Understanding the multidimensional structure of poverty in Argentine households

Understanding the multidimensional structure of poverty in Argentine households | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Understanding the multidimensional structure of poverty in Argentine households Adrián Maximiliano Moneta Pizarro This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4736007/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract A question of great interest for the design of multidimensional poverty measures is whether they should include a monetary poverty indicator. One of the most common arguments for keeping income poverty separate from non-monetary poverty is that they reflect different dimensions of the phenomenon. This paper explores the multidimensional structure of poverty in Argentina and investigate whether monetary poverty should be considered as another indicator of multidimensional poverty using generalized structural equation modeling (GSEM) methods. Using categorical data from a household survey, a generalized confirmatory factor analysis model (GCFA) and a GSEM with a second order factor are analyzed. The GCFA model postulates the traditional assumption that monetary poverty is just another dimension of poverty, while the GSEM supports the hypothesis that monetary poverty is a cause of non-monetary poverty. The results show that the data fit well in both cases, but that it is more plausible to consider that the non-monetary factors are indicators of a higher order dimension and that this non-monetary poverty, as a whole, is explained by monetary poverty. Finally, the implications of these results for the design of multidimensional poverty indicators in Argentina are discussed. JEL Classification: C38, I32 multidimensional poverty monetary poverty generalized structural equation modeling Argentina Figures Figure 1 Figure 2 1. Introduction The multidimensional approach, a phenomenon increasingly promoted and accepted by researchers, governments and society, considers poverty as a complex phenomenon related to multiple deprivations present in dimensions where the monetary element could be just one of these deprivations (Salecker, Ahmadov & Karimli, 2020 ). This line of sight is opposed to the unidimensional feature of poverty, which traditionally considers low monetary income as the only conceivable source of deprivation. One of the challenges in assuming a multidimensional position emerges when we ask ourselves which dimensions must be thought over, and this argument is closely related to the specific definition of poverty (Kim, 2016 ). According to Walker ( 2015 ), people are usually aware of poverty at once when they come across it, but many times they face difficulties in asserting exactly what it implies. Although general agreement has been reached when considering poverty as a multidimensional phenomenon, there is little unanimity as to which poverty dimensions should be included (Ntsalaze & Ikhide, 2018 ; Kim, 2016 ) and how they interrelate with each other (Chan y Wong, 2020). In this sense, a question of great interest for the design of multidimensional poverty measures is whether they should include a monetary poverty indicator. One of the most common arguments for keeping income or consumption poverty separate from non-monetary poverty is that they reflect different dimensions of the phenomenon (Santos et al., 2015 ). The standard method of current income captures cyclical fluctuations in welfare related to the labour market. Instead, non-monetary measures of multidimensional poverty capture the deprivations reflected in less volatile and more structural indicators such as precarious housing and unfavourable socio-environmental conditions. Some empirical studies show that monetary measures of poverty are imperfect predictors of non-monetary measures (Bader et al., 2016; Bourguignon et al., 2010; Roelen, 2017; 2018; Roelen et al., 2009; 2012; Ruggeri et al., 2003; Wang et al., 2016). An analysis of the mismatch between income poverty and the multidimensional poverty index carried out in Chile showed that, although 20.4% of the population was multidimensionally poor and 14.4% was income poor, only 5.5% it was poor on both measures (Ministry of Social Development, 2015). These mismatches may be due to the fact that these measures might be capturing different phenomena or differences in how each indicator is captured and calculated. These findings justify the development of new research to understand the causes of these mismatches and their policy implications (UNDP, 2019). The most widespread method to solve the dimension selection problem is to embrace a normative approach, such as the Alkire-Foster axiomatic counting methodology (Alkire & Foster, 2011 ). Methods such as this one assign the dimension definitions and their indicators to subjective decisions based on sociopolitical agreements and data availability. Their advantages are that they can seize deprivation joint distribution, identify the poor and provide a rate to summarize multidimensional poverty measurement in a single indicator (Alkire, Foster, Seth, Santos, Roche & Ballon, 2015 ). The use of statistical techniques is a further way to work out the issue associated with dimension identification. Those who stand up for the usage of these methods are very interested in exploring the complex nature and structure of poverty starting from the data themselves. Alkire et al. ( 2015 ) divide them into two large groups: one related to descriptive techniques and the other based on latent variable modelling. The first group comprises cluster analysis (CA), principal component analysis (PCA) and multiple correspondence analysis (MCA). These methods are quite involved in the problem of dimension reduction. The second group covers factor analysis (FA), latent class analysis (LCA) and structural equation models (SEM). According to Walker ( 2015 ), this second group can be subdivided into two other groups. On the one hand, methods such as FA and LCA are intended to identify poverty dimensions, and on the other hand, SEM is mainly used for contrasting theories about the relationships between dimensions. In SEM, the poverty dimensions and the path relationship among them are specified beforehand by applying the proper theoretical basis or the previous exploratory analysis, and the actual presence is then contrasted against reality according to that captured by the available data. Obviously, the normative approach is not the most appropriate to analyse whether poverty based on income and non-monetary poverty are indicators of the same construct or reflect different dimensions of the phenomenon, as it assumes that dimensions are predetermined. Statistical methods seem more suitable (Nájera Catalán & Gordon, 2020 ). In the case of Argentina, there is still no regular and systematic official measurement of multidimensional poverty due to the lack of consensus on its composition and limitations in the available data sources. Therefore, the question of whether monetary poverty should be part of a multidimensional poverty index is still unresolved and is a very relevant issue. However, the analysis of multidimensional poverty in this country has been focused mainly on the development of composite indicators where poverty is a linear combination of independent factors built by applying PCA, more appropriate for continuous indicators. Thus, the possibility of a reflective measurement model is ignored, which contemplates the dimensions of poverty as latent variables, not directly observable, and the indicators as particular manifestations of these different facets of the phenomenon, possibly correlated with each other. It also omits that the available indicators are mainly binary variables. Works carried out by Conconi and Ham ( 2007 ), Conconi ( 2011 ), Carranza Mena, Pagani and Sánchez Fernández (2011) and Gasparini, Sosa Escudero, Marchionni and Olivieri (2013) give examples of descriptive techniques application, basically PCA, for the identification of dimensions. On the other hand, the works of Fagnola and Moneta Pizarro ( 2021 ) and Moneta Pizarro and Satorres Bechara ( 2021 ) are examples of applications with FA. There are no records in Argentina of the application of more appropriate techniques to explore the relationships between the dimensions of poverty, such as SEM, and even less of methods that take into account the binary nature of the available data, for example, the generalized structural equations models (GSEM). The research background in Argentina represents significant advances in terms of the analysis of multidimensional poverty, but does not establish hypotheses that allow us to investigate the relationships between its dimensions. These antecedents are mostly studies based on Sen’s capabilities approach (1984, 1985, 19982, 2000), but they only contribute a supporting conceptual framework for the understood multidimensionality hypothesis. This means that this framework is not used to assume theoretical models that make sense of causal relationships and serve to confirm multidimensional structure. This paper demonstrates how SEM can be implemented to explore the multidimensional structure of poverty in Argentina and investigate whether monetary poverty should be considered as another indicator of multidimensional poverty. However, given the categorical nature of the available data, the generalized version of these models, GSEM, is used. Specifically, a generalized confirmatory factor analysis (GCFA) model is compared with a full GSEM that includes a second-order factor. The first model proposes to relate each factor with a dimension of poverty without a causal structure that links them. The second model includes structural relationships and proposes two alternative hypotheses at the same time. On the one hand, it assumes the unidimensionality of the non-monetary dimension of poverty, where the non-monetary factors are first-order measures of a single higher-order dimension; on the other hand, it is proposed that non-monetary poverty, represented by the higher order construct, is a consequence of the monetary dimension of poverty. Considering Walker ( 2015 ) as regards this concept, he asserts that poverty is not only the lack of monetary resources necessary to meet specific needs but also refers to the multiple consequences of this scarcity, such as deficiency in education, healthcare, housing and employment (poverty nonmonetary factors), suffered by poor people all at the same time. It also follows Chan and Wong ( 2020 ), who applied SEM to Hong Kong data and found that monetary income has a significant impact on the nonmonetary dimensions of poverty. 2. Literature Review At the international level, there are some precedents in the application of SEM to the analysis of multidimensional poverty that focus more on the development of causal models and less on obtaining synthetic indices. Some examples are Di Tommaso ( 2007 ) with data from India, Ballon and Krishnakumar ( 2008 ) with an application to child poverty in Bolivia, Wagle ( 2009 ) for Nepal and the United States, Kim ( 2016 ) using data from the United Kingdom, Ballon ( 2018 ) for female empowerment in Cambodia, Chan and Wong ( 2020 ) with data from Hong Kong, Zhang and Huai ( 2023 ) in a paper applied to poverty among farmers in China, and Clausen et al. ( 2024 ) to explore the association between multidimensional poverty and depression using data from Perú. However, the models proposed in this background involve the use of dimensions and variables that seem to be exclusively adjusted to the context of those countries or to the availability of data specific to each case. Despite the great global acceptance that progress in multidimensional poverty studies has had (Alkire & Santos, 2010 , 2013 ) and the growth of poverty in Argentina, research works in this area are very few at the national level in this country (Arévalo & Paz, 2015 ). Among the main argentine bibliographic antecedents, outstanding works are those by Conconi and Ham ( 2007 ), Conconi ( 2011 ), Santos, Villatoro, Mancero and Gerstenfeld ( 2015 ), Arévalo and Paz ( 2015 ), Salvia, Bonfiglio and Vera ( 2017 ), Durán and Condorí ( 2017 ), Ignacio-González and Santos ( 2020 ), Fares, Favata and Martínez ( 2021 ), Macció and Mitchell ( 2023 ), Sione ( 2024 ), and Poggiese and Ibañez Martín ( 2024 ). These studies share the particularity that they are mainly centred in the construction and use of synthetic indicators such as those corresponding to Bourguignon and Chakravarty ( 2003 ) and Alkire and Foster ( 2011 ). Some relevant works applying statistical techniques are of great interest for this research. Exemplary antecedents belong to Conconi and Ham ( 2007 ), Conconi ( 2011 ), Carrazán Mena et al. ( 2011 ) and Gasparini et al. ( 2013 ). However, in these cases, factor identification is performed through PCA, and therefore, there is neither progress for the structural model contrast nor attention to the lack of continuity and normality of the variables used as poverty indicators. Recently, some works were included in the literature about argentine multidimensional poverty where robust FA methods are applied with tetrachoric and polychoric correlation matrices for dimension validity. In this way, problems related to the indicator normality absence are overcome. Works by Fagnola and Moneta Pizarro ( 2021 ), Moneta Pizarro and Satorres Bechara ( 2021 ) and Gutiérrez Montecino and Moneta Pizarro ( 2021 ) refined the indicators of multidimensional poverty obtained from the Permanent Household Survey (EPH, for its acronym in Spanish) through strong techniques of exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). The model originally proposed by Fagnola and Moneta Pizarro ( 2021 ) has 15 indicators distributed among 5 factors (exclusion, sanitation, health, infrastructure and economic capacity) and takes into account only one EPH wave (third quarter 2017). After considering EFA and CFA tools, 13 indicators remained and were grouped into 4 factors. The work of Moneta Pizarro and Satorres Bechara ( 2021 ) is an extension of the previous work to longitudinal data of representative periods related to the different phases of the argentine economic cycle, allowing the contrast of the longitudinal invariance of the factor structure. These authors enlarged the proposal to 18 indicators in 5 factors, but after consistency analyses, construct validity and longitudinal invariance, there remained 10 selected indicators in 3 factors (housing, environment and income). Lastly, Gutiérrez Montecino and Moneta Pizarro ( 2021 ) also contrast multiregional invariance, and a measuring model with 7 indicators and 3 factors remains validated. These papers represent significant advances in the identification with robust methods of the factor structure of multidimensional poverty in Argentina, but they do not analyse possible relationships between the latent constructs. At both international and national level, there is no evidence of applications of GSEM, the most advanced version of SEM, in the available literature on multidimensional poverty. GSEM is a novel alternative that is considered more appropriate for dealing with complex data structures (Skrondal & Rabe-Hesketh, 2004 ). 