty matrices: concordant and discordant pairs

6:47 Dimensional transfer

12:44 Example of inequality among the poor using deprivation count vector

15:28 Two relevant practical properties of M0

16:32 Impossibility result

19:17 First approach using M0 and another poverty measure

21:30 Second approach using analysis on inequality separately

23:06 Example of Madagascar and Rwanda

24:30 Other inequality measures that can be used

25:16 Use of union approach

33:20 Deprivations versus attainments

34:16 Example using attainment score to reflect inequality

37:50 Elimination of inequality measures

38: 56 Decomposition between groups

40:08 Policy relevant property: within-group mean independence

43:39 The use of the variance as an inequality measure

47:53 Inequality decomposition: within group and between group component

48:30 Applications: Inequality among the poor and inequality across population subgroups

50:24 Disparity in poverty across subgroup

51:07 Example using Yemen, India , Togo and Bangladesh

53:20 Summary and concluding remarks

Robustness Analysis & Statistical Inference

boubaBouba Housseini Suman photoSuman Seth
  • Stochastic dominance for the Alkire Foster method
  • Rank robustness; Kendall’s Tau and Spearman
  • Rank concordance methods.
  • Computation of standard errors and confidence intervals


Video (with guide)

Guide to the video


Part 1 Robustness analysis

1:04 Sources for the lecture

3:28 Policy areas requiring robustness analysis

4:34 Importance of robustness analyses illustrated using Global MPI data

6:15 Implications of conclusions based on a sample

8:04 Parameters of M0 for robustness analysis: poverty cutoff, weighting vector and deprivation cutoffs

9:48 Rank robustness analysis

11:03 First Order unidimensional dominance

13:15 Dominance for H and M0 in AF

16:24 Calculation using two deprivation count vectors

21:10 Discussion of the M0 curve

23:24 Stochastic dominance conditions

26:42 Methods for comparing robustness of ranking

28:38 Kendall’s Tau

32:43 Spearman’s Rho

34:00 Some illustrations using the MPI: Robustness to weights

37:18 Robustness to poverty cutoff (k)

Part 2 Statistical inference

46:30 Estimation from samples about population level characteristics

47:18 Common concerns

49:20 Standard error (SE) and construction of confidence intervals (CI)

54:40 How to obtain the standard error

58:06 How to use a confidence interval

1:00:00 Difference between analytical SE and bootstrap methods

1:08:10 Concerns with sampling

1:11:38 Example using analytical method for SE using data from India

1:13:55 Hypothesis tests and equivalent procedures

1:19:50 Statistical inference in MPI comparisons

1:34:35 Conclusions

OPHI Summer School on Multidimensional Poverty Analysis 2015 ~ Washington D.C, USA


Organised by the Oxford Poverty and Human Development Initiative (OPHI), University of Oxford, this annual summer school on multidimensional poverty analysis will be be hosted at Georgetown University, Washington, DC, USA, 3-15 August 2015.

The purpose of this intensive summer school is to provide a thorough conceptual and technical introduction to some techniques of measuring multidimensional poverty with a strong emphasis on the Alkire Foster method. Participants will revise axiomatic poverty measures, and will learn about different techniques of multidimensional poverty measurement and which problems they are best suited to solve. The empirical motivation for measuring multidimensional poverty will be presented as well as the conceptual motivation, drawing on Amartya Sen’s capability approach.

The following topics will be covered:

  • Axiomatic approaches to unidimensional and multidimensional poverty;
  • Methodologies to analyse multidimensional poverty – dashboard, stochastic dominance, information theory,fuzzy set, multiple correspondence analysis, unmet basic needs and counting approaches – and the problems each methodology best solves;
  •  The Alkire Foster methodology of multidimensional poverty measurement;
  •  Selection of parameters – purpose, unit of measure, dimensions, indicators, cut-offs and weights;
  •  Estimation of multidimensional poverty and interpretation of the results;
  •  Subgroup decomposition and mapping;
  •  Multidimensional poverty dynamics;
  •  Disparity among the poor and across groups;
  •  Econometric analysis of multidimensional poverty;
  •  Institutions, policies, and communication.

