Abstract
By means of probabilistic graphical models, in this paper, we present a new framework for exploring relationships among indicators commonly included in the Multidimensional Poverty Index (MPI). In particular, we propose an Ising model with covariates for modeling the MPI as an undirected graph. First, we prove why Ising models are consistent with the theoretical distribution of MPI indicators. Then, a comparison between our estimates and the association measures typically used in the literature is provided. Finally, we show how undirected graphs can complement the MPI policy-relevant properties, apart from discovering further insightful patterns that can be useful for policy purposes. This novel approach is illustrated with an empirical application for the global MPI indicators of Guinea and Ecuador, taking living areas and monetary poverty as covariates, respectively.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11205-024-03325-8/MediaObjects/11205_2024_3325_Fig7_HTML.png)
Similar content being viewed by others
Data Availability
The data that support the findings of this study are available in https://www.openicpsr.org/openicpsr/project/192052/version/V1/view.
Notes
This means that if we collapse over C, the conditional association between A and B does not change. Similarly, between B and C, if we collapse over A.
Matlab code available in http://onlinelibrary.wiley.com/doi/10.1111/biom.12202/
The codes to compute the global MPI for both countries were taken from the Stata do-files provided by OPHI corresponding to the global MPI 2021, which are available in https://cloud-ophi.qeh.ox.ac.uk/index.php/s/SgLdxLScG9Tg6gq?path=%2F2021. For more details about technical aspects of the global MPI, see Alkire et al. (2021).
The data for Guinea and Ecuador that support the findings of this study are available on the DHS website (https://dhsprogram.com/methodology/survey/survey-display-539.cfm) and in the official ECV 13-14 website (https://www.ecuadorencifras.gob.ec//documentos/web-inec/ECV/ECV_2015/)
References
Agresti, A. (2012). Categorical data analysis (Vol. 792). Wiley.
Agresti, A. (2015). Foundations of linear and generalized linear models. Wiley.
Alkire, S. (2020). Multidimensional poverty measures as policy tools. In V. Beck, H. Hahn, & R. Lepenies (Eds.), Dimensions of poverty: Measurement, epistemic injustices, activism (pp. 197–214). Springer.
Alkire, S., & Foster, J. (2011). Counting and multidimensional poverty measurement. Journal of Public Economics, 95(7), 476–487.
Alkire, S., Kanagaratnam, U., & Suppa, N. (2021). The global multidimensional poverty index (MPI) 2021 (methodological note No. 51. Oxford Poverty and Human Development Initiative, University of Oxford.
Alkire, S., Roche, J. M., Ballon, P., Foster, J., Santos, M. E., & Seth, S. (2015). Multidimensional poverty measurement and analysis. Oxford University Press.
Alkire, S., & Santos, M. (2014). Measuring acute poverty in develo** world: Robustness and scope of the multidimensional poverty index. World Development, 59, 251–274.
Alkire, S., & Seth, S. (2013). Selecting a targeting method to identify BPL households in India. Social Indicators Research, 112(2), 417–446. https://doi.org/10.1007/s11205-013-0254-6
Alkire, S., UlHaq, R., & Alim, A. (2019). The state of multidimensional child poverty in south Asia: A contextual and gendered view (OPHI Working Papers No. 127). University of Oxford.
Azevedo, V., & Robles, M. (2013). Multidimensional targeting: Identifying beneficiaries of conditional cash transfer programs. Social Indicators Research, 112(2), 447–475.
Ballon, P. (2023). Statistical issues in multidimensional poverty measurement: redundancy analysis. In Research handbook on measuring poverty and deprivation (pp. 463–474). Edward Elgar Publishing.
Berk, R., Brown, L., Buja, A., Zhang, K., & Zhao, L. (2013). Valid post-selection inference. The Annals of Statistics, 41, 802–837.
Bishop, C. (2006). Pattern recognition and machine learning. Springer.
Brown, C., Ravallion, M., & Van de Walle, D. (2018). A poor means test? Econometric targeting in Africa. Journal of Development Economics, 134, 109–124.