3. Data The elements used are microdata at the household level from the cross-section corresponding to the 2022 first quarter of the EPH from the National Institute of Statistics and Census (INDEC, for its acronym in Spanish). It is a sample of 16898 observations corresponding to households in 32 urban agglomerates in all regions of the country. Using this kind of data represents some disadvantages: they correspond to urban conglomerate samples with more than one hundred thousand inhabitants, and the questionnaire applied was not specifically designed to demonstrate multidimensional poverty characteristics. Survey conclusions are then limited by these restrictions. Another problem is that prospect variables to set up poverty indicators are mostly of dichotomous type (for example, whether the household head is employed or unemployed, whether the family has medical coverage, or if the housing has running water, among others). Therefore, this reveals limitations for the multivariate analysis and particularly for SEM application, working with preferably normal continuous variables and distributions. Nevertheless, this drawback is overcome by using GSEM, a more advanced and appropriate strategy of statistical modelling for these kinds of data. With respect to the variables, the measurement model validated by Gutiérrez Montecino and Moneta Pizarro ( 2021 ) is used as the starting point, considering that it has 7 indicators grouped into 3 factors. The first factor is related to the conditions of the housing infrastructure, and the indicators are: Roof: It takes the value 1 if the roof is of low quality (plastic, cardboard, cane, planks or sheets without ceiling or inner lining) and 0 otherwise. Bathroom: it takes the value 1 if it does not have a bathroom with drainage inside the housing and 0 otherwise. The second factor is related to the environment or surroundings of the housing, and the indicators are as follows: Dumpsite: it takes the value 1 if the housing is less than 3 blocks from a dumpsite and 0 otherwise. Floodable area: it takes the value 1 if the housing is in a floodable area (in the last 12 months) and 0 otherwise. Finally, the third factor includes indicators related to economic home resources: External support takes the value 1 if the home receives external monetary or material support in the form of subsidies, assistance programs, charity and others and 0 otherwise. Medical coverage: It takes the value 1 if any of the members of the housing unit does not pay or is deducted payment for medical coverage services and 0 otherwise. TFI < TBB: it takes the value 1 in the case that the total family income is less than the total specific basic basket of that home and 0 otherwise. 1 The first two factors are related to nonmonetary dimensions of poverty, while the third factor is strictly associated with a monetary dimension. Descriptive statistics for all these variables are presented in the following table, where the means, since all variables are binary, represent the proportion of observations with values equal to 1. Table 1 Descriptive statistics Variable Obs. Mean Std. dev. Roof 16898 0.0728 0.2598 Bathroom 16898 0.0867 0.2814 Dumpsite 16898 0.0518 0.2216 Floodable 16898 0.0541 0.2262 External support (Ext_supp) 16898 0.2310 0.4215 Medical coverage (Med_cov) 16898 0.3628 0.4808 TFI < TBB 14854 0.3161 0.4650 4. Methodology As stated above, GSEM techniques are applied in this work. This modelling combines SEM capacities with generalized linear model (GLM) capacities. Similar to the econometric methods of simultaneous equations, SEM allows the simultaneous examination of a group of dependency relationships where some variables act as predictor and dependent variables at the same time but with the capacity to estimate and evaluate the relationship among latent (unobservable) variables. These variables are constructs assumed from the theory that they can be measured through observable variables (Cupani, 2012 ). Compared with other analysis techniques where the constructs are represented by a single measurement and the measurement error is not modelled, multiple measures are used in SEM to represent each construct and control the specific measurement error of each variable, thus allowing the assessment of the validity of each construct (Ruiz, Pardo & San Martín, 2010 ). As Kline ( 2015 ) shows, every SEM has two elements: (a) a measurement model representing the relationships between the latent variables and their manifest indicators and (b) a structural model describing the interrelationship among the latent constructs. By using the measurement model, the aim is to verify the adequacy of the indicators selected in the measurement of relevant constructs. The structural relationship model is the one truly desired to be estimated. It includes the effects and relationships among the constructs, which are normally latent variables. It is similar to a regression model, but it may have concatenated effects and loops among variables. Another particular characteristic of SEM is that it has several statistical tests and a group of goodness of fit indicators, but the adjustment is verified when the values of the estimated parameters reproduce the observed covariance matrix as strictly as possible (Kahn, 2006 ). For this purpose, in SEM, the model estimation is based on the existing correlations among measured variables in a cross-sectional sample. Instead of reducing the difference between the predicted values and the ones observed at the individual level (least squares method), the difference between the covariances observed in the sample and the covariances predicted by the structure model is reduced. According to Long ( 1983 ), this is the reason why these models are also called covariance structure models. Therefore, the model residuals are the differences between the covariances observed and the covariances predicted by the theoretical structure model (Ruiz et al., 2010 ). By combining SEM with GLM, in GSEM, it is possible to work with response variables that may be continuous, binary, ordinal, count or multinominal variables. In addition, both normal linear regressions and the large spectrum of regressions of the exponential family (gamma, logit, probit, Poisson, negative binomial and their variants) can be modelled (Skrondal & Rabe-Hesketh, 2004 ). In this research, the presence of binary variables makes it necessary to apply GSEM instead of SEM. In this work, first, a measurement model (without a structural part) with a logit link is specified, that is, a CFA model adapted for variables of dichotomous or binary response (GCFA). In this model, as indicated in Fig. 1 below, it is assumed that each latent factor is one poverty dimension that can be measured through observable indicators and that the factors may be correlated. This first model is therefore made up of three measurement submodels, one for each factor. Two observed indicators, roof and bathroom, help us to approach the housing conditions; two other observed indicators, dumpsite and floodable area, are assumed to be manifestations of the environment conditions; and three indicators observed measure the economic home resources. The three measurement models are jointly estimated, and this permits correlations among factors. As McGartland Rubio et al. ( 2001 ) indicate, the verification of these correlations is usually understood as the result of the existence of a higher-order factor. Nevertheless, this may not necessarily be the reason, since the correlation can be because the factors measure different dimensions of one construct, poverty in this case. Using matrix notation, this first model can be represented through the measurement Eq. ( 1 ): $$\:\text{logit}\left[\text{Pr}\left(\left.w=1\right|\xi\:\right)\right]=\:{{\Lambda\:}}_{w}\xi\:$$ 1 where \(\:w\) is the vector of observed indicators, \(\:1\) is a vector of ones, \(\:\xi\:\) is the vector of latent factors (poverty dimensions) with covariance matrix Φ and \(\:{{\Lambda\:}}_{w}\) is the matrix of model coefficients (factor loadings). Second, a full GSEM is specified where two factors, housing and environment, are demonstrations of nonmonetary poverty, that is, a second-order construct, and the factor related to economic resources or monetary poverty is an exogenous latent variable explaining nonmonetary poverty. This model is represented in Fig. 2 . It should be noted that the measurement indicators for each factor are the same as those in the first model. The full GSEM can be expressed in matrix form using equations ( 2 ), ( 3 ) and (4). $$\:\eta\:=\:\text{B\:}\eta\:+{\Gamma\:}\:\tau\:+\epsilon\:$$ 2 $$\:\text{logit}\left[\text{Pr}\left(\left.y=1\right|\eta\:\right)\right]=\:{{\Lambda\:}}_{y}\eta\:$$ 3 $$\:\text{logit}\left[\text{Pr}\left(\left.x=1\right|\tau\:\right)\right]=\:{{\Lambda\:}}_{x}\tau\:$$ 4 Equation ( 2 ) represents the structural part of the model, while equations ( 3 ) and ( 4 ) are the measurement model for latent factors. In the structural part, \(\:\eta\:\) is the vector of endogenous latent variables ( \(\:{\eta\:}_{1},\:{\eta\:}_{2},\:{\eta\:}_{3}\) ), with \(\:{\eta\:}_{1}\) and \(\:{\eta\:}_{2}\) being the first-order factors (housing and environment) and \(\:{\eta\:}_{3}\) being the second-order factor (nonmonetary poverty), \(\:\tau\:\) is the only exogenous latent variable (economic resources), \(\:\text{B}\) is the 3×3 coefficient matrix of the endogenous latent variables, \(\:{\Gamma\:}\) is a coefficient vector of order 3 whose elements are all equal to zero except the third one that represents the effect of \(\:\tau\:\) on \(\:{\eta\:}_{3}\) and \(\:\epsilon\:\) is the vector of disturbances ( \(\:{\epsilon\:}_{1},\:{\epsilon\:}_{2},\:{\epsilon\:}_{3}\) ) associated with the endogenous latent variables included in \(\:\eta\:\) with a diagonal matrix of covariances \(\:{\Psi\:}\) . In the measurement model, \(\:y\) is the vector of indicators used to measure \(\:{\eta\:}_{1}\) and \(\:{\eta\:}_{2}\) , \(\:x\) is the vector of indicators for the measurement of \(\:\tau\:\) , \(\:{{\Lambda\:}}_{y}\) and \(\:{{\Lambda\:}}_{x}\) are factor loading matrices and \(\:1\) is a vector of ones. With the help of Stata 17, both models are estimated with maximum likelihood (ML), which is the only available method for GSEM. In the case of the second model, its high complexity hindered difficulties in achieving convergence in the estimation process. This problem was temporally solved by modifying the numerical integration model and reducing the number of integration points. The results were satisfactory by using the model of the non-adaptive Gauss-Hermite quadrature and by reducing the points of numerical integration to 3. Afterwards, these estimations were used as improved initial values to re-estimate the model with the default options, which allowed arriving at a solution with greater accuracy. This was achieved with the adaptive Gauss-Hermite quadrature algorithm based on the mean and the variance and 7 points of numerical integration. 5. Results Table 2 indicates the estimation results of the pure measurement model, that is, the GCFA model, which permits us to contrast the convergent and discriminant validity of the three factors proposed for multidimensional poverty. By observing such results, all the indicators are significantly related to their corresponding latent constructs. This supports the convergent validity of the model, that is, that the indicator variables of each factor are strongly correlated with each other, sharing a high proportion of the variance (Aldás & Uriel, 2017 ). The covariances among the latent factors are also significantly different from zero. Departing from the estimated variances and covariances, the correlations obtained are equal to 0.51 between housing and environment, 0.60 between housing and income and 0.32 between environment and income. These moderate correlation values indicate discriminant validity. Table 2 Results of the GCFA model Variables Coefficient Std. Err. z P > | z | [95% Conf. Int.] Roof Housing constant 1 (restricted) -4.232543 0.1449429 -29.2 0.000000 -4.516626 -3.94846 Bathroom Housing constant 0.9593894 0.0813283 11.8 0.000000 0.7999889 1.11879 -3.85143 0.1272349 -30.27 0.000000 -4.100806 -3.602054 Dumpsite Environment constant 1 (restricted) -4.199538 0.167694 -25.04 0.000000 -4.528213 -3.870864 Floodable area Environment constant 1.288781 0.1995092 6.46 0.000000 0.8977506 1.679812 -4.802548 0.3051847 -15.74 0.000000 -5.400699 -4.204397 External support Economic resources constant 1 (restricted) -1.809089 0.0375684 -48.15 0.000000 -1.882722 -1.735457 Medical coverage Economic resources constant 1.474708 0.0745719 19.78 0.000000 1.32855 1.620866 -1.093027 0.0417698 -26.17 0.000000 -1.174894 -1.011159 TFI < TBB Economic resources constant 1.263382 0.0554054 22.8 0.000000 1.15479 1.371975 -1.422182 0.0424695 -33.49 0.000000 -1.505421 -1.338944 var(Housing) 5.305203 0.5600140 4.313697 6.524607 var(Environment) 3.453833 0.5148337 2.578813 4.625756 var(Economic resources) 3.105572 0.1776731 2.776153 3.474079 cov(Housing/Environment) 2.188527 0.2328391 9.4 0.000000 1.73217 2.644883 cov(Housing, Econ. res.) 2.471096 0.1573252 15.71 0.000000 2.162744 2.779448 cov(Environment, Econ. res.) 1.075617 0.1115909 9.64 0.000000 0.8569024 1.294331 The fact that the covariances among factors are significant demonstrates that the presence of a second-order factor and structural relationships among the latent variables are reasonable. In this way, the specification corresponding to the full GSEM is plausible. Table 3 exhibits the results from this estimation. According to the measurement modelling, all the indicators of this second model are also significantly related to their corresponding latent factors, and as expected, the estimated coefficients assume very similar values to those of the pure measurement model. By analysing Table 3, estimated covariances among constructs are not included since they have been replaced by structural relationships. All these structural coefficients were also significant. Therefore, the hypotheses presented for this second model are supported by the empirical evidence. Table 3. Results of full GSEM Variables Coefficient Std. Err. z P>|z| [95% Conf. Int.] Measurement model Roof 1 (restricted) Housing constant -4.246725 0.1571272 -27.03 0.000000 -4.554688 -3.938761 Bathroom Housing 0.955539 0.0866052 11.03 0.000000 0.7857958 1.125282 constant -3.854903 0.1302366 -29.6 0.000000 -4.110162 -3.599644 Dumpsite Environment 1 (restricted) constant -4.19829 0.1645436 -25.51 0.000000 -4.52079 -3.875791 Floodable Area Environment 1.289765 0.1964518 6.57 0.000000 0.9047261 1.674803 constant -4.803195 0.3025952 -15.87 0.000000 -5.396271 -4.21012 External support Economic resources 1 (restricted) Constant -1.809797 0.0395883 -45.72 0.000000 -1.887389 -1.732206 Medical coverage Economic resources 1.474086 0.0779903 18.9 0.000000 1.321228 1.626944 Constant -1.093534 0.0417848 -26.17 0.000000 -1.17543 -1.011637 TFI < TBB Economic resources 1.263164 0.0579683 21.79 0.000000 1.149548 1.376779 constant -1.422828 0.0424854 -33.49 0.000000 -1.506098 -1.339558 Structural model Housing Non-monetary poverty 1 (restricted) Environment Non-monetary poverty 0.4383168 0.0493818 8.88 0.000000 0.3415303 0.5351033 Non-monetary poverty Economic resources 0.7941819 0.057259 13.87 0.000000 0.6819563 0.9064074 var(e.Non-monetary poverty) 3.001571 0.4559609 2.228665 4.042523 var(e.Housing) 0.383752 0.3345924 0.0694848 2.119391 var(e.Environment) 2.492091 0.3747607 1.855924 3.346321 var(e.Economic resources) 3.104728 0.1953692 2.744483 3.51226 Comparing the models with the Akaike (AIC) and Schwarz (BIC) information criteria, the full GSEM has lower values in both cases. The GCFA model has an AIC equal to 82658.00 and a BIC equal to 82789.49, while the GSEM model achieves an AIC equal to 82657.17 and a BIC equal to 82788.66. Consequently, GSEM is the relatively best fit to the data. 6. Discussion The results show that with the EPH data, in spite of their limitations for multidimensional poverty analysis, it is possible to confirm the presence of at least three dimensions of poverty. One of them is associated with the material conditions of the housing, another one is related to the environmental surrounding and the last one corresponds to the monetary income. However, it is more realistic to consider that the nonmonetary factors of poverty, such as housing and environmental conditions, are different indicators of the same dimension of higher order and that this nonmonetary poverty, as a whole, is explained by the monetary dimension. Empirical support is then provided for the identification of two dimensions instead of three. One dimension is associated with the lack of monetary resources, and the other is related to nonmonetary deprivations. In this way, it cannot be excluded that poverty is multidimensional in Argentina, but a structural relationship is supported among its dimensions. This means, as Walker ( 2015 ) points out, that poverty is manifested not only by the presence of inadequate monetary income but also by the multiple consequences of this absence regarding housing and environment, which are part of a nonmonetary dimension of the phenomenon. This conclusion, seeming quite obvious, is opposed to a great part of the literature about multidimensional poverty developed up to present time, which sometimes implicitly and other times explicitly associates the correlation among poverty factors to the presence of different dimensions or attributes of one same construct. Kim ( 2016 ), who in his measurement model includes economic resources as one more poverty dimension and is based on Kangas and Ritakallio ( 1998 ), Lelli ( 2001 ), Whelan ( 1993a ) and Whelan ( 1993b ), affirms that this it is common because these resources can be applied to other functions related to poverty, for example, to healthy food purchases. The criticism that can be made of this type of position is that they do not inquire in depth about these links and the possible presence of structural relationships between the dimensions. The exception to this may be the trap models of poverty. Beneath the theoretical framework of these models, poverty is characterized as a vicious circle due to the presence of self-reinforcing mechanisms that provoke its continuation (Santos, 2014 ) and therefore postulate in this way the possibility of causal relationships among their dimensions. 