1 / 43

Consensual Approach to Measuring Poverty Professor David Gordon, University of Bristol

Consensual Approach to Measuring Poverty Professor David Gordon, University of Bristol Mr Viliami Konifelenisi Fifita, Government Statistician, Kingdom of Tonga Dr Shailen Nandy, Cardiff University Government Statisticians & Planners PSSC Members Meeting Tanoa Skylodge, Nadi, Fiji

carlae
Télécharger la présentation

Consensual Approach to Measuring Poverty Professor David Gordon, University of Bristol

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Consensual Approach to Measuring Poverty Professor David Gordon, University of Bristol Mr Viliami Konifelenisi Fifita, Government Statistician, Kingdom of Tonga Dr Shailen Nandy, Cardiff University Government Statisticians & Planners PSSC Members Meeting Tanoa Skylodge, Nadi, Fiji 15th November 2016

  2. Sustainable Development Goals (SDGs) 2015 to 2030 17 Goals, 169 targets, ??? Indicators

  3. SDGs and Multidimensional Poverty The current proposal for the Sustainable Development Goals (SDGs) includes a target to reduce multidimensional poverty, including child poverty: “1.2 By 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to national definitions.” This idea has been made feasible in part by UNICEF’s pioneering Global Study of Child Poverty and Disparities (2007 to 2011) http://www.unicef.org/socialpolicy/index_45357.html

  4. Poverty Theory

  5. Peter Townsend’s concept of relative deprivation Argues that poverty can only be measured: “objectively and applied consistently only in terms of the concept of relative deprivation.…. The term is understood objectively rather than subjectively. Individuals, families and groups in the population can be said to be in poverty when they lack the resources to obtain the types of diet, participate in the activities and have the living conditions and amenities which are customary, or at least widely encouraged or approved, in the society to which they belong” (1979, p 31)

  6. Universal Needs and Relative Deprivation Measurement of Poverty The key ideas • Poverty is a sociological phenomena which can only be meaningfully measured relative to the society to which a person/household belongs. • There are certain universal needs that people require/ think of as necessities in ALL societies e.g. food, clothing, shelter/housing, health care/medicine, children’s education, leisure activities, social activities/obligations/participation such as present giving and marking major life events such as births, deaths, weddings, etc. • The exact way these universal needs are met varies from society to society but the needs remain universal

  7. Scientific Definitions of Poverty Poverty can be defined as; Command over insufficient resources over time The result of poverty is deprivation

  8. Poverty Groups

  9. Relative Deprivation & Consensual Poverty

  10. Identifying the Essentials of Life and Deprivation (Australian Method)

  11. PSE Survey Questions The interviewer asks the respondent: On these cards are a number of different items which relate to our standard of living. I would like you to indicate the living standards you feel all adults should have in Britain today by placing the cards in the appropriate box. BOX A is for items which you think are necessary – which all adults should be able to afford and which they should not have to do without. BOX B is for items which may be desirable but are not necessary.

  12. Consensual Poverty – if more than 50% agree an item is a necessity of life it is then included in the poverty analysis

  13. South Pacific Consensual Deprivation Module (Viliami Fifita)

  14. Child Consensual Deprivation Questions: Tonga DHS 2012 (Viliami Fifita)

  15. Subjective Poverty Validator Variables (Viliami Fifita)

  16. An Example from the RegionSolomon Islands DHS 2015

  17. Primary Analytical Aim • To identify an optimal adult and child optimal deprivation index; • Each index needs to be; • Suitable • Reliable • Valid • Additive • Suitable – A majority of the population (50% or more) saying that each item is a necessity which everybody should be able to afford • i.e. a ‘customary’ possession or activity in the Island society • Reliability – Classical Test Theory & Latent Trait Models • Validity – Logistic Regression of each deprivation indicator by Subjective Poverty (Standard of Living), Debt and Asset Index • Additivity – ANOVA model, second order interactions of deprivation items by equivalised disposable household income or Asset Index score where income is not available • A common analytical framework was agreed based on an updating of the 1999 Poverty & Social Exclusion Survey deprivation index construction methodology (Pantaziset al, 2006).

