Category Archives: Short Courses

Review of the Course

sabina_alkire

Sabina Alkire

  • Recapping key lessons and topics covered during the course

Audio 

Video (with guide)

 

Guide to the video

00:00: Introduction

01:06: OPHI at a glance

02:05: OPHI’s research – two themes

04:35: Why do we measure multidimensional poverty? (ethical, technical, empirical and policy motivations)

08.33: Unidimensional poverty measurement

10:20: Axioms of multidimensional poverty measurement

14:43: Methods of multidimensional poverty measurement and analysis

19:04: Global Multidimensional Poverty Index

23:48: Alkire Foster methodology

25:49: What is the Capability Approach?

29:16: Normative considerations for multidimensional poverty measures

34.05: Case studies – National multidimensional poverty measures

37:33: Data issues in multidimensional poverty measurement

40:39: Decomposition of multidimensional poverty measures

41:21: Changes in poverty over time

43:42: Chronic poverty measurement

44:45: Dominance and robustness

46:14: Standard errors and confidence intervals

47:20: Inequality among the poor

49:50: Regression analysis

51:45: Policy and communication

54:18: Targeting and impact

57:05: Institutions and policy

Institutions and Policy

 JohnHammock_100

John Hammock

mauricio

Mauricio Apablaza

  • Why use the MPI?
  • Policy considerations
  • Political considerations
  • How to use the MPI

Audio 

Video (with guide)

Guide to the video

00.00: Introduction

01.25: Why use the MPI?

06.00: Policy considerations

06.30: Political support

14.45: Policy applications of the MPI

29.10: Political considerations

32.30: Who needs to be on board?

35.00: Key stakeholders

38.20 Technical/policy considerations

43.00 National considerations

48.45: Observations of institutions responsible for creating, analysing and updating poverty measures

52.30 How to use the MPI – country examples

Targeting and Impact Evaluation

sabina_alkire

Sabina Alkire

Ana Vaz

  • Targeting poverty reduction programmes: Identifying the poor
  • Evaluating multidimensional impact of poverty reduction programmes
  • Introduction to resources on the OPHI website

Audio 

Video (with guide)

Guide to the video

00:00 Introduction

Part 1: Targeting

01.25: The challenge of targeting

07.34: Resources on the Alkire Foster(AF) method and targeting

09.04: How to identify a target group?

11.27: Census questions used for targeting

16.41: Requirements to design a targeting instrument

17.28: Tools of targeting

18.45: Which is the best proxy?

20.00: How many indicators to use?

21.52: How many are to be targeted in each region?

22.45: Conclusions

Part 2: Impact Evaluation

33.30: Motivation

36.07: Purpose – how can the AF method be used in impact evaluation?

36.25: Background

37.06: Why use the AF method?

39.35: How to use the AF method?

45.26: Empirical application

52.37: Sample

53.17: Differences at baseline

55.53: Impact

01.01.24: Decomposition

01.08.50: Resources on the OPHI website

Policy and communications

 JohnHammock_100

John Hammock

Paddy-Coulter_100

Paddy Coulter

  • Why communicate your measure?
  • How to communicate your measure?
  • The Alkire Foster method simplified
  • The Multidimensional Poverty Peer Network

Audio

Video (with guide)

Guide to the video

00.00: Introduction

01.20: Principles of effective communications

09.00: Why communicate your measure at all?

14.50: How to communicate your measure – challenges and principles

17.00: Starting points for communicating your measure: audiences, messages, channels

40.40: Finding factoids

50.20: The Alkire Foster method simplified

01.00.15: Common questions

01.12.00: The Global Multidimensional Poverty Peer Network (MPPN)

Multidimensional Poverty Dynamics with Time Series and Panel Data

adriana_100

Sabina Alkire

Ana Vaz

  • Descriptive analysis using time series data

  • Chronic poverty measurements using multidimensional data

  • Subgroup analysis using panel data

Audio 

Video (with guide)

Guide to the video

Introduction

00:44 Outline of lecture

Descriptive analysis using time series data

1:28 Introduction to time series analysis and importance of surveys

3: 46 Notation

4:49 The rate of absolute change

8:55 Reference periods in comparison of two countries

10:00 The rate of relative change

11:51 Example for annualized changes (absolute and relative) using data from Nepal, Peru, Rwanda and Senegal

15:34 Annualized absolute change in MPI across time

18:42 Performance calculated according to relative change

19:20 Accounting for population numbers

20:26 Rate of reduction in the headcount ratio

21:19 Importance of demographic shifts

23:44 Calculating and interpreting dimensional changes using the example of Nepal

30:10 Comparison of censored and uncensored headcount

36:00 Subgroup decompositions

Chronic poverty measurement using panel data

41:12 Introduction

42:45 Concept of chronic poverty in multidimensional data

44:34 Focus of lecture

45:20 Calculate M0 to measure chronic poverty using achievement matrix

54:40 Transient poor

1:00: 22 Dimensional indices for chronic multidimensional poverty

1:00:46 Period specific indices for chronic multidimensional poverty

1:01:08 Introduction to dynamic subgroup analysis

1:03:07 Analyses that can be conducted with panel data

1:03:29 Construction of subgroups and calculation of dynamic subgroup analysis

1:06:19 Interpreting an illustration of dynamic subgroup analysis

1:09:03 Importance for policy

1:15:12 Shapley decomposition for illustration

1:19:52 Q & A

Population Subgroup Decomposition and Dimensional Breakdown

Suman Seth

  • The calculation of decomposition by subgroup and dimensions.
  • The calculation of the contribution of each dimension’s censored headcount to overall poverty, as well as each subgroup’s.