The summer school will be led by the Researchers and Director of OPHI, University of Oxford. OPHI instructors include Sabina Alkire, Adriana Conconi, James Foster, John Hammock, Bouba Housseini, Gisela Robles Aguilar., Suman Seth, Yangyang Shen & Ana Vaz.

Course Format:

The summer school will consist of 10.5 days of instruction and working group sessions, taught in English. Each participant needs to bring a laptop with STATA with them to do the problem sets. Throughout the summer school, participants will be actively involved in discussions and working through problem sets, and will be invited to present their research work as well as share their experiences.

You can see readings, presentations, exercises and other materials from last year’s Summer School here.

Dates and Location:

The course runs from Monday 3 August to Friday 14 August with a final exam on Saturday 15 August (non- compulsory). Classes will be at Georgetown University, Washington, DC, USA.


The course fee is varied as follows:

  •  $750 USD for all students, and for academic researchers based in developing countries;
  •  $1500 USD for professionals based in developing countries; and
  •  $2500 USD for professionals based in developed countries.
  • In addition, participants will need to pay for accommodation (from $31 per night for a shared twin room, and from $46 per night for a single occupancy room, plus a single $34 linen cost per person for the two weeks), meals ($25/day for modest fare), travel fares and cost of STATA software programme. Additionally, please note a USA visa may need to be applied for and obtained at the participant’s cost (more details can be found on the US Visas website.

Course Application Information


Audience The summer school is addressed to those who are working on, or actively interested in gaining skills in, multidimensional poverty measurement; particularly, academics, post-graduate students, professionals and government officials. Applicants who are currently pursuing a research project or work on measurement are particularly welcome. Other applicants having a demonstrated research interest in empirical analysis in these topics will be considered on the basis of their experience and space availability.
Prerequisites A demonstrable knowledge of STATA is an absolute pre-requisite for attending the course. Every attendant at the summer school will need to have STATA 10 or higher installed on his/her laptop. The software is not provided by the summer school.
Where possible, participants should have a strong knowledge of quantitative methods (e.g. econometrics, statistics etc.), and a strong interest in poverty measurement and analysis.
Financial Support Limited partial and full financial support will be available. Competition for financial support will be very strong; however, well qualified and committed applicants with financial need are strongly encouraged to apply. Successful applicants are also highly encouraged to seek support from their local governments and institutions. We would be happy to provide support letters for these funding applications.
Application Form Please note application is now closed. Applicants will be evaluated on the basis of the information provided in their application.
Important Dates – 23 March 2015: Deadline for applicants requesting financial support
– 30 March 2015: Final deadline
– 14  April 2015: First acceptance emails will be sent starting from this date
Questions? Please write to James Brown: 


Alkire Foster en Stata (Mapas)

Mauricio Apablaza
Paola Ballon
  • Creación de mapas en Stata

Ver el video

Escuchar audio



Escuchar audio


Introduccion al enfoque de Capacidades

Sabina Alkire

  • La historia y motivación del trabajo de Sen
  • Definición de capacidades, funcionalidades, libertad y agencia
  • La relación entre el enfoque de capacidades y temas de medición de pobreza

Ver el video

Escuchar audio



Escuchar audio


Revisión del Curso

Sabina Alkire

  • Literatura clave sobre mediciones de pobreza multidimensional

Ver el video

Escuchar audio



 Escuchar audio


Instituciones y Políticas Públicas

John Hammock

  • Consideraciones prácticas sobre políticas públicas en la implementación en la implementación de mediciones de pobreza multidimensional
  • Objetivos políticos: usuarios he incentivos creados
  • El proceso de desarrollar una medición. ¿Cómo será usada? ¿Quién estaría interesado en usarla?
  • ¿Cómo se actualizan las mediciones? ¿Qué instituciones deberían participar?

Ver el video

Escuchar audio



Escuchar audio


Comunicardo el IPM

Sabina Alkire

  • Key literature on multidimensional poverty measures

Ver el video

Escuchar audio



Escuchar audio



Paola Ballon

  • Factor analysis; latent variable analysis help to define weights and final indicators.
  • ??

Ver el video

Escuchar audio



Escuchar audio