Ceriani, L., & Gigliarano, C. (2020). Multidimensional well-being: A Bayesian networks approach. Social Indicators Research, 152, 237–263.
Chan, L., Silverman, B. W., & Vincent, K. (2021). Multiple systems estimation for sparse capture data: Inferential challenges when there are nonoverlap** lists. Journal of the American Statistical Association, 116(535), 1297–1306.
Chen, H., Cohen, P., & Chen, S. (2010). How big is a big odds ratio? Interpreting the magnitudes of odds ratios in epidemiological studies. Communications in Statistics—Simulation and Computation, 39(4), 860–864.
Cheng, J., Levina, E., Wang, P., & Zhu, J. (2014). A sparse Ising model with covariates. Biometrics, 70(4), 943–953.
Cheng, J., Li, T., Levina, E., & Zhu, J. (2017). High-dimensional mixed graphical models. Journal of Computational and Graphical Statistics, 26(2), 367–378.
Cohen, J. (2013). Statistical power analysis for the behavioral sciences. Routledge.
Dai, B. (2013). MVB: Multivariate Bernoulli log-linear model [Computer software manual]. Retrieved from https://CRAN.R-project.org/package=MVB (R package version 1.1)
Dai, B., Ding, S., & Wahba, G. (2013). Multivariate Bernoulli distribution. Bernoulli, 19, 1465–1483.
Duarte, S. (2016). Modelos lineales generalizados: Regresión de rango reducidoy reducción suficiente de dimensiones. Ph.D Dissertation, FIQ-Universidad Nacional del Litoral.
Duarte, S., Forzani, L., García Arancibia, R., Llop, P., & Tomassi, D. (2021). Socioeconomic index for income and poverty prediction: A sufficient dimension reduction approach. Review of Income and Wealth. https://doi.org/10.1111/roiw.12529
Epskamp, S., Maris, G., Waldorp, L. J., & Borsboom, D. (2018). Network psychometrics. In The Wiley handbook of psychometric testing: A multidisciplinary reference on survey, scale and test development (pp. 953–986).
Evans, M., Nogales, R., & Robson, M. (2020). Monetary and multidimensional poverty: Correlations, mismatches and joint distributions (OPHI Working Papers No. 133). Queen Elizabeth House, University of Oxford.
Evans, R. (1996). An analysis of criterion variable reliability in conjoint analysis. Perceptual and Motor Skills, 82(3), 988–990.
FAO, & OPHI. (2022). Measuring rural poverty with a multidimensional approach: The rural multidimensional poverty index (FAO Statistical Development Series No. 19). FAO.
Fiszbein, A., & Schady, N. R. (2009). Conditional cash transfers: Reducing present and future poverty. World Bank Publications.
Gabel, S.G., & Kamerman, S.B. (2013). Conditional cash transfers (CCTs): A child policy strategy in Asia. In Economic stress, human capital, and families in Asia: Research and policy challenges (pp. 197–220). Springer.
Garcia, M., & Moore, C. (2012). The cash dividend: The rise of cash transfer programs in sub-Saharan Africa. World Bank Publications.
Hall, O., Dompae, F., Wahab, I., & Dzanku, F. M. (2023). A review of machine learning and satellite imagery for poverty prediction: Implications for development research and applications. Journal of International Development. https://doi.org/10.1002/jid.3751
Ising, E. (1925). Beitrag zur theorie des ferromagnetismus. Zeitschrift für Physik, 31, 253–258.
Kidd, S., & Wylde, E. (2011). Targeting the poorest: An assessment of the proxy means test methodology (AusAID Report). AusAID.
Klasen, S., & Lange, S. (2015). Targeting performance and poverty effects of proxy means-tested transfers: Trade-offs and challenges (IAI Discussion Papers). Ibero-America Institute for Economic Research.
Koller, D., & Friedman, N. (2009). Probabilistic graphical models: Principles and techniques. MIT Press.
Kowal, D. R. (2022). Bayesian subset selection and variable importance for interpretable prediction and classification. The Journal of Machine Learning Research, 23(1), 4661–4698.