7. Implications The implication of these conclusions for future research and public policy designs is that, on the basis of the data used for this research and the evidence found, poverty in Argentina should not be measured by using one multidimensional index but by using two. These should be a monetary poverty indicator on the one hand and a nonmonetary poverty indicator on the other. In other words, income poverty should not be mixed with other dimensions of multidimensional poverty since it is very likely the reason for the nonmonetary dimensions. Thus, and differently from what was proposed by Santos et al. ( 2015 ), the recommendation to keep the income poverty and nonmonetary poverty indicators separate is supported. The arguments that are generally put forward to recommend this are very well compiled in the manual of the United Nations Development Programme and the Oxford Poverty and Human Development Initiative of the University of Oxford (UNDP & OPHI, 2019), where it is stated that if the objective to elaborate a multidimensional poverty measure is to complement the current statistics of monetary poverty, then the inclusion of a dimension related to income adds noise. In these cases, it is appropriate to look beyond the manifestations of economic capacity, broadening the understanding of the phenomenon towards nonmonetary dimensions not captured by the traditional measures of monetary poverty. Among other arguments, it also points out the importance of recognizing that monetary and nonmonetary indicators capture poverty differently. Monetary indicators are generally considered indirect measures of poverty because they focus on the scarcity of resources for the acquisition of basic goods and services, while multidimensional indexes developed on the basis of nonmonetary indicators are considered direct measures of lack of well-being because they reflect real deprivations. However, this study provides another reason for the monetary and nonmonetary poverty measures to be kept separate, and it is the finding of empirical evidence, at least for the case of Argentina, in favor of the fact that monetary poverty is a good predictor of nonmonetary poverty. This result could be associated with a close relationship between the indicators of one type of poverty and the other, suggesting a certain coincidence between both measures and that they could be combined without problems in a single measurement. But the correlation is not perfect, as the results show for the structural regression of the nonmonetary dimension as a function of the monetary one. The fact that economic resources partially explain nonmonetary poverty does not mean that they are useful for measuring the same concept. In contrast, both the measurement model proposed through the GCFA and the complete GSEM assume that the household economic capacity indicators are manifestations of one construct, while the rest, the nonmonetary ones, are manifestations of others. The empirical evidence found here confirms that these measures should not be combined, but not because they are independent or do not coincide in the estimation of the poor population but because it is more plausible, as stated in the complete GSEM, that one is the cause of the other. These results can be a constructive contribution to the debate on the relationship between monetary and multidimensional poverty measures, which for the Latin American case is very appropriately discussed in Santos et al. ( 2015 ). 8. Conclusion The aim of this paper was to deepen the multidimensional analysis of poverty in Argentina. By exploring the multidimensional nature of poverty, the study aimed to highlight the importance of considering various dimensions of deprivation beyond monetary income. Furthermore, it set out to address the complexities and challenges associated with determining the specific dimensions that should be incorporated into multidimensional poverty measures, thereby contributing to a more comprehensive understanding and improving the effectiveness of poverty alleviation efforts in Argentina. Key aspects of the methodology employed included the identification and selection of relevant dimensions, the comparison of rival models to analyse the interrelationships between these dimensions, and the use of advanced statistical techniques such as GSEM to assess multidimensional poverty in Argentina. Ideas from previous research on robust measurement methods were incorporated and empirical data from the EPH were used to validate the proposed models. The results of the study highlighted the effectiveness of the multidimensional approach in capturing the complexities of poverty beyond mere monetary measures. The testing of the models developed provided valuable insights for a more complete understanding and identified the interrelationships between the different facets of the phenomenon in Argentina. The empirical validation of these models corroborated the robustness of the multidimensional approach in capturing the nuances of poverty and guiding interventions for the development of more appropriate indicators. By exploring various dimensions of deprivation, the research provided evidence on the interconnected nature of the different aspects of poverty and underlined the need to separate the monetary dimension from the non-monetary ones. In this way, the study provides a framework for designing appropriate indicators of multidimensional poverty, as well as for the implementation of more effective and targeted comprehensive policies to combat this scourge. In summary, this research on multidimensional poverty not only contributes significantly to the academic literature, but also has important implications for policy and practice. By delving into the complexities of poverty measurement beyond income and exploring the multidimensional nature of deprivation, the study offers a more comprehensive understanding of poverty. The conclusions highlight the need to adopt a multidimensional approach, complementary to the traditional monetary one, when developing poverty measures and designing programs for its alleviation. Declarations Funding This research was supported by Secretariat of Science and Technology of the National University of Córdoba under grant code 33620230100448CB. Competing interests The author has no financial or non-financial interests to disclose. Ethics approval This is an observational study using open data from the Encuesta Permanente de Hogares (EPH) published by the Instituto Nacional de Estadísticas y Censos (INDeC) of Argentina. No ethical approval is required. Author Contribution The entire manuscript has been produced and written by a single author. References Aldás, J., & Uriel, E. (2017). Análisis multivariante aplicado con R (2nd Ed.). Ed. Paraninfo. Alkire, S., & Foster, J. (2011). Counting and Multidimensional Poverty Measurement. Journal of Public Economics , 95 (7-8), 476–487. https://doi.org/10.1016/j.jpubeco.2010.11.006 Alkire, S., Foster, J. E., Seth, S., Santos, M. E., Roche, J. M., & Ballon, P. (2015). Multidimensional Poverty Measurement and Analysis: Chapter 3 – Overview of Methods for Multidimensional Poverty Assessment . OPHI Working Paper, 84. https://ophi.org.uk/multidimensional-poverty-measurement-and-analysis-chapter-3-overview-of-methods-for-multidimensional-poverty-assessment/ Alkire, S., & Santos, M. E. (2010). Acute Multidimensional Poverty: A New Index for Developing Countries . OPHI Working Paper, 38. https://ophi.org.uk/acute-multidimensional-poverty-a-new-index-for-developing-countries/ Alkire, S. & Santos, M. E. (2013). A Multidimensional Approach: Poverty Measurement & Beyond. Social Indicators Research , 112 , 239–257. https://doi.org/10.1007/s11205-013-0257-3 Arévalo, C., & Paz, J. A. (2015). P obreza en la Arg en tina. Priv aciones m u ´ l tiples y asimetrías regionales . Documento de Trabajo No. 15, Instituto de Estudios Laborales y del Desarrollo Econo´mico (IELDE), Facultad de Ciencias Econo´micas, Jurídicas y Sociales, Universidad Nacional de Salta (UNSa). https://www.economicas.unsa.edu.ar/ielde/index.php/documentos-de-trabajo/167-nro-15-primavera-2015-carla-arevalo-y-jorge-paz-pobreza-en-la-argentina-privaciones-multiples-y-asimetrias-regionales Ballon, P., & Krishnakumar, J. (2008, august). A Model-Based Multidimensional Capability Deprivation Index (Paper). 30th General Conference of The International Association for Research in Income and Wealth, Portoroz, Slovenia. http://old.iariw.org/papers/2008/ballon.pdf Ballon, P. (2018). A Structural Model of Female Empowerment. The Journal of Development Studies , 54 (8), 1303-1320. https://doi.org/10.1080/00220388.2017.1414189 Bourguignon, F., & Chakravarty, S. (2003). The Measurement of Multidimensional Poverty . Journal of Economic Inequality , 1 , 25–49. https://doi.org/10.1023/A:1023913831342 Carrazán Mena, G., Pagani, P. A., & Sánchez Fernández, D. (2011). Análisis multivariado de la pobreza en la Ciudad de Salta. Una aproximación a partir del Censo Social 2009-2010 . Documento de Trabajo No. 3, Instituto de Investigaciones Económicas (IEE), Universidad Nacional de Salta. https://www.economicas.unsa.edu.ar/iie/archivos/dt/2011S%C3%A1nchez.pdf Chan, S. M., & Wong, H. (2020). Impact of Income, Deprivation and Social Exclusion on Subjective Poverty: A Structural Equation Model of Multidimensional Poverty in Hong Kong. Social Indicators Research , 152, 971–99. https://doi.org/10.1007/s11205-020-02476-8 Clausen, J., Barrantes, N., Caballero, E., & Guillén, H. (2024). Exploring the Association between Multidimensional Poverty and Depression Using Structural Equation Models. Applied Research Quality Life , 19, 727–747. https://doi.org/10.1007/s11482-023-10262-0 Conconi, A., & Ham, A. (2007). Pobreza Multidimensional Relativa: Una Aplicaci o ´ n a la Arg en tina . Documento de Trabajo No. 57, Centro de Estudios Distributivos, Laborales y Sociales (CEDLAS), Universidad Nacional de La Plata. http://sedici.unlp.edu.ar/handle/10915/3616 Conconi, A. (2011). Pobreza Multidimensional en Argentina: Ampliando las Medidas Tradicionales de Pobreza por Ingreso y NBI . Documento de Trabajo No. 90, Departamento de Economía, Facultad de Ciencias Económicas, Universidad Nacional de La Plata. https://www.depeco.econo.unlp.edu.ar/wp/wp-content/uploads/2017/05/doc90.pdf Cupani, M. (2012). Análisis de Ecuaciones Estructurales: conceptos, etapas de desarrollo y un ejemplo de aplicación. Revista Tesis , 1 , 186–199. https://rdu.unc.edu.ar/bitstream/handle/11086/22039/16.pdf?sequence=1&isAllowed=y Di Tommaso, M. (2007). Children Capabilities: A Structural Equation Model for India. Journal of Socio-Economics , 36 (3), 436–450. https://doi.org/10.1016/j.socec.2006.12.006 Durán, R. J., & Condorí, M. A. (2017). Deprivation Index for Small Areas Based on Census Data in Argentina. Social Indicators Research , 141 , 331–363. https://doi.org/10.1007/s11205-017-1827-6 Fagnola, B., & Moneta Pizarro, A. M. (2021). Identificación de la pobreza multidimensional en Argentina con métodos robustos de análisis factorial. Revista Cuadernos del CIMBAGE , 23 . https://ojs.econ.uba.ar/index.php/CIMBAGE/article/view/2053 Fares, F. M., Favata, F., & Martínez, R. G. (2021). Una propuesta para la medición de la pobreza multidimensional en la Argentina (2004-2019). Economía y Desafíos del Desarrollo , 1 (7), 4-48. https://revistaedd.unsam.edu.ar/?p=1454 Gasparini, L., Sosa Escudero, W., Marchionni, M., & Olivieri, S. (2013). Multidimensional poverty in Latin America and the Caribbean: new evidence from the Gallup World Poll. Journal of Economic Inequality , 11 (2), 195–214. https://doi.org/10.1007/s10888-011-9206-z Gutiérrez Montecino, D. A., & Moneta Pizarro, A. M. (2021, october). Invarianza mul tirregional de la pobreza multidimensional en Argentina (Paper). XIV Congreso Latinoamericano de Sociedades de Estadística “Laura Nalbarte”, Montevideo, Uruguay. https://sue.org.uy/wp-content/uploads/2022/05/Libro_de_resumenes_XIV_CLATSE.pdf Ignacio-González, F. A., & Santos, M. E. (2020). Pobreza multidimensional urbana en Argentina. ¿Reducción de las disparidades entre el norte grande argentino y centro-cuyo-sur? (2003-2016). Cuadernos de Economía , 39 (81), 795-822. https://doi.org/10.15446/cuad.econ.v39n81.76486 Kahn, J. H. (2006). Factor analysis in Counseling Psychology research, training and practice: Principles, advances and applications. The Counseling Psychologist , 34 (5), 1–36. https://doi.org/10.1177/0011000006286347 Kangas, O., & Ritakallio, V.-M. (1998). Different methods—different results? approaches to multidimensional poverty. In H.-J. Andress (Ed.), Empirical poverty research in a comparative perspective (pp. 167–203). Ashgate. Kim, S-G. (2016). What Have We Called as “Poverty”? A Multidimensional and Longitudinal Perspective. Social Indicators Research , 129 , 229–276. https://doi.org/10.1007/s11205-015-1101-8 Kline, R. B. (2015). Principles and Practice of Structural Equation Modeling (4th Ed.). Guilford Press. Lelli, S. (2001). Factor analysis vs. Fuzzy sets theory: Assessing the influence of different techniques on Sen’s functioning approach . Discussion Paper Series (DPS) 01.21, Center for Economic Studies, K. U. Leuven. http://www.econ.kuleuven.be/ces/discussionpapers/default.htm Long, J. S. (1983). Covariance Structure Models: An introduction to LISREL. Sage University Paper Series on Quantitative Applications in the Social Sciences , 34 . Macció, J. & Mitchell, A. (2023). Medición multidimensional de pobreza en ciudades segregadas: evidencia de la ciudad de Buenos Aires. Revista Desarrollo y Sociedad , 1 (93), 101-137. https://doi.org/10.13043/DYS.93.3 McGartland Rubio, D., Berg-Weger, M., & Tebb, S. S. (2001). Using Structural Equation Modeling to Test for Multidimensionality. Structural Equation Modeling , 8 (4), 613–626. https://doi.org/10.1207/S15328007SEM0804_06 Moneta Pizarro, A. M., & Satorres Bechara, A. P. (2021). Estabilidad dina´mica de la pobreza multidimensional en Argentina. In R. Pérez Calle, E. Trincado Aznar & E. Gallego Abaroa (Eds.), Economía, empresa y justicia. Nuevos retos para el futuro (pp. 1337–1365). Editorial Dykinson S.L. https://www.dykinson.com/libros/economia-empresa-y-justicia-nuevos-retos-para-el-futuro/9788413773261/ Nájera Catalán, H. E., & Gordon, D. (2020). The Importance of Reliability and Construct Validity in Multidimensional Poverty Measurement: An Illustration Using the Multidimensional Poverty Index for Latin America (MPI-LA) . The Journal of Development Studies , 56 (9). https://doi.org/10.1080/00220388.2019.1663176 Ntsalaze, L., & Ikhide, S. (2018). Rethinking Dimensions: The South African Multidimensional Poverty Index. Social Indicators Research , 135 , 195–213. https://doi.org/10.1007/s11205-016-1473-4 Poggiese, M., & Ibañez Martín, M. M. (2024). Privaciones multidimensionales: El origen de la pobreza y la exclusión social. Un estudio para Argentina. Visión de futuro , 28 (1), 36-60. https://dx.doi.org/https://doi.org/10.36995/j.visiondefuturo.2023.28.01.002.es Ruiz, M. A., Pardo, A., & San Martín, R. (2010). Modelos de Ecuaciones Estructurales. Papeles del Psicólogo , 31 (1), 34-45. http://www.redalyc.org/articulo.oa?id=77812441004 Salecker, L., Ahmadov, A. K. & Karimli, L. (2020). Contrasting Monetary and Multidimensional Poverty Measures in a Low‑Income Sub‑Saharan African Country. Social Indicators Research , 151 , 547–574. https://doi.org/10.1007/s11205-020-02382-z Salvia, A., Bonfiglio, J. I., & Vera, J. (2017). La pobreza multidimensional en la argentina urbana 2010-2016. Un ejercicio de aplicación de los métodos OPHI y CONEVAL al caso argentino (Online). Observatorio de la Deuda Social Argentina, Barómetro de la Deuda Social Argentina, Serie del Bicentenario (2010-2016). https://repositorio.uca.edu.ar/handle/123456789/8206 Santos, M. E. (2014). El índice multidimensional y trampas de pobreza en el Cono Sur. Revista Problemas del Desarrollo , 178 (45), 89–112. https://doi.org/10.1016/S0301-7036(14)70877-6 Santos, M. E., Villatoro, P., Mancero, X., & Gerstenfeld, P. (2015). A multidimensional poverty index for Latin America . OPHI Working Paper, 79. https://www.ophi.org.uk/wp-content/uploads/OPHIWP079.pdf Sen, A. (1984). Rights and capabilities . Basil Blackwell. —— (1985). Commodities and Capabilities . Elsevier. —— (1992). Inequality reexamined . Oxford University Press. —— (2000). Chapter 1 Social justice and the distribution of income. In A. B. Atkinson & F. Bourguignon (Eds.), Handbook of Income Distribution (pp. 59-85). Elsevier. Sione, C. A. (2024). La medición multidimensional de la pobreza en Argentina: propuesta metodológica. Ciencia, Docencia Y Tecnología , 35 (70), 1-43. https://doi.org/10.33255/3570/1642 Skrondal, A., & Rabe-Hesketh, S. (2004). Generalized Latent Variable Modeling. Multilevel, Longitudinal, and Structural Equation Models . Chapman & Hall/CRC. United Nations Development Programme & Oxford Poverty and Human Development Initiative. (2019). How to Build a National Multidimensional Poverty Index (MPI): Using the MPI to inform the SDGs . https://www.undp.org/publications/how-build-national-multidimensional-poverty-index Wagle, U. (2009). Multidimensional Poverty Measurement: Concepts and Applications. Springer. Walker, R (2015). Multidimensional Poverty . GSDRC Professional Development Reading Pack No. 22, University of Birmingham. https://gsdrc.org/wp-content/uploads/2015/10/Multidimensional-Poverty_RP.pdf Whelan, B. J. (1993a). Non-monetary Indicators of Poverty. In J. Berghman & B. Cantillon (Eds.), The European face of social security: Essays in honour of Herman Deleeck (pp. 24-42). Avebury. Whelan, C. T. (1993b). The role of social support in mediating the psychological consequences of economic stress. Sociology of Health and Illness , 15 (1), 86–101. https://doi.org/10.1111/1467-9566.ep11343797 Zhang, Y., & Huai, J. (2023). A Case Study of Farmers’ Behavioral Motivation Mechanisms to Crack the Fractal Multidimensional Relative Poverty Trap in Shaanxi, China. Agriculture , 13 (11), 20-43. https://doi.org/10.3390/agriculture13112043 Footnotes This is the classic indicator of monetary poverty (unidimensional) in Argentina. The total basic basket (TBB) is the value of a set of goods and services for basic consumption. Households whose total family income (TFI) is not enough to cover the TBB are considered poor. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4736007","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":336362886,"identity":"9e147459-3c84-4297-ac75-ffb1c796ddab","order_by":0,"name":"Adrián Maximiliano Moneta Pizarro","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYFCCBIYDcPYHCGVArBZmBsYZxGphgGlh5iFGC3977sMDP3dskzNn7z/42bbtjjwDe/M2CYaaezi1SJx5bnCw98xtY8uew8zSuW3PDBt4jpVJMBwrxm3NjTSGA7xttxM33EhmAGo5nMAgkWMmwcCWgFOHPFDLwb8gLfcfM/+2BGmRfwPU8g+3FgOglsMQW5jZpBnBtvCYSTC24dZieOYZw2HZttvGBmeSzSx7zh02bONJK7ZI7MOtRe54GvPHt2235QyOH3x840fZYXl+9sMbb3z4hlsLJmADEaRoGAWjYBSMglGACQCWo1ZsdpGDkgAAAABJRU5ErkJggg==","orcid":"","institution":"National University of Córdoba","correspondingAuthor":true,"prefix":"","firstName":"Adrián","middleName":"Maximiliano Moneta","lastName":"Pizarro","suffix":""}],"badges":[],"createdAt":"2024-07-13 17:07:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4736007/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4736007/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":62652411,"identity":"169d83ef-ad4d-49c5-81b3-d9a986b21811","added_by":"auto","created_at":"2024-08-17 01:00:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":5300,"visible":true,"origin":"","legend":"\u003cp\u003eGCFA model\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4736007/v1/355edb9f402b14a670547c10.png"},{"id":62653320,"identity":"8329bc22-69c8-44f3-92dc-c0de5e162757","added_by":"auto","created_at":"2024-08-17 01:08:32","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":11051,"visible":true,"origin":"","legend":"\u003cp\u003eFull GSEM path diagram\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4736007/v1/814f496dafae1dd2893018e9.png"},{"id":68557796,"identity":"ae99746a-10d8-4943-91fc-807c71c3fb6f","added_by":"auto","created_at":"2024-11-08 13:47:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":682338,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4736007/v1/37693547-b1c9-4646-8de4-c6a1d3bfb480.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Understanding the multidimensional structure of poverty in Argentine households","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe multidimensional approach, a phenomenon increasingly promoted and accepted by researchers, governments and society, considers poverty as a complex phenomenon related to multiple deprivations present in dimensions where the monetary element could be just one of these deprivations (Salecker, Ahmadov \u0026amp; Karimli, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This line of sight is opposed to the unidimensional feature of poverty, which traditionally considers low monetary income as the only conceivable source of deprivation. One of the challenges in assuming a multidimensional position emerges when we ask ourselves which dimensions must be thought over, and this argument is closely related to the specific definition of poverty (Kim, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAccording to Walker (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), people are usually aware of poverty at once when they come across it, but many times they face difficulties in asserting exactly what it implies. Although general agreement has been reached when considering poverty as a multidimensional phenomenon, there is little unanimity as to which poverty dimensions should be included (Ntsalaze \u0026amp; Ikhide, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Kim, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) and how they interrelate with each other (Chan y Wong, 2020).\u003c/p\u003e \u003cp\u003eIn this sense, a question of great interest for the design of multidimensional poverty measures is whether they should include a monetary poverty indicator. One of the most common arguments for keeping income or consumption poverty separate from non-monetary poverty is that they reflect different dimensions of the phenomenon (Santos et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The standard method of current income captures cyclical fluctuations in welfare related to the labour market. Instead, non-monetary measures of multidimensional poverty capture the deprivations reflected in less volatile and more structural indicators such as precarious housing and unfavourable socio-environmental conditions. Some empirical studies show that monetary measures of poverty are imperfect predictors of non-monetary measures (Bader et al., 2016; Bourguignon et al., 2010; Roelen, 2017; 2018; Roelen et al., 2009; 2012; Ruggeri et al., 2003; Wang et al., 2016). An analysis of the mismatch between income poverty and the multidimensional poverty index carried out in Chile showed that, although 20.4% of the population was multidimensionally poor and 14.4% was income poor, only 5.5% it was poor on both measures (Ministry of Social Development, 2015). These mismatches may be due to the fact that these measures might be capturing different phenomena or differences in how each indicator is captured and calculated. These findings justify the development of new research to understand the causes of these mismatches and their policy implications (UNDP, 2019).\u003c/p\u003e \u003cp\u003eThe most widespread method to solve the dimension selection problem is to embrace a normative approach, such as the Alkire-Foster axiomatic counting methodology (Alkire \u0026amp; Foster, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Methods such as this one assign the dimension definitions and their indicators to subjective decisions based on sociopolitical agreements and data availability. Their advantages are that they can seize deprivation joint distribution, identify the poor and provide a rate to summarize multidimensional poverty measurement in a single indicator (Alkire, Foster, Seth, Santos, Roche \u0026amp; Ballon, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe use of statistical techniques is a further way to work out the issue associated with dimension identification. Those who stand up for the usage of these methods are very interested in exploring the complex nature and structure of poverty starting from the data themselves. Alkire et al. (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) divide them into two large groups: one related to descriptive techniques and the other based on latent variable modelling. The first group comprises cluster analysis (CA), principal component analysis (PCA) and multiple correspondence analysis (MCA). These methods are quite involved in the problem of dimension reduction. The second group covers factor analysis (FA), latent class analysis (LCA) and structural equation models (SEM). According to Walker (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), this second group can be subdivided into two other groups. On the one hand, methods such as FA and LCA are intended to identify poverty dimensions, and on the other hand, SEM is mainly used for contrasting theories about the relationships between dimensions. In SEM, the poverty dimensions and the path relationship among them are specified beforehand by applying the proper theoretical basis or the previous exploratory analysis, and the actual presence is then contrasted against reality according to that captured by the available data.\u003c/p\u003e \u003cp\u003eObviously, the normative approach is not the most appropriate to analyse whether poverty based on income and non-monetary poverty are indicators of the same construct or reflect different dimensions of the phenomenon, as it assumes that dimensions are predetermined. Statistical methods seem more suitable (N\u0026aacute;jera Catal\u0026aacute;n \u0026amp; Gordon, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the case of Argentina, there is still no regular and systematic official measurement of multidimensional poverty due to the lack of consensus on its composition and limitations in the available data sources. Therefore, the question of whether monetary poverty should be part of a multidimensional poverty index is still unresolved and is a very relevant issue. However, the analysis of multidimensional poverty in this country has been focused mainly on the development of composite indicators where poverty is a linear combination of independent factors built by applying PCA, more appropriate for continuous indicators. Thus, the possibility of a reflective measurement model is ignored, which contemplates the dimensions of poverty as latent variables, not directly observable, and the indicators as particular manifestations of these different facets of the phenomenon, possibly correlated with each other. It also omits that the available indicators are mainly binary variables. Works carried out by Conconi and Ham (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), Conconi (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), Carranza Mena, Pagani and S\u0026aacute;nchez Fern\u0026aacute;ndez (2011) and Gasparini, Sosa Escudero, Marchionni and Olivieri (2013) give examples of descriptive techniques application, basically PCA, for the identification of dimensions. On the other hand, the works of Fagnola and Moneta Pizarro (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and Moneta Pizarro and Satorres Bechara (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) are examples of applications with FA.\u003c/p\u003e \u003cp\u003eThere are no records in Argentina of the application of more appropriate techniques to explore the relationships between the dimensions of poverty, such as SEM, and even less of methods that take into account the binary nature of the available data, for example, the generalized structural equations models (GSEM). The research background in Argentina represents significant advances in terms of the analysis of multidimensional poverty, but does not establish hypotheses that allow us to investigate the relationships between its dimensions. These antecedents are mostly studies based on Sen\u0026rsquo;s capabilities approach (1984, 1985, 19982, 2000), but they only contribute a supporting conceptual framework for the understood multidimensionality hypothesis. This means that this framework is not used to assume theoretical models that make sense of causal relationships and serve to confirm multidimensional structure.\u003c/p\u003e \u003cp\u003eThis paper demonstrates how SEM can be implemented to explore the multidimensional structure of poverty in Argentina and investigate whether monetary poverty should be considered as another indicator of multidimensional poverty. However, given the categorical nature of the available data, the generalized version of these models, GSEM, is used. Specifically, a generalized confirmatory factor analysis (GCFA) model is compared with a full GSEM that includes a second-order factor. The first model proposes to relate each factor with a dimension of poverty without a causal structure that links them. The second model includes structural relationships and proposes two alternative hypotheses at the same time. On the one hand, it assumes the unidimensionality of the non-monetary dimension of poverty, where the non-monetary factors are first-order measures of a single higher-order dimension; on the other hand, it is proposed that non-monetary poverty, represented by the higher order construct, is a consequence of the monetary dimension of poverty. Considering Walker (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) as regards this concept, he asserts that poverty is not only the lack of monetary resources necessary to meet specific needs but also refers to the multiple consequences of this scarcity, such as deficiency in education, healthcare, housing and employment (poverty nonmonetary factors), suffered by poor people all at the same time. It also follows Chan and Wong (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), who applied SEM to Hong Kong data and found that monetary income has a significant impact on the nonmonetary dimensions of poverty.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eAt the international level, there are some precedents in the application of SEM to the analysis of multidimensional poverty that focus more on the development of causal models and less on obtaining synthetic indices. Some examples are Di Tommaso (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) with data from India, Ballon and Krishnakumar (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) with an application to child poverty in Bolivia, Wagle (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) for Nepal and the United States, Kim (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) using data from the United Kingdom, Ballon (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) for female empowerment in Cambodia, Chan and Wong (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) with data from Hong Kong, Zhang and Huai (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) in a paper applied to poverty among farmers in China, and Clausen et al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) to explore the association between multidimensional poverty and depression using data from Per\u0026uacute;. However, the models proposed in this background involve the use of dimensions and variables that seem to be exclusively adjusted to the context of those countries or to the availability of data specific to each case.