Using sparse categorical principal components to estimate asset indices: new methods with an application to rural southeast asia

Merola, Giovanni Maria; Baulch, Bob

doi:10.1111/rode.12568

Cited by 10 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data on assets were not available in early life for India. The limited availability of asset data also prevented us from inferring if other metrics associated with assets – quantity, quality or functioning, technological generation, availability of substitutes – biased our findings ( Johnston & Abreu, 2016 ; Merola & Baulch, 2018 ). Our sensitivity analysis using data from the Pelotas 1993 (Brazil) cohort suggested that a harmonized asset index created using counts of assets such as televisions, cars and housekeepers as well as number of bathrooms in the house was correlated (r = 0.98) with the benchmark asset index.…”

Section: Discussionmentioning

confidence: 99%

Changes in asset-based wealth across the life course in birth cohorts from five low- and middle-income countries

Varghese

Adair

Patel

et al. 2021

SSM - Population Health

View full text Add to dashboard Cite

Background Temporally-harmonized asset-based measures of wealth can be used to study the association of life-course wealth exposures in the same scale with health outcomes in low- and middle-income countries (LMICs). The within-individual longitudinal stability of asset-based indices of wealth in LMICs is poorly understood. Methods Using data from five birth cohorts from three continents, we developed temporally-harmonized asset indices over the life course through polychoric principal component analysis of a common set of assets collected consistently over time (18 years in Brazil to 50 years in Guatemala). For each cohort, we compared the harmonized index to cross-sectional indices created using more comprehensive asset measures using rank correlations. We evaluated the rank correlation of the harmonized index in early life and adulthood with maternal schooling and own attained schooling, respectively. Results Temporally-harmonized asset indices developed from a consistently-collected set of assets (range: 10 in South Africa to 30 in Philippines) suggested that mean wealth improved over time for all birth cohorts. Cross-sectional indices created separately for each study wave were correlated with the harmonized index for all cohorts (Brazil: r = 0.78 to 0.96; Guatemala: r = 0.81 to 0.95; India: 0.75 to 0.93; Philippines: r = 0.92 to 0.99; South Africa: r = 0.84 to 0.96). Maternal schooling (r = 0.15 to 0.56) and attained schooling (r = 0.23 to 0.53) were positively correlated with the harmonized asset index in childhood and adulthood respectively. Conclusions Temporally-harmonized asset indices displayed coherence with cross-sectional indices as well as construct validity with schooling.

show abstract

Section: Discussionmentioning

confidence: 99%

Changes in asset-based wealth across the life course in birth cohorts from five low- and middle-income countries

Varghese

Adair

Patel

et al. 2021

SSM - Population Health

View full text Add to dashboard Cite

show abstract

“…The positive scoring factor is closely associated with higher socio-economic status and the negative scoring factor explains the lower socio-economic status. The first principal component will provide a measure of wealth (27) . Hence, the first component of the scoring factor of each asset variables is used for assessing socio-economic status of the sample households.…”

Section: Asset Based Scoring Using Principal Component Analysismentioning

confidence: 99%

“…Later, asset index is sorted and quintile cut-off points created to differentiate economic groups and by caste categories (Government of India classification). The approaches used are based on the assumption that socio-economic status is uniformly distributed or data driven (27) . Arbitrary quintile cut-off points are categorized into five groups: lowest 20 percent of the households into 'poorest' , second lowest 20 percent of the households into 'poor' , mid 20 percent of the households into 'middle' , second highest 20 percent of the households into 'rich' and highest 20 percent of the households into 'richest' .…”

Section: Asset Index Of Different Economic Groups and By Caste Categoriesmentioning

confidence: 99%

Measuring the economic status of different caste groups in rural India: An application of principal component analysis