Kruis, J., & Maris, G. (2016). Three representations of the Ising model. Scientific Reports, 6(34175), 1–11. https://doi.org/10.1038/srep34175
Lee, J. D., Sun, D. L., Sun, Y., & Taylor, J. E. (2016). Exact post-selection inference, with application to the lasso.
Meinshausen, N., & Bühlmann, P. (2006). High-dimensional graphs and variable selection with the lasso. The Annals of Statistics, 34(3), 1436–1462.
Nardi, Y., & Rinaldo, A. (2012). The log-linear group-lasso estimator and its asymptotic properties. Bernoulli, 18(3), 945–974. https://doi.org/10.3150/11-BEJ364
Ravikumar, P., Wainwright, M. J., & Lafferty, J. D. (2010). High-dimensional Ising model selection using l-regularized logistic regression. The Annals of Statistics, 38(3), 1287–1319.
Santos, M. E. (2019). Challenges in designing national multidimensional poverty measures (Statistics series No. 100). ECLAC.
Santos, M. E., & Villatoro, P. (2018). A multidimensional poverty index for Latin America. Review of Income and Wealth, 64(1), 52–82.
Sen, A. (1976). Poverty: An ordinal approach to measurement. Econometrica: Journal of the Econometric Society, 44, 219–231.
Sen, A. (1985). Well-being, agency and freedom. Journal of Philosophy, 82, 169–221.
Sen, A. (1999). Development as freedom. Oxford University Press.
Stiglitz, J., Sen, A., & Fitoussi, J.-P. (2009). The measurement of economic performance and social progress revisited: Reflections and overview (Sciences Po publications No. 2009-33). Sciences Po.
Suppa, N., Alkire, S., & Nogales, R. (2022). The many forms of poverty: Analyses of deprivation interlinkages in the develo** world (OPHI Research in Progress Series No. 63a). Queen Elizabeth House, University of Oxford.
Taylor, J., & Tibshirani, R. J. (2015). Statistical learning and selective inference. Proceedings of the National Academy of Sciences, 112(25), 7629–7634.
Wainwright, M. J., & Jordan, M. I. (2008). Graphical models, exponential families, and variational inference. Foundations and Trends in Machine Learning, 1(1–2), 1–305.
Wang, W., Gurgone, A., Martínez, H., Barbieri Góes, M. C., Gallo, E., Kerényi, A., & Andrade, P. D. S. (2022). COVID-19 mortality and economic losses: The role of policies and structural conditions. Journal of Risk and Financial Management, 15(8), 354.
Zhao, S., Witten, D., & Shojaie, A. (2021). In defense of the indefensible: A very Naive approach to high-dimensional inference. Statistical Science, 36(4), 562–577.
Funding
This work was supported by ANPCYT Grant PICT 2018-04377, CONICET Grant PIP 11220200101595CO and UNL Grant CAI+D 506-201901-00041LI.
Author information
Authors and Affiliations
Contributions
Authors contributed equally to this work.
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1: The Global MPI
The global MPI has been published annually by OPHI and UNPD since 2010 and covers more than 100 develo** countries with the aim of complementing globally comparable monetary poverty measures such as the \(\$ 1.90\) day line. Table 5 describes all the selected dimensions, indicators, deprivation cutoffs, and weights.
Poverty cutoff k is equal to 0.33. This means that a person is identified as poor if she is deprived in a third or more of ten (weighted) indicators or, in other words, if she is deprived in one dimension.
Appendix 2: Parameters Estimates of Ising Models
Appendix 3: Redundancy Measures
Figures 8 and 9 show the estimated unconditional association measures (Cramer’s V and Redundancy R) for Guinea and Ecuador, respectively.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
García Arancibia, R., Girela, I. Graphical Representation of Multidimensional Poverty: Insights for Index Construction and Policy Making. Soc Indic Res 172, 595–634 (2024). https://doi.org/10.1007/s11205-024-03325-8
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11205-024-03325-8