\u003c/p\u003e \u003cp\u003eDespite the great global acceptance that progress in multidimensional poverty studies has had (Alkire \u0026amp; Santos, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) and the growth of poverty in Argentina, research works in this area are very few at the national level in this country (Ar\u0026eacute;valo \u0026amp; Paz, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Among the main argentine bibliographic antecedents, outstanding works are those by Conconi and Ham (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), Conconi (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), Santos, Villatoro, Mancero and Gerstenfeld (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), Ar\u0026eacute;valo and Paz (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), Salvia, Bonfiglio and Vera (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), Dur\u0026aacute;n and Condor\u0026iacute; (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), Ignacio-Gonz\u0026aacute;lez and Santos (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), Fares, Favata and Mart\u0026iacute;nez (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), Macci\u0026oacute; and Mitchell (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), Sione (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and Poggiese and Iba\u0026ntilde;ez Mart\u0026iacute;n (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These studies share the particularity that they are mainly centred in the construction and use of synthetic indicators such as those corresponding to Bourguignon and Chakravarty (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2003\u003c/span\u003e) and Alkire and Foster (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSome relevant works applying statistical techniques are of great interest for this research. Exemplary antecedents belong to Conconi and Ham (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), Conconi (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), Carraz\u0026aacute;n Mena et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) and Gasparini et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). However, in these cases, factor identification is performed through PCA, and therefore, there is neither progress for the structural model contrast nor attention to the lack of continuity and normality of the variables used as poverty indicators.\u003c/p\u003e \u003cp\u003eRecently, some works were included in the literature about argentine multidimensional poverty where robust FA methods are applied with tetrachoric and polychoric correlation matrices for dimension validity. In this way, problems related to the indicator normality absence are overcome. Works by Fagnola and Moneta Pizarro (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), Moneta Pizarro and Satorres Bechara (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and Guti\u0026eacute;rrez Montecino and Moneta Pizarro (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) refined the indicators of multidimensional poverty obtained from the Permanent Household Survey (EPH, for its acronym in Spanish) through strong techniques of exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). The model originally proposed by Fagnola and Moneta Pizarro (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) has 15 indicators distributed among 5 factors (exclusion, sanitation, health, infrastructure and economic capacity) and takes into account only one EPH wave (third quarter 2017). After considering EFA and CFA tools, 13 indicators remained and were grouped into 4 factors. The work of Moneta Pizarro and Satorres Bechara (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) is an extension of the previous work to longitudinal data of representative periods related to the different phases of the argentine economic cycle, allowing the contrast of the longitudinal invariance of the factor structure. These authors enlarged the proposal to 18 indicators in 5 factors, but after consistency analyses, construct validity and longitudinal invariance, there remained 10 selected indicators in 3 factors (housing, environment and income). Lastly, Guti\u0026eacute;rrez Montecino and Moneta Pizarro (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) also contrast multiregional invariance, and a measuring model with 7 indicators and 3 factors remains validated. These papers represent significant advances in the identification with robust methods of the factor structure of multidimensional poverty in Argentina, but they do not analyse possible relationships between the latent constructs.\u003c/p\u003e \u003cp\u003eAt both international and national level, there is no evidence of applications of GSEM, the most advanced version of SEM, in the available literature on multidimensional poverty. GSEM is a novel alternative that is considered more appropriate for dealing with complex data structures (Skrondal \u0026amp; Rabe-Hesketh, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2004\u003c/span\u003e).\u003c/p\u003e"},{"header":"3. Data","content":"\u003cp\u003eThe elements used are microdata at the household level from the cross-section corresponding to the 2022 first quarter of the EPH from the National Institute of Statistics and Census (INDEC, for its acronym in Spanish). It is a sample of 16898 observations corresponding to households in 32 urban agglomerates in all regions of the country. Using this kind of data represents some disadvantages: they correspond to urban conglomerate samples with more than one hundred thousand inhabitants, and the questionnaire applied was not specifically designed to demonstrate multidimensional poverty characteristics. Survey conclusions are then limited by these restrictions. Another problem is that prospect variables to set up poverty indicators are mostly of dichotomous type (for example, whether the household head is employed or unemployed, whether the family has medical coverage, or if the housing has running water, among others). Therefore, this reveals limitations for the multivariate analysis and particularly for SEM application, working with preferably normal continuous variables and distributions. Nevertheless, this drawback is overcome by using GSEM, a more advanced and appropriate strategy of statistical modelling for these kinds of data.\u003c/p\u003e \u003cp\u003eWith respect to the variables, the measurement model validated by Guti\u0026eacute;rrez Montecino and Moneta Pizarro (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) is used as the starting point, considering that it has 7 indicators grouped into 3 factors. The first factor is related to the conditions of the housing infrastructure, and the indicators are:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eRoof: It takes the value 1 if the roof is of low quality (plastic, cardboard, cane, planks or sheets without ceiling or inner lining) and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eBathroom: it takes the value 1 if it does not have a bathroom with drainage inside the housing and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe second factor is related to the environment or surroundings of the housing, and the indicators are as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eDumpsite: it takes the value 1 if the housing is less than 3 blocks from a dumpsite and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eFloodable area: it takes the value 1 if the housing is in a floodable area (in the last 12 months) and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eFinally, the third factor includes indicators related to economic home resources:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eExternal support takes the value 1 if the home receives external monetary or material support in the form of subsidies, assistance programs, charity and others and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eMedical coverage: It takes the value 1 if any of the members of the housing unit does not pay or is deducted payment for medical coverage services and 0 otherwise.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eTFI\u0026thinsp;\u0026lt;\u0026thinsp;TBB: it takes the value 1 in the case that the total family income is less than the total specific basic basket of that home and 0 otherwise.\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe first two factors are related to nonmonetary dimensions of poverty, while the third factor is strictly associated with a monetary dimension. Descriptive statistics for all these variables are presented in the following table, where the means, since all variables are binary, represent the proportion of observations with values equal to 1.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eObs.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStd. dev.\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoof\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.2598\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBathroom\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.2814\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDumpsite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0518\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.2216\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFloodable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0541\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.2262\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExternal support (Ext_supp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.2310\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.4215\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedical coverage (Med_cov)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.3628\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.4808\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTFI\u0026thinsp;\u0026lt;\u0026thinsp;TBB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.3161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.4650\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"4. Methodology","content":"\u003cp\u003eAs stated above, GSEM techniques are applied in this work. This modelling combines SEM capacities with generalized linear model (GLM) capacities. Similar to the econometric methods of simultaneous equations, SEM allows the simultaneous examination of a group of dependency relationships where some variables act as predictor and dependent variables at the same time but with the capacity to estimate and evaluate the relationship among latent (unobservable) variables. These variables are constructs assumed from the theory that they can be measured through observable variables (Cupani, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Compared with other analysis techniques where the constructs are represented by a single measurement and the measurement error is not modelled, multiple measures are used in SEM to represent each construct and control the specific measurement error of each variable, thus allowing the assessment of the validity of each construct (Ruiz, Pardo \u0026amp; San Mart\u0026iacute;n, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAs Kline (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) shows, every SEM has two elements: (a) a measurement model representing the relationships between the latent variables and their manifest indicators and (b) a structural model describing the interrelationship among the latent constructs. By using the measurement model, the aim is to verify the adequacy of the indicators selected in the measurement of relevant constructs. The structural relationship model is the one truly desired to be estimated. It includes the effects and relationships among the constructs, which are normally latent variables. It is similar to a regression model, but it may have concatenated effects and loops among variables.\u003c/p\u003e \u003cp\u003eAnother particular characteristic of SEM is that it has several statistical tests and a group of goodness of fit indicators, but the adjustment is verified when the values of the estimated parameters reproduce the observed covariance matrix as strictly as possible (Kahn, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). For this purpose, in SEM, the model estimation is based on the existing correlations among measured variables in a cross-sectional sample. Instead of reducing the difference between the predicted values and the ones observed at the individual level (least squares method), the difference between the covariances observed in the sample and the covariances predicted by the structure model is reduced. According to Long (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1983\u003c/span\u003e), this is the reason why these models are also called covariance structure models. Therefore, the model residuals are the differences between the covariances observed and the covariances predicted by the theoretical structure model (Ruiz et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBy combining SEM with GLM, in GSEM, it is possible to work with response variables that may be continuous, binary, ordinal, count or multinominal variables. In addition, both normal linear regressions and the large spectrum of regressions of the exponential family (gamma, logit, probit, Poisson, negative binomial and their variants) can be modelled (Skrondal \u0026amp; Rabe-Hesketh, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). In this research, the presence of binary variables makes it necessary to apply GSEM instead of SEM.\u003c/p\u003e \u003cp\u003eIn this work, first, a measurement model (without a structural part) with a logit link is specified, that is, a CFA model adapted for variables of dichotomous or binary response (GCFA). In this model, as indicated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e below, it is assumed that each latent factor is one poverty dimension that can be measured through observable indicators and that the factors may be correlated. This first model is therefore made up of three measurement submodels, one for each factor. Two observed indicators, roof and bathroom, help us to approach the housing conditions; two other observed indicators, dumpsite and floodable area, are assumed to be manifestations of the environment conditions; and three indicators observed measure the economic home resources. The three measurement models are jointly estimated, and this permits correlations among factors. As McGartland Rubio et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2001\u003c/span\u003e) indicate, the verification of these correlations is usually understood as the result of the existence of a higher-order factor. Nevertheless, this may not necessarily be the reason, since the correlation can be because the factors measure different dimensions of one construct, poverty in this case.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eUsing matrix notation, this first model can be represented through the measurement Eq.\u0026nbsp;(\u003cspan refid=\"Equ1\" class=\"InternalRef\"\u003e1\u003c/span\u003e):\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:\\text{logit}\\left[\\text{Pr}\\left(\\left.w=1\\right|\\xi\\:\\right)\\right]=\\:{{\\Lambda\\:}}_{w}\\xi\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:w\\)\u003c/span\u003e\u003c/span\u003e is the vector of observed indicators, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:1\\)\u003c/span\u003e\u003c/span\u003e is a vector of ones, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\xi\\:\\)\u003c/span\u003e\u003c/span\u003e is the vector of latent factors (poverty dimensions) with covariance matrix Φ and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\Lambda\\:}}_{w}\\)\u003c/span\u003e\u003c/span\u003e is the matrix of model coefficients (factor loadings).\u003c/p\u003e \u003cp\u003eSecond, a full GSEM is specified where two factors, housing and environment, are demonstrations of nonmonetary poverty, that is, a second-order construct, and the factor related to economic resources or monetary poverty is an exogenous latent variable explaining nonmonetary poverty. This model is represented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. It should be noted that the measurement indicators for each factor are the same as those in the first model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe full GSEM can be expressed in matrix form using equations (\u003cspan refid=\"Equ2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), (\u003cspan refid=\"Equ3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) and (4).