Audio 

Video (with guide)

Guide to the video

Introduction

00:48 Focus of the lecture

5:15 Properties of subgroups in AF method

6:48 Population subgroup decomposability

8:38 Example of a calculation with achievement matrix

17:01 Demonstration with Nigeria and India data

20:34 Illustration of national and sub-national disparities evinced from 2011 MPI data

22:22 Clarification of basic concepts underlying dimensional breakdown: Dimensional decompositions, or the censored headcount for each dimensions. M1 and M2 have the same decomposition properties and calculations are similar.

33:19 Calculation using an achievement matrix

36:20 Computing censored headcount ration of each dimensional

37:23 Discussion of uncensored versus censored headcount ratio

38:36 Calculation of contribution of dimension to overall poverty

39:00 Relationship between M0 and uncensored headcount ratio in union approach

43:06 Calculation using example of achievement matrix

44:12 Using equal weights when calculating the contribution of each dimension

44:55 Example of countries with same MPI but different contributions of dimensions across indicators

Data Issues in Multidimensional Poverty Measurement

 IMG_1675

Adriana Conconi

Ana Vaz

  • Sources of multidimensional data
  • Household surveys: survey design
  • Design of indicators
  • Applicable population
  • Combined measures
  • Missing values and inconsistencies

Audio 

 

Video (with guide)

Guide to the video

00.00: Introduction

01.05: Outline

1. Sources of multidimensional data

03.00: Census

06.32: Administrative data

08.38: Household surveys

13.44: Household surveys: metadata

2. Household surveys: survey design

26.24: Sample weights

27.45: Samples and subsamples

31.25: Non-response rate and other non-sampling error

3. Indicators’ design

32.25: Unit-level indicator accuracy

38.32: Indicators transformation to match unit of identification

4. Applicable population

39.36: Applicable population

5. Combined measures

46.40: Combined measures

01.03.45: Assessing combined measures

6. Missing values and inconsistencies

01.11.55: Missing values and inconsistencies

01.16.10: How should we treat a missing value when computing the Adjusted Headcount Ratio?

01.16.48: Sample drop and bias analysis

Further resources

Watch a presentation on ‘Issues of Data Analysis’ by OPHI Research Associate José Manuel Roche at OPHI’s 2013 Summer School in Washington, D. C.

Normative issues in Multidimensional Poverty Measurement

 sabina_alkire

Sabina Alkire

  • Eight core choices for designing a multidimensional poverty measures:
  • Purpose of poverty measure; choice of space; units of identification and analysis; choice of dimensions; choice of indicators; deprivation cuttoffs; weights; poverty cutoff

Audio 

Video (with guide)

Guide to the video

00.00: Introduction

02.15: Diversity of people

03.45: What are ‘normative’ choices?

05.55: Normative reasoning

11.50: Overview of eight core choices for designing a multidimensional poverty measures

17.55: 1. Purpose of poverty measure

29.07: 2. Choice of space

31.00: 3. Unit(s) of identification and analysis

36.27: 4. Choice of dimensions

52.09: Dimensions used in national MPIs of Colombia and Mexico

57.35: 5. Choice of indicators

01.01.54: 6. Deprivation cuttoffs

01.06.50: Field studies in Bhutan

01.11.20: 7. Weights

01.24.05: 8. Poverty cutoff

Properties of Multidimensional Poverty Measures

Suman photo

Suman Seth

  • Two major steps of measurement: Identification and aggregation
  • Classification of properties of multidimensional measures
  • Invariance, dominance, subgroup and technical properties

Audio 

Video (with guide)

Guide to the video

01.48: Main sources of the lecture

02.30: Preliminaries

09.46: Measurement step 1: Identification – who is poor?

20.15: Measurement step 2: Aggregation – how poor is the society?

20.45: Classification of properties of multidimensional measures

Invariance properties

24.00: Symmetry

25.00 Replication invariance

27.14: Scale invariance

28:40 Focus: poverty focus and deprivation focus

45.40: Focus axioms and types of identification

59.20: Ordinality

Dominance properties

01.08.24: Monotonicity

01.10.42: Dimensional Monotonicity

01.15.05: Transfer in unidimensional context

01.17.01: Bistochastic matrix

01.19.05: Uniform majorisation

01;19.55: Transfer

01.21.17: Rearrangements

01.35.10: Subgroup properties

01.37.24: Technical properties

Why Multidimensional Poverty Measures?

sabina_alkire

Sabina Alkire

  • Overview of technical, empirical and policy
    motivations for multidimensional poverty measures

Audio 

Video (with guide)

Guide to the video

Introduction

00.18: Ethical motivations for multidimensional poverty measures

02.00: What is poverty? Who is poor?

Part 1: Technical motivations

05.57: Relevant data are increasing

10:00: Computational and methodological developments

Part 2: Empirical motivations

12:24 Monetary and non-monetary household deprivation levels

27.30: Associations across non-monetary deprivations

35.55: Economic growth and non-income deprivations

Part 3: Policy motivations 

55.01: National and international demand

57:35: Political space for new metrics