Multivariate data has been routinely used in the form of feature vectors for identifying/determining, assessing, and analyzing
child malnutrition. Using machine learning techniques on such data for eventually deriving social health determinants has been
a recent methodology used in concerned research communities. However, data visualization has been grossly underused in this
regard, for studying trends and further analysis of the data. Visualization of multivariate data in identifying child malnutrition
enables exploration of the data, and summarizes its multidimensional analysis. Even though the multivariate data used for data
analytics across various studies, comprise of common variables/parameters predominantly, there are slight variations depending
on the purpose of the research study and the control population used. Here, we enumerate all the parameters used in these studies
into an exhaustive list of variables, and propose using this list as anunified feature vector, which can be used for holistic analysis
of the problem of malnutrition. Thus, in this work, we survey existing literature to find and propose the following: (a) a unified
feature vector, which is a subject-wise data object, derived across several research studies in identifying, assessing, and analyzing
child malnutrition, and (b) in the existence of dataset/collection of unified feature vectors across a population, the visualization
tasks that can be performed on it, in lines of the five W’s of knowledge discovery in journalistic reporting, and (c) the appropriate
set of visualization techniques to accomplish those tasks. We further propose the design of the visual analytic framework for such
an analysis of dataset.