\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:\\eta\\:=\\:\\text{B\\:}\\eta\\:+{\\Gamma\\:}\\:\\tau\\:+\\epsilon\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\:\\text{logit}\\left[\\text{Pr}\\left(\\left.y=1\\right|\\eta\\:\\right)\\right]=\\:{{\\Lambda\\:}}_{y}\\eta\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\:\\text{logit}\\left[\\text{Pr}\\left(\\left.x=1\\right|\\tau\\:\\right)\\right]=\\:{{\\Lambda\\:}}_{x}\\tau\\:$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eEquation (\u003cspan refid=\"Equ2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) represents the structural part of the model, while equations (\u003cspan refid=\"Equ3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) and (\u003cspan refid=\"Equ4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) are the measurement model for latent factors. In the structural part, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\eta\\:\\)\u003c/span\u003e\u003c/span\u003e is the vector of endogenous latent variables (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{1},\\:{\\eta\\:}_{2},\\:{\\eta\\:}_{3}\\)\u003c/span\u003e\u003c/span\u003e), with \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{1}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{2}\\)\u003c/span\u003e\u003c/span\u003e being the first-order factors (housing and environment) and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{3}\\)\u003c/span\u003e\u003c/span\u003e being the second-order factor (nonmonetary poverty), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\tau\\:\\)\u003c/span\u003e\u003c/span\u003e is the only exogenous latent variable (economic resources), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{B}\\)\u003c/span\u003e\u003c/span\u003e is the 3\u0026times;3 coefficient matrix of the endogenous latent variables, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Gamma\\:}\\)\u003c/span\u003e\u003c/span\u003e is a coefficient vector of order 3 whose elements are all equal to zero except the third one that represents the effect of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\tau\\:\\)\u003c/span\u003e\u003c/span\u003e on \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{3}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e is the vector of disturbances (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\epsilon\\:}_{1},\\:{\\epsilon\\:}_{2},\\:{\\epsilon\\:}_{3}\\)\u003c/span\u003e\u003c/span\u003e) associated with the endogenous latent variables included in \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\eta\\:\\)\u003c/span\u003e\u003c/span\u003e with a diagonal matrix of covariances \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Psi\\:}\\)\u003c/span\u003e\u003c/span\u003e. In the measurement model, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:y\\)\u003c/span\u003e\u003c/span\u003e is the vector of indicators used to measure \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{1}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\eta\\:}_{2}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:x\\)\u003c/span\u003e\u003c/span\u003e is the vector of indicators for the measurement of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\tau\\:\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\Lambda\\:}}_{y}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\Lambda\\:}}_{x}\\)\u003c/span\u003e\u003c/span\u003e are factor loading matrices and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:1\\)\u003c/span\u003e\u003c/span\u003e is a vector of ones.\u003c/p\u003e \u003cp\u003eWith the help of Stata 17, both models are estimated with maximum likelihood (ML), which is the only available method for GSEM. In the case of the second model, its high complexity hindered difficulties in achieving convergence in the estimation process. This problem was temporally solved by modifying the numerical integration model and reducing the number of integration points.\u003c/p\u003e \u003cp\u003eThe results were satisfactory by using the model of the non-adaptive Gauss-Hermite quadrature and by reducing the points of numerical integration to 3. Afterwards, these estimations were used as improved initial values to re-estimate the model with the default options, which allowed arriving at a solution with greater accuracy. This was achieved with the adaptive Gauss-Hermite quadrature algorithm based on the mean and the variance and 7 points of numerical integration.\u003c/p\u003e"},{"header":"5. Results","content":"\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e indicates the estimation results of the pure measurement model, that is, the GCFA model, which permits us to contrast the convergent and discriminant validity of the three factors proposed for multidimensional poverty. By observing such results, all the indicators are significantly related to their corresponding latent constructs. This supports the convergent validity of the model, that is, that the indicator variables of each factor are strongly correlated with each other, sharing a high proportion of the variance (Ald\u0026aacute;s \u0026amp; Uriel, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The covariances among the latent factors are also significantly different from zero. Departing from the estimated variances and covariances, the correlations obtained are equal to 0.51 between housing and environment, 0.60 between housing and income and 0.32 between environment and income. These moderate correlation values indicate discriminant validity.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResults of the GCFA model\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStd. Err.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ez\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eP\u003cem\u003e\u0026gt;\u003c/em\u003e|\u003cem\u003ez\u003c/em\u003e|\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e[95% Conf. Int.]\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoof\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eHousing\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c8\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-4.232543\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1449429\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-29.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-4.516626\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-3.94846\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBathroom\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eHousing\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9593894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0813283\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.7999889\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.11879\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-3.85143\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1272349\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-30.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-4.100806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-3.602054\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDumpsite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEnvironment\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-4.199538\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.167694\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-25.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-4.528213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-3.870864\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFloodable area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEnvironment\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.288781\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1995092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.8977506\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.679812\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-4.802548\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.3051847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-15.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-5.400699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-4.204397\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eExternal support\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c8\" namest=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.809089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0375684\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-48.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.882722\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-1.735457\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMedical coverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c8\" namest=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.474708\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0745719\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e19.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.32855\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.620866\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.093027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0417698\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-26.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.174894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-1.011159\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eTFI\u0026thinsp;\u0026lt;\u0026thinsp;TBB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c8\" namest=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.263382\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0554054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.15479\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.371975\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.422182\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0424695\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-33.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.505421\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-1.338944\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003evar(Housing)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.305203\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.5600140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e4.313697\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e6.524607\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003evar(Environment)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.453833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.5148337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.578813\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e4.625756\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003evar(Economic resources)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.105572\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1776731\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.776153\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3.474079\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003ecov(Housing/Environment)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.188527\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.2328391\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.73217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2.644883\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003ecov(Housing, Econ. res.)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.471096\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1573252\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e15.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.162744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2.779448\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003ecov(Environment, Econ. res.)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.075617\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1115909\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.8569024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.294331\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe fact that the covariances among factors are significant demonstrates that the presence of a second-order factor and structural relationships among the latent variables are reasonable. In this way, the specification corresponding to the full GSEM is plausible. Table\u0026nbsp;3 exhibits the results from this estimation. According to the measurement modelling, all the indicators of this second model are also significantly related to their corresponding latent factors, and as expected, the estimated coefficients assume very similar values to those of the pure measurement model. By analysing Table\u0026nbsp;3, estimated covariances among constructs are not included since they have been replaced by structural relationships. All these structural coefficients were also significant. Therefore, the hypotheses presented for this second model are supported by the empirical evidence.\u003c/p\u003e \u003cp\u003eTable 3. Results of full GSEM\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStd. Err.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ez\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP\u0026gt;|z|\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e[95% Conf. Int.]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeasurement model\u003c/p\u003e \u003cp\u003eRoof\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" morerows=\"1\" nameend=\"c4\" namest=\"c2\" rowspan=\"2\"\u003e \u003cp\u003e1 (restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-4.246725\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1571272\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-27.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-4.554688\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-3.938761\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBathroom\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.955539\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0866052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.7857958\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.125282\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-3.854903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1302366\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-29.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-4.110162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-3.599644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDumpsite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnvironment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-4.19829\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1645436\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-25.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-4.52079\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-3.875791\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFloodable Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnvironment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.289765\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1964518\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.9047261\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.674803\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-4.803195\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3025952\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-15.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-5.396271\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-4.21012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExternal support\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.809797\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0395883\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-45.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-1.887389\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.732206\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedical coverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.474086\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0779903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.321228\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.626944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.093534\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0417848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-26.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-1.17543\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.011637\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTFI\u0026thinsp;\u0026lt;\u0026thinsp;TBB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.263164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0579683\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.149548\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.376779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003econstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.422828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0424854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-33.