  18. ‘The work is considered technically as providing a "gold standard" for the list of MD variables and indicator's construction and has unanimous support’ Eurostat Task Force on Material Deprivation (2011)

  19. Peer Review by the UN Expert Group on Poverty Statistics

  20. Establishing Consensus

  21. Checking Item Validity

  22. Reliability

  23. Deprivation Scale Reliability, Solomon Islands

  24. Item Response Theory Reliability Results

  25. IRT Severity & Discrimination Results: Adult Items Solomon Islands

  26. Additivity

  27. Drawing the Poverty Line

  28. Modal Deprivation by Logarithm of Income as a Percentage of Supplementary Benefit Scale Rates (Townsend, 1979)

  29. Checking Scale Validity

  30. Solomon Islands 2015: Initial Child Poverty Results • Threshold used: deprivation of 5+ Socially Perceived Necessities (SPNs) • Prevalence rates shown are for children (<18 yrs) Headline figures • National prevalence of Consensual Child Poverty: 53% • Urban areas: 37% • Rural areas: 56% • Regions: Malaita 64%, Central 61%, Honiara 36%, Western 35%

  31. Conclusions: Consensual Poverty • The best method currently available to measure multidimensional poverty as; • Based on a clear scientific theory and definition of poverty • 30 year history of continuous methodological development • Proven track record in over 45 countries • Incorporates the views of the public in the measurement of poverty and adapts to social changes. • Method applicable to all countries and societies and it is the only method that can produce meaningful and comparable results in low, middle and high income countries • Easy to implement in diverse survey situations – with generally high response rates and respondent satisfaction • Results have been shown to be valid, reliable and repeatable • Results are not easy to obscure or distort • Socially realistic method that is easily understood & supported by both the public and policy makers • Results are easy to understand and policy relevant

  32. Problems of Food Based Poverty Lines (FPL)

  33. Food Poverty Lines (FPL) Usually Based on Engel’s Law (1857) - as expenditure increases, so the proportion devoted to food will decline. Research has found that there is only a weak association between calorie intake and income in many poor countries (Behrman and Deolalikar, 1987). This is very problematic as “Even the relatively poor individuals weigh heavily food attributes other than calorie content when they make food choice” (p. 666). The variety of food and not its nutritional value is often affected by changes in income (Behrman & Deolalikar,1989; Meenakshi and Vishwanathan, 2003) - “Nutritional requirements are dependent on the work roles of people at different points of history and in different cultures and on foods available in local markets.” (Townsend, 1985)

  34. Orshansky’s Multiplier Method Orshansky multipliers are sometimes used to estimate the non-food poverty line. However, this methodology is based on a number of questionable assumptions (Rio Group, 2006). Firstly, it “assumes that the households that satisfy their nutritional needs are satisfying, at the same time, the minimum standards of the other basic needs”, this assumption is not supported by the empirical evidence (Feres, 1997). Second, Streeten (1989) argues that “there may be an inconsistency in this way of arriving at a poverty line. The minimum food requirements are derived normatively, by calculating how much the minimum requirements would cost; while the non-food items are determined by observing how much people actually spend. In order to remove the inconsistency, we would have to assume that what people actually happen to spend is what they need to spend on non-food items, a clearly unrealistic assumption.”

  35. The US Official Food Poverty Line Even in the 1960s this methodology, which Orshansky (1988) claimed was “relatively absolute” rather than “absolute”, was criticised as being out-dated and unscientific (Fisher, 1992). In 1992, the USA National Academy of Sciences (NAS) Panel on Poverty and Family Assistance concluded that the Orshansky multiplier methodology should be abandoned and a budget standard developed which included food, clothing, shelter (including utilities) and other needs (Citro and Michael, 1995). More recently, a 2004 review by USA Committee on National Statistics concluded “that the current measure needs to be revised: it no longer provides an accurate picture of the differences in the extent of economic poverty among population groups or geographic areas of the country, nor an accurate picture of trends over time.” (Iceland, 2005).

  36. Why Not Use the MPI Method? Example of the problems resulting from the equal weights requirement of the Alkire and Foster MPI Method

  37. OPHI Multidimensional Poverty Index Both the dimensions and the weights are based on judgement and lack empirical support

  38. Testing the OPHI MPI using Confirmatory Factor Analysis CFI & TLI – > 0.9=adequate fit: 0.95 = good fit; RMSEA cut off is <0.06 Model 1 – OPHI MPI - fixed weights and dimensions Model 2 – OPHI MPI – weights can change freely Model 3 – OPHi MPI – weights can change freely, one dimension The fixed weights and dimensions requirement of the OPHI MPI method (Model 1) result in a model to which the data for India do not come close to achieving an adequate fit. However, if the weights and dimensions are allowed to change freely, the India data fit the one dimensional model well (Model 3)

  39. OPHI MPI Indicators but with free weights and one dimension for India

More Related