www.uhasselt.be
DSpace

Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/24865

Title: Cluster Analysis of Ethiopian children (6-60 months of age) living in extreme poverty in Jimma town of Ethiopia: Using Denver-II subscales
Authors: Safari, Wende Clarence
Worku, Berhanu Nigussie
Abessa, Teklu Gemechu
Bruckers, Liesbeth
Granitzer, Marita
Issue Date: 2017
Citation: ISCI 2017: 6th Conference of the International Society for Child Indicators, Montreal - Quebec, Canada, 28-30/06/2017
Abstract: Introduction: Child development consists of several interdependent domains, including language (ability to use and understand language), sensory-motor (ability to use small and large muscles), cognitive (thinking skills), and social-emotional (ability to relate to other people). During the first five years of life, children lay the groundwork for lifelong development. Thus, it is vital to assess children during this vulnerable period in order to determine their status of development and setup development interventions in case they do not optimally develop. Children who live in poverty are more exposed to multiple risks, including; malnutrition, poor health, and unstimulating home environments, which are likely to affect their development. The goals of the paper are, first, to investigate if Ethiopian children who live in extreme poverty compromise of a homogeneous group in terms of developmental profile. And the second goal is to determine what factors predict a subgroup a child belongs to, in case of non-homogeneity (heterogeneity) in developmental profile. Methods: Children were assessed using the Denver-II sub-scales (personal-social, language, fine motor and gross motor performance ratios), which were adopted in the local cultural context of Jimma, Ethiopia. The Latent Class Cluster Analysis was used to investigate population heterogeneity by utilizing finite mixture multivariate normal densities. The obtained subgroups of heterogeneity population were labelled as latent classes/clusters. The Bayesian Information Criterion (BIC) was used to compare multiple models and identify the optimum number of clusters. Results: Of the 819 children studied, three clusters of children were identified and labelled as Delayed Developmental group, Questionable Developmental group and Normal Developmental group. Compared to the Normal group, children in the Delayed group had smaller values of personal-social, language and gross motor performance ratios. Moreover, children in this group had a significant relationship with stunting and a negative relationship with age. The Questionable Developmental group showed delay in language and gross motor development. Further, younger and stunted children were more likely to be in this group, compared to the normal group. In contrast, the Normal group showed normal development based on all domains of Denver-II test. Older and non-stunting children are more likely to be in this group. Conclusion, Policy Implications and Recommendation: The latent class cluster analysis provided an effective method on finding three clusters in the data based on Denver-II sub-scales. Findings suggest that children of younger age and stunting are more likely to be in the Delayed and Questionable Developmental groups as compared to the Normal group. Furthermore, children with poor performance in terms of personal social, gross motor and language are also more likely to be in that group. If the problem won’t be addressed especially in extreme poverty areas, the effects of the problem will affect tomorrow’s workforce. Therefore, it is important to identify developmental delays early so that interventions can minimize the effects of the problem. There should be strong emphases in the children’s health policies and interventions during early childhood development. Moreover, policies should focus more on the improved nutrition’s and be implemented especially in extreme poverty areas.
URI: http://hdl.handle.net/1942/24865
Category: C2
Type: Conference Material
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
N/A246.61 kBMicrosoft Powerpoint

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.