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-1.506098\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-1.339558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStructural model\u003c/p\u003e \u003cp\u003eHousing\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-monetary poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c7\" namest=\"c3\"\u003e \u003cp\u003e(restricted)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnvironment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-monetary poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.4383168\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0493818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.3415303\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.5351033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-monetary poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEconomic resources\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.7941819\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.057259\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.6819563\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.9064074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003evar(e.Non-monetary poverty)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.001571\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.4559609\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.228665\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e4.042523\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003evar(e.Housing)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.383752\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3345924\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0694848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.119391\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003evar(e.Environment)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.492091\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3747607\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.855924\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.346321\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003evar(e.Economic resources)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.104728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1953692\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.744483\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.51226\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eComparing the models with the Akaike (AIC) and Schwarz (BIC) information criteria, the full GSEM has lower values in both cases. The GCFA model has an AIC equal to 82658.00 and a BIC equal to 82789.49, while the GSEM model achieves an AIC equal to 82657.17 and a BIC equal to 82788.66. Consequently, GSEM is the relatively best fit to the data.\u003c/p\u003e"},{"header":"6. Discussion","content":"\u003cp\u003eThe results show that with the EPH data, in spite of their limitations for multidimensional poverty analysis, it is possible to confirm the presence of at least three dimensions of poverty. One of them is associated with the material conditions of the housing, another one is related to the environmental surrounding and the last one corresponds to the monetary income. However, it is more realistic to consider that the nonmonetary factors of poverty, such as housing and environmental conditions, are different indicators of the same dimension of higher order and that this nonmonetary poverty, as a whole, is explained by the monetary dimension. Empirical support is then provided for the identification of two dimensions instead of three. One dimension is associated with the lack of monetary resources, and the other is related to nonmonetary deprivations.\u003c/p\u003e \u003cp\u003eIn this way, it cannot be excluded that poverty is multidimensional in Argentina, but a structural relationship is supported among its dimensions. This means, as Walker (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) points out, that poverty is manifested not only by the presence of inadequate monetary income but also by the multiple consequences of this absence regarding housing and environment, which are part of a nonmonetary dimension of the phenomenon. This conclusion, seeming quite obvious, is opposed to a great part of the literature about multidimensional poverty developed up to present time, which sometimes implicitly and other times explicitly associates the correlation among poverty factors to the presence of different dimensions or attributes of one same construct. Kim (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), who in his measurement model includes economic resources as one more poverty dimension and is based on Kangas and Ritakallio (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1998\u003c/span\u003e), Lelli (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), Whelan (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1993a\u003c/span\u003e) and Whelan (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e1993b\u003c/span\u003e), affirms that this it is common because these resources can be applied to other functions related to poverty, for example, to healthy food purchases. The criticism that can be made of this type of position is that they do not inquire in depth about these links and the possible presence of structural relationships between the dimensions. The exception to this may be the trap models of poverty. Beneath the theoretical framework of these models, poverty is characterized as a vicious circle due to the presence of self-reinforcing mechanisms that provoke its continuation (Santos, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) and therefore postulate in this way the possibility of causal relationships among their dimensions.\u003c/p\u003e"},{"header":"7. Implications","content":"\u003cp\u003eThe implication of these conclusions for future research and public policy designs is that, on the basis of the data used for this research and the evidence found, poverty in Argentina should not be measured by using one multidimensional index but by using two. These should be a monetary poverty indicator on the one hand and a nonmonetary poverty indicator on the other. In other words, income poverty should not be mixed with other dimensions of multidimensional poverty since it is very likely the reason for the nonmonetary dimensions.\u003c/p\u003e \u003cp\u003eThus, and differently from what was proposed by Santos et al. (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), the recommendation to keep the income poverty and nonmonetary poverty indicators separate is supported. The arguments that are generally put forward to recommend this are very well compiled in the manual of the United Nations Development Programme and the Oxford Poverty and Human Development Initiative of the University of Oxford (UNDP \u0026amp; OPHI, 2019), where it is stated that if the objective to elaborate a multidimensional poverty measure is to complement the current statistics of monetary poverty, then the inclusion of a dimension related to income adds noise. In these cases, it is appropriate to look beyond the manifestations of economic capacity, broadening the understanding of the phenomenon towards nonmonetary dimensions not captured by the traditional measures of monetary poverty. Among other arguments, it also points out the importance of recognizing that monetary and nonmonetary indicators capture poverty differently. Monetary indicators are generally considered indirect measures of poverty because they focus on the scarcity of resources for the acquisition of basic goods and services, while multidimensional indexes developed on the basis of nonmonetary indicators are considered direct measures of lack of well-being because they reflect real deprivations.\u003c/p\u003e \u003cp\u003eHowever, this study provides another reason for the monetary and nonmonetary poverty measures to be kept separate, and it is the finding of empirical evidence, at least for the case of Argentina, in favor of the fact that monetary poverty is a good predictor of nonmonetary poverty. This result could be associated with a close relationship between the indicators of one type of poverty and the other, suggesting a certain coincidence between both measures and that they could be combined without problems in a single measurement. But the correlation is not perfect, as the results show for the structural regression of the nonmonetary dimension as a function of the monetary one. The fact that economic resources partially explain nonmonetary poverty does not mean that they are useful for measuring the same concept. In contrast, both the measurement model proposed through the GCFA and the complete GSEM assume that the household economic capacity indicators are manifestations of one construct, while the rest, the nonmonetary ones, are manifestations of others. The empirical evidence found here confirms that these measures should not be combined, but not because they are independent or do not coincide in the estimation of the poor population but because it is more plausible, as stated in the complete GSEM, that one is the cause of the other. These results can be a constructive contribution to the debate on the relationship between monetary and multidimensional poverty measures, which for the Latin American case is very appropriately discussed in Santos et al. (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e"},{"header":"8. Conclusion","content":"\u003cp\u003eThe aim of this paper was to deepen the multidimensional analysis of poverty in Argentina. By exploring the multidimensional nature of poverty, the study aimed to highlight the importance of considering various dimensions of deprivation beyond monetary income. Furthermore, it set out to address the complexities and challenges associated with determining the specific dimensions that should be incorporated into multidimensional poverty measures, thereby contributing to a more comprehensive understanding and improving the effectiveness of poverty alleviation efforts in Argentina.\u003c/p\u003e \u003cp\u003eKey aspects of the methodology employed included the identification and selection of relevant dimensions, the comparison of rival models to analyse the interrelationships between these dimensions, and the use of advanced statistical techniques such as GSEM to assess multidimensional poverty in Argentina. Ideas from previous research on robust measurement methods were incorporated and empirical data from the EPH were used to validate the proposed models.\u003c/p\u003e \u003cp\u003eThe results of the study highlighted the effectiveness of the multidimensional approach in capturing the complexities of poverty beyond mere monetary measures. The testing of the models developed provided valuable insights for a more complete understanding and identified the interrelationships between the different facets of the phenomenon in Argentina. The empirical validation of these models corroborated the robustness of the multidimensional approach in capturing the nuances of poverty and guiding interventions for the development of more appropriate indicators. By exploring various dimensions of deprivation, the research provided evidence on the interconnected nature of the different aspects of poverty and underlined the need to separate the monetary dimension from the non-monetary ones. In this way, the study provides a framework for designing appropriate indicators of multidimensional poverty, as well as for the implementation of more effective and targeted comprehensive policies to combat this scourge.\u003c/p\u003e \u003cp\u003eIn summary, this research on multidimensional poverty not only contributes significantly to the academic literature, but also has important implications for policy and practice. By delving into the complexities of poverty measurement beyond income and exploring the multidimensional nature of deprivation, the study offers a more comprehensive understanding of poverty. The conclusions highlight the need to adopt a multidimensional approach, complementary to the traditional monetary one, when developing poverty measures and designing programs for its alleviation.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by Secretariat of Science and Technology of the National University of C\u0026oacute;rdoba under grant code 33620230100448CB.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author has no financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis is an observational study using open data from the Encuesta Permanente de Hogares (EPH) published by the Instituto Nacional de Estad\u0026iacute;sticas y Censos (INDeC) of Argentina. No ethical approval is required.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eThe entire manuscript has been produced and written by a single author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAld\u0026aacute;s, J., \u0026amp; Uriel, E. (2017). \u003cem\u003eAn\u0026aacute;lisis multivariante aplicado con R\u003c/em\u003e (2nd Ed.). Ed. Paraninfo. \u003c/li\u003e\n\u003cli\u003eAlkire, S., \u0026amp; Foster, J. (2011). Counting and Multidimensional Poverty Measurement. \u003cem\u003eJournal of Public Economics\u003c/em\u003e, \u003cem\u003e95\u003c/em\u003e(7-8), 476\u0026ndash;487. https://doi.org/10.1016/j.jpubeco.2010.11.006 \u003c/li\u003e\n\u003cli\u003eAlkire, S., Foster, J. E., Seth, S., Santos, M. E., Roche, J. M., \u0026amp; Ballon, P. (2015). \u003cem\u003eMultidimensional Poverty Measurement and Analysis: Chapter 3 \u0026ndash; Overview of Methods for Multidimensional Poverty Assessment\u003c/em\u003e. OPHI Working Paper, 84. https://ophi.org.uk/multidimensional-poverty-measurement-and-analysis-chapter-3-overview-of-methods-for-multidimensional-poverty-assessment/ \u003c/li\u003e\n\u003cli\u003eAlkire, S., \u0026amp; Santos, M. E. (2010). \u003cem\u003eAcute Multidimensional Poverty: A New \u003c/em\u003e\u003cem\u003eIndex for Developing Countries\u003c/em\u003e. OPHI Working Paper, 38. https://ophi.org.uk/acute-multidimensional-poverty-a-new-index-for-developing-countries/ \u003c/li\u003e\n\u003cli\u003eAlkire, S. \u0026amp; Santos, M. E. (2013). A Multidimensional Approach: Poverty Measurement \u0026amp; Beyond. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cem\u003e112\u003c/em\u003e, 239\u0026ndash;257. https://doi.org/10.1007/s11205-013-0257-3 \u003c/li\u003e\n\u003cli\u003eAr\u0026eacute;valo, C., \u0026amp; Paz, J. A. (2015). \u003cstrong\u003e\u003cem\u003e \u003c/em\u003e\u003c/strong\u003e\u003cem\u003eP\u003c/em\u003e\u003cem\u003eobreza\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003een\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003ela\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003eArg\u003c/em\u003e\u003cem\u003een\u003c/em\u003e\u003cem\u003etina.\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003ePriv\u003c/em\u003e\u003cem\u003eaciones\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003em\u003c/em\u003e\u003cem\u003eu\u003c/em\u003e\u003cem\u003e\u0026acute;\u003c/em\u003e\u003cem\u003el\u003c/em\u003e\u003cem\u003etiples \u003c/em\u003e\u003cem\u003ey asimetr\u0026iacute;as regionales\u003c/em\u003e. Documento de Trabajo No. 15, Instituto de Estudios Laborales y del Desarrollo Econo\u0026acute;mico (IELDE), Facultad de Ciencias Econo\u0026acute;micas, Jur\u0026iacute;dicas y Sociales, Universidad Nacional de Salta (UNSa). https://www.economicas.unsa.edu.ar/ielde/index.php/documentos-de-trabajo/167-nro-15-primavera-2015-carla-arevalo-y-jorge-paz-pobreza-en-la-argentina-privaciones-multiples-y-asimetrias-regionales \u003c/li\u003e\n\u003cli\u003eBallon, P., \u0026amp; Krishnakumar, J. (2008, august). \u003cem\u003eA Model-Based Multidimensional Capability Deprivation Index\u003c/em\u003e (Paper). 30th General Conference of The International Association for Research in Income and Wealth, Portoroz, Slovenia. http://old.iariw.org/papers/2008/ballon.pdf \u003c/li\u003e\n\u003cli\u003eBallon, P. (2018). A Structural Model of Female Empowerment. \u003cem\u003eThe Journal of Development Studies\u003c/em\u003e, \u003cem\u003e54\u003c/em\u003e(8), 1303-1320. https://doi.org/10.1080/00220388.2017.1414189 \u003c/li\u003e\n\u003cli\u003eBourguignon, F., \u0026amp; Chakravarty, S. (2003). The Measurement of Multidimensional Poverty\u003cstrong\u003e\u003cem\u003e.