Manaloor¹,

Radhakrishnan²,

Raghu³

et al. 2021

IJED

View full text Add to dashboard Cite

Background/Objectives: To prevent leakages and to efficiently roll out development programs, it is important that policy-makers and researchers are able to identify poor households that need intervention. Such identification will aid in adopting a targeted approach for poverty reduction strategies and improvement of livelihoods. Methods/Statistical analysis: The present study conducted amongst 3845 farm households in three agro-biodiversity hotspot regions in India, uses non-money metrics method to measure household economic status. Principal component analysis is used to build an asset index score of each household and determine the economic grouping based on caste categories. Findings: The results show a high level of inter-regional differentiation among the household economic status and intra-sub caste groups. The overall economic status reveals that Meenangadi falls in the 'rich/richest' group compared to the Kolli Hills in the 'middle' group and Koraput in the 'poorest' group. Our results are also important for household classification based on the social stratification within a regional context. Hence, it is reasonable to state that the asset based method is relevant for fine tuning identification of marginalized communities, notably in the deprivation hotspots, paving the path for their inclusive growth. Novelty/Applications: The present study attempts not only to identify economic status of rural households in different states of India, but also looked at the economic status of different castes, particularly at scheduled tribe groups that are dominant in the study area.

show abstract

“…NLPCA produces a similar output to that of PCA with eigenvalues, component loadings and communalities [9]. Reference [12], also argue the use of NLPCA over other methods stating that the interpretability of the NLPCA solutions are much enhanced compared to MCA solutions as NLPCA produces components that are combinations of the variables instead of categories. Hence, weights for the index can be derived from the NLPCA solution in a similar fashion to that of weights derived using PCA or FA.…”

Section: Introductionmentioning

confidence: 98%

“…In fact, MCA is the oldest application of PCA to categorical variables. However, [12] raise concerns over the interpretability of MCA solutions as they are expressed in terms of combinations of categories for each of the individual variables. This will complicate the interpretability of the MCA results, because MCA yields factorial scores for each level of the categorical variables.…”

Section: Introductionmentioning

confidence: 99%

Improving the Statistical Capacity Index: A Statistical Approach

Dharmaratne¹,

Mdt²

2018

AJAMS

View full text Add to dashboard Cite

Good quality, timely and accurate statistics lie at the heart of a country's effort to improve development effectiveness. As a response to the challenge of measuring the institutional capacity of a country in producing timely and accurate statistics, the World Bank developed its framework for the Statistical Capacity Index (SCI). Although the World Bank's framework is acknowledged for its simplistic approach, it has received extensive critique for the ad-hoc allocation of weights. This research attempts to find a solution to this criticism using a statistical methodology. Country information used by the World Bank to create the SCI for the year 2014 was considered. The data consisted information on 25 categorical variables out of which 16 were binary variables and 9 were ordinal variables. Nonlinear Principal Component Analysis (NLPCA) was conducted on the categorical data to reduce the observed variables to uncorrelated principal components. Consequently, the optimally scaled variables were used as input for factor analysis with principal component extraction. The results of the factor analysis were used to weight the new SCI. The dimension, availability and periodicity of economic and financial indicators explained most of the variance in the data set. The research proposes a simpler version of the new SCI with only 23 variables. In the proposed new index, the variables enrolment reporting to UNESCO, gender equality in education and primary completion indicators were the three variables receiving the largest weight. These three indicators measure the periodicity of reporting data on educational statistics to UNESCO; periodicity of observing the gross enrolment rate of girls to boys in primary and secondary education; and periodicity of observing the PCR indicator which is the number of children reaching the last year of primary school net of repeaters respectively. This research represents the first attempt to create a SCI using multivariate statistical techniques and especially index construction with NLPCA. The research concluded with a comparison of the proposed new index and the index created by the World Bank, which justified that the proposed index be used as a solution for the arbitrary allocation of weights in creating the SCI.

show abstract

Using sparse categorical principal components to estimate asset indices: new methods with an application to rural southeast asia

Cited by 10 publications

References 26 publications

Changes in asset-based wealth across the life course in birth cohorts from five low- and middle-income countries

Changes in asset-based wealth across the life course in birth cohorts from five low- and middle-income countries

Measuring the economic status of different caste groups in rural India: An application of principal component analysis

Improving the Statistical Capacity Index: A Statistical Approach

Contact Info

Product

Resources

About