\u003c/em\u003e\u003c/strong\u003e \u003cem\u003eJournal of Economic Inequality\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 25\u0026ndash;49. https://doi.org/10.1023/A:1023913831342 \u003c/li\u003e\n\u003cli\u003eCarraz\u0026aacute;n Mena, G., Pagani, P. A., \u0026amp; S\u0026aacute;nchez Fern\u0026aacute;ndez, D. (2011). \u003cem\u003eAn\u0026aacute;lisis multivariado de la pobreza en la Ciudad de Salta. Una aproximaci\u0026oacute;n a partir del Censo Social 2009-2010\u003c/em\u003e. Documento de Trabajo No. 3, Instituto de Investigaciones Econ\u0026oacute;micas (IEE), Universidad Nacional de Salta. https://www.economicas.unsa.edu.ar/iie/archivos/dt/2011S%C3%A1nchez.pdf \u003c/li\u003e\n\u003cli\u003eChan, S. M., \u0026amp; Wong, H. (2020). Impact of Income, Deprivation and Social Exclusion on Subjective Poverty: A Structural Equation Model of Multidimensional Poverty in Hong Kong. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, 152, 971\u0026ndash;99. https://doi.org/10.1007/s11205-020-02476-8 \u003c/li\u003e\n\u003cli\u003eClausen, J., Barrantes, N., Caballero, E., \u0026amp; Guill\u0026eacute;n, H. (2024). Exploring the Association between Multidimensional Poverty and Depression Using Structural Equation Models. \u003cem\u003eApplied Research Quality Life\u003c/em\u003e, 19, 727\u0026ndash;747. https://doi.org/10.1007/s11482-023-10262-0 \u003c/li\u003e\n\u003cli\u003eConconi, A., \u0026amp; Ham, A. (2007). \u003cem\u003ePobreza Multidimensional Relativa: Una Aplicaci\u003c/em\u003e\u003cem\u003eo\u003c/em\u003e\u003cem\u003e\u0026acute;\u003c/em\u003e\u003cem\u003en\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003ea\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003ela\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003eArg\u003c/em\u003e\u003cem\u003een\u003c/em\u003e\u003cem\u003etina\u003c/em\u003e. Documento de Trabajo No. 57, Centro de Estudios Distributivos, Laborales y Sociales (CEDLAS), Universidad Nacional de La Plata. http://sedici.unlp.edu.ar/handle/10915/3616 \u003c/li\u003e\n\u003cli\u003eConconi, A. (2011). \u003cem\u003ePobreza Multidimensional en Argentina: Ampliando \u003c/em\u003e\u003cem\u003elas Medidas Tradicionales de Pobreza por Ingreso y NBI\u003c/em\u003e. Documento de Trabajo No. 90, Departamento de Econom\u0026iacute;a, Facultad de Ciencias Econ\u0026oacute;micas, Universidad Nacional de La Plata. https://www.depeco.econo.unlp.edu.ar/wp/wp-content/uploads/2017/05/doc90.pdf \u003c/li\u003e\n\u003cli\u003eCupani, M. (2012). An\u0026aacute;lisis de Ecuaciones Estructurales: conceptos, etapas de desarrollo y un ejemplo de aplicaci\u0026oacute;n. \u003cem\u003eRevista Tesis\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e, 186\u0026ndash;199. https://rdu.unc.edu.ar/bitstream/handle/11086/22039/16.pdf?sequence=1\u0026amp;isAllowed=y \u003c/li\u003e\n\u003cli\u003eDi Tommaso, M. (2007). Children Capabilities: A Structural Equation Model for India. \u003cem\u003eJournal of Socio-Economics\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e(3), 436\u0026ndash;450. https://doi.org/10.1016/j.socec.2006.12.006 \u003c/li\u003e\n\u003cli\u003eDur\u0026aacute;n, R. J., \u0026amp; Condor\u0026iacute;, M. A. (2017). Deprivation Index for Small Areas Based on Census Data in Argentina. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cem\u003e141\u003c/em\u003e, 331\u0026ndash;363. https://doi.org/10.1007/s11205-017-1827-6 \u003c/li\u003e\n\u003cli\u003eFagnola, B., \u0026amp; Moneta Pizarro, A. M. (2021). \u003cstrong\u003e\u003cem\u003e \u003c/em\u003e\u003c/strong\u003eIdentificaci\u0026oacute;n de la pobreza multidimensional en Argentina con m\u0026eacute;todos robustos de an\u0026aacute;lisis factorial. \u003cem\u003eRevista Cuadernos del CIMBAGE\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e. https://ojs.econ.uba.ar/index.php/CIMBAGE/article/view/2053 \u003c/li\u003e\n\u003cli\u003eFares, F. M., Favata, F., \u0026amp; Mart\u0026iacute;nez, R. G. (2021). Una propuesta para la medici\u0026oacute;n de la pobreza multidimensional en la Argentina (2004-2019). \u003cem\u003eEconom\u0026iacute;a y Desaf\u0026iacute;os del Desarrollo\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(7), 4-48. https://revistaedd.unsam.edu.ar/?p=1454 \u003c/li\u003e\n\u003cli\u003eGasparini, L., Sosa Escudero, W., Marchionni, M., \u0026amp; Olivieri, S. (2013). Multidimensional poverty in Latin America and the Caribbean: new evidence from the Gallup World Poll. \u003cem\u003eJournal of Economic Inequality\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(2), 195\u0026ndash;214. https://doi.org/10.1007/s10888-011-9206-z \u003c/li\u003e\n\u003cli\u003eGuti\u0026eacute;rrez Montecino, D. A., \u0026amp; Moneta Pizarro, A. M. (2021, october). \u003cem\u003eInvarianza\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cem\u003emul\u003c/em\u003e\u003cem\u003etirregional de la pobreza multidimensional en Argentina\u003c/em\u003e (Paper). XIV Congreso Latinoamericano de Sociedades de Estad\u0026iacute;stica \u0026ldquo;Laura Nalbarte\u0026rdquo;, Montevideo, Uruguay. https://sue.org.uy/wp-content/uploads/2022/05/Libro_de_resumenes_XIV_CLATSE.pdf \u003c/li\u003e\n\u003cli\u003eIgnacio-Gonz\u0026aacute;lez, F. A., \u0026amp; Santos, M. E. (2020). Pobreza multidimensional urbana en Argentina. \u0026iquest;Reducci\u0026oacute;n de las disparidades entre el norte grande argentino y centro-cuyo-sur? (2003-2016). \u003cem\u003eCuadernos de Econom\u0026iacute;a\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(81), 795-822. https://doi.org/10.15446/cuad.econ.v39n81.76486 \u003c/li\u003e\n\u003cli\u003eKahn, J. H. (2006). Factor analysis in Counseling Psychology research, training and practice: Principles, advances and applications. \u003cem\u003eThe Counseling Psychologist\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(5), 1\u0026ndash;36. https://doi.org/10.1177/0011000006286347 \u003c/li\u003e\n\u003cli\u003eKangas, O., \u0026amp; Ritakallio, V.-M. (1998). Different methods\u0026mdash;different results? approaches to multidimensional poverty. In H.-J. Andress (Ed.), \u003cem\u003eEmpirical poverty research in a comparative perspective\u003c/em\u003e (pp. 167\u0026ndash;203). Ashgate.\u003c/li\u003e\n\u003cli\u003eKim, S-G. (2016). What Have We Called as \u0026ldquo;Poverty\u0026rdquo;? A Multidimensional and Longitudinal Perspective. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cem\u003e129\u003c/em\u003e, 229\u0026ndash;276. https://doi.org/10.1007/s11205-015-1101-8 \u003c/li\u003e\n\u003cli\u003eKline, R. B. (2015). \u003cem\u003ePrinciples and Practice of Structural Equation Modeling\u003c/em\u003e (4th Ed.). Guilford Press.\u003c/li\u003e\n\u003cli\u003eLelli, S. (2001). \u003cem\u003eFactor analysis vs. Fuzzy sets theory: Assessing the influence of different techniques on Sen\u0026rsquo;s functioning approach\u003c/em\u003e. Discussion Paper Series (DPS) 01.21, Center for Economic Studies, K. U. Leuven. http://www.econ.kuleuven.be/ces/discussionpapers/default.htm \u003c/li\u003e\n\u003cli\u003eLong, J. S. (1983). Covariance Structure Models: An introduction to LISREL. \u003cem\u003eSage University Paper Series on Quantitative Applications in the Social Sciences\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eMacci\u0026oacute;, J. \u0026amp; Mitchell, A. (2023). Medici\u0026oacute;n multidimensional de pobreza en ciudades segregadas: evidencia de la ciudad de Buenos Aires. \u003cem\u003eRevista Desarrollo y Sociedad\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(93), 101-137. https://doi.org/10.13043/DYS.93.3 \u003c/li\u003e\n\u003cli\u003eMcGartland Rubio, D., Berg-Weger, M., \u0026amp; Tebb, S. S. (2001). Using Structural Equation Modeling to Test for Multidimensionality. \u003cem\u003eStructural Equation Modeling\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(4), 613\u0026ndash;626. https://doi.org/10.1207/S15328007SEM0804_06 \u003c/li\u003e\n\u003cli\u003eMoneta Pizarro, A. M., \u0026amp; Satorres Bechara, A. P. (2021). Estabilidad dina\u0026acute;mica de la pobreza multidimensional en Argentina. In R. P\u0026eacute;rez Calle, E. Trincado Aznar \u0026amp; E. Gallego Abaroa (Eds.), \u003cem\u003eEconom\u0026iacute;a, empresa y justicia. Nuevos retos para el futuro \u003c/em\u003e(pp. 1337\u0026ndash;1365). Editorial Dykinson S.L. https://www.dykinson.com/libros/economia-empresa-y-justicia-nuevos-retos-para-el-futuro/9788413773261/ \u003c/li\u003e\n\u003cli\u003eN\u0026aacute;jera Catal\u0026aacute;n, H. E., \u0026amp; Gordon, D. (2020). \u003cstrong\u003e\u003cem\u003e \u003c/em\u003e\u003c/strong\u003eThe Importance of Reliability and Construct Validity in Multidimensional Poverty Measurement: An Illustration Using the Multidimensional Poverty Index for Latin America (MPI-LA)\u003cstrong\u003e\u003cem\u003e. \u003c/em\u003e\u003c/strong\u003e\u003cem\u003eThe Journal of Development Studies\u003c/em\u003e, \u003cem\u003e56\u003c/em\u003e(9). https://doi.org/10.1080/00220388.2019.1663176\u003c/li\u003e\n\u003cli\u003eNtsalaze, L., \u0026amp; Ikhide, S. (2018). Rethinking Dimensions: The South African Multidimensional Poverty Index. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cem\u003e135\u003c/em\u003e, 195\u0026ndash;213. https://doi.org/10.1007/s11205-016-1473-4 \u003c/li\u003e\n\u003cli\u003ePoggiese, M., \u0026amp; Iba\u0026ntilde;ez Mart\u0026iacute;n, M. M. (2024). Privaciones multidimensionales: El origen de la pobreza y la exclusi\u0026oacute;n social. Un estudio para Argentina. \u003cem\u003eVisi\u0026oacute;n de futuro\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(1), 36-60. https://dx.doi.org/https://doi.org/10.36995/j.visiondefuturo.2023.28.01.002.es \u003c/li\u003e\n\u003cli\u003eRuiz, M. A., Pardo, A., \u0026amp; San Mart\u0026iacute;n, R. (2010). Modelos de Ecuaciones Estructurales. \u003cem\u003ePapeles del Psic\u0026oacute;logo\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e(1), 34-45. http://www.redalyc.org/articulo.oa?id=77812441004 \u003c/li\u003e\n\u003cli\u003eSalecker, L., Ahmadov, A. K. \u0026amp; Karimli, L. (2020). Contrasting Monetary and Multidimensional Poverty Measures in a Low‑Income Sub‑Saharan African Country. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cem\u003e151\u003c/em\u003e, 547\u0026ndash;574. https://doi.org/10.1007/s11205-020-02382-z \u003c/li\u003e\n\u003cli\u003eSalvia, A., Bonfiglio, J. I., \u0026amp; Vera, J. (2017). \u003cem\u003eLa pobreza multidimensional en la argentina urbana 2010-2016. Un ejercicio de aplicaci\u0026oacute;n de los m\u0026eacute;todos OPHI y CONEVAL al caso argentino \u003c/em\u003e(Online). Observatorio de la Deuda Social Argentina, Bar\u0026oacute;metro de la Deuda Social Argentina, Serie del Bicentenario (2010-2016). https://repositorio.uca.edu.ar/handle/123456789/8206 \u003c/li\u003e\n\u003cli\u003eSantos, M. E. (2014). El \u0026iacute;ndice multidimensional y trampas de pobreza en el Cono Sur. \u003cem\u003eRevista Problemas del Desarrollo\u003c/em\u003e, \u003cem\u003e178\u003c/em\u003e(45), 89\u0026ndash;112. https://doi.org/10.1016/S0301-7036(14)70877-6 \u003c/li\u003e\n\u003cli\u003eSantos, M. E., Villatoro, P., Mancero, X., \u0026amp; Gerstenfeld, P. (2015). \u003cem\u003eA \u003c/em\u003e\u003cem\u003emultidimensional \u003c/em\u003e\u003cem\u003epoverty index for Latin America\u003c/em\u003e. OPHI Working Paper, 79. https://www.ophi.org.uk/wp-content/uploads/OPHIWP079.pdf \u003c/li\u003e\n\u003cli\u003eSen, A. (1984). \u003cem\u003eRights and capabilities\u003c/em\u003e. Basil Blackwell.\u003c/li\u003e\n\u003cli\u003e\u0026mdash;\u0026mdash; (1985). \u003cem\u003eCommodities and Capabilities\u003c/em\u003e. Elsevier.\u003c/li\u003e\n\u003cli\u003e\u0026mdash;\u0026mdash; (1992). \u003cem\u003eInequality reexamined\u003c/em\u003e. Oxford University Press.\u003c/li\u003e\n\u003cli\u003e\u0026mdash;\u0026mdash; (2000). Chapter 1 Social justice and the distribution of income. In A. B. Atkinson \u0026amp; F. Bourguignon (Eds.), \u003cem\u003eHandbook of Income Distribution\u003c/em\u003e (pp. 59-85). Elsevier.\u003c/li\u003e\n\u003cli\u003eSione, C. A. (2024). La medici\u0026oacute;n multidimensional de la pobreza en Argentina: propuesta metodol\u0026oacute;gica. \u003cem\u003eCiencia, Docencia Y Tecnolog\u0026iacute;a\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(70), 1-43. https://doi.org/10.33255/3570/1642 \u003c/li\u003e\n\u003cli\u003eSkrondal, A., \u0026amp; Rabe-Hesketh, S. (2004). \u003cem\u003eGeneralized Latent Variable Modeling. Multilevel, Longitudinal, and Structural Equation Models\u003c/em\u003e. Chapman \u0026amp; Hall/CRC.\u003c/li\u003e\n\u003cli\u003eUnited Nations Development Programme \u0026amp; Oxford Poverty and Human Development Initiative. (2019). \u003cem\u003eHow to Build a National Multidimensional Poverty Index (MPI): Using the MPI to inform the SDGs\u003c/em\u003e. https://www.undp.org/publications/how-build-national-multidimensional-poverty-index \u003c/li\u003e\n\u003cli\u003eWagle, U. (2009). \u003cem\u003eMultidimensional Poverty Measurement: Concepts and Applications.\u003c/em\u003e Springer.\u003c/li\u003e\n\u003cli\u003eWalker, R (2015). \u003cem\u003eMultidimensional Poverty\u003c/em\u003e. GSDRC Professional Development Reading Pack No. 22, University of Birmingham. https://gsdrc.org/wp-content/uploads/2015/10/Multidimensional-Poverty_RP.pdf \u003c/li\u003e\n\u003cli\u003eWhelan, B. J. (1993a). Non-monetary Indicators of Poverty. In J. Berghman \u0026amp; B. Cantillon (Eds.), \u003cem\u003eThe European face of social security: Essays in honour of Herman Deleeck \u003c/em\u003e(pp. 24-42). Avebury.\u003c/li\u003e\n\u003cli\u003eWhelan, C. T. (1993b). The role of social support in mediating the psychological consequences of economic stress. \u003cem\u003eSociology of Health and Illness\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(1), 86\u0026ndash;101. https://doi.org/10.1111/1467-9566.ep11343797\u003c/li\u003e\n\u003cli\u003eZhang, Y., \u0026amp; Huai, J. (2023). A Case Study of Farmers\u0026rsquo; Behavioral Motivation Mechanisms to Crack the Fractal Multidimensional Relative Poverty Trap in Shaanxi, China. \u003cem\u003eAgriculture\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(11), 20-43. https://doi.org/10.3390/agriculture13112043 \u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e This is the classic indicator of monetary poverty (unidimensional) in Argentina. The total basic basket (TBB) is the value of a set of goods and services for basic consumption. Households whose total family income (TFI) is not enough to cover the TBB are considered poor.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"multidimensional poverty, monetary poverty, generalized structural equation modeling, Argentina","lastPublishedDoi":"10.21203/rs.3.rs-4736007/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4736007/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eA question of great interest for the design of multidimensional poverty measures is whether they should include a monetary poverty indicator. One of the most common arguments for keeping income poverty separate from non-monetary poverty is that they reflect different dimensions of the phenomenon. This paper explores the multidimensional structure of poverty in Argentina and investigate whether monetary poverty should be considered as another indicator of multidimensional poverty using generalized structural equation modeling (GSEM) methods. Using categorical data from a household survey, a generalized confirmatory factor analysis model (GCFA) and a GSEM with a second order factor are analyzed. The GCFA model postulates the traditional assumption that monetary poverty is just another dimension of poverty, while the GSEM supports the hypothesis that monetary poverty is a cause of non-monetary poverty. The results show that the data fit well in both cases, but that it is more plausible to consider that the non-monetary factors are indicators of a higher order dimension and that this non-monetary poverty, as a whole, is explained by monetary poverty. Finally, the implications of these results for the design of multidimensional poverty indicators in Argentina are discussed.\u003c/p\u003e\n\u003cp\u003eJEL Classification: C38, I32\u003c/p\u003e","manuscriptTitle":"Understanding the multidimensional structure of poverty in Argentine households","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-17 01:00:27","doi":"10.21203/rs.3.rs-4736007/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4800a230-fb90-4406-8fc3-59d4dc854924","owner":[],"postedDate":"August 17th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-11-08T13:38:57+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-17 01:00:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4736007","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4736007","identity":"rs-4736007","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}