selecting indicators of healthy early childhood development

Home Forums Measurement: Assessments and Measures selecting indicators of healthy early childhood development

This topic contains 2 replies, has 3 voices, and was last updated by  Elizabeth Prado 3 years, 11 months ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #938

    On the Al Jazeera TV news this evening (9 September), Dr Priscilla Atwani Idele, a statistics expert at UNICEF HQ in New York, with considerable field experience in sub-Saharan Africa, was interviewed on the remarkable reduction of infant mortality around the world since 1990. Over the past 25 years the rate has been reduced by more than 50%.

    Asked what are the main action priorities for maintaining that momentum and reducing still further the terribly high rate of infant mortality in Sub-Saharan Africa, she replied first: to maintain and strengthen the assault on the known major causes of infant death in the region, pneumonia, malaria, diarrhea and perinatal complications of delivery. Then she said that no matter how well medical technology progresses it will never make as much of an impact as changing behavior in communities. And finally she added that one of the international priorities should be investment in good data, since it is attention to data that has enabled effective monitoring of infant health and strategic targeting of key factors for intervention.

    What came to mind, as I reflected on those wise and carefully chosen remarks, was that she was making a strong case for the value of the enterprise that I was privileged to take a peep at in Cape Town last week. I attended the 5th Conference of the International Society for Child Indicators, and commented on the papers presented at a symposium entitled “Meeting the challenge of cultural, linguistic and socio-economic diversity in the measurement of developmental domains during early childhood.” Papers were presented on research under way or completed in “culturally variant, low and middle income settings of the Global South”, including India, East Asia, Polynesia, Latin America and many different countries in Africa.

    The focus of this diverse, multinational collection of papers centered on a question posed by the Convenor, Andy Dawes of UCT and HSRC South Africa: “Are internationally appropriate measures of children’s core competencies– essential developmental skills that may manifest in particular culturally embedded ways – feasible or realistic?”

    Given the importance of community change in effecting reductions in child mortality, and indeed in effecting enhancements in the quality of life for children, I think it is important to unpack Andy’s challenge. Embedded in his question is the supposition that what is essential in one particular cultural context is equally essential in another. If so, the challenges are to

    i.find a way of defining a skill in a cross-culturally equivalent way

    ii.Establish a reliable and valid way of measuring that skill in a fair and acceptable way across different contexts.

    And that is where much of the effort of applied research seems to be getting invested in this field.

    But, as Penny Holding of the Aga Khan University in Kenya, noted, indicators represent constructs, and constructs are selective representations of conceptual domains. And, as Sally Brinkman of the Fraser Mustard Centre in Adelaide, Australia, illustrated in detail with her account of research in Tonga, according to local expert informants the “basic building blocks of being a Tongan” include: “respect”, “dignity”, “appropriate obedience”, “knowing your place”, personal attributes that are also widely valued in agrarian African societies, but accorded little attention in Western social and psychological theories of child development.

    In an impressive ongoing project in South Africa, the ELOM (Early Learning Outcomes Measure) is to be empirically validated in evidence of the (descriptive) norms of performance by a representative sample of children in various ses, “deep rural” and language communities. But as Linda Biersteker, co-PI of the project with Andy Dawes, emphasised, the validity of the measure will ultimately depend on how well it matches what children SHOULD know and be able to do in various domains (i.e. prescriptive norms). So a key issue that applied researchers in this field need to address is: what is the relation between the two meanings of “normative” (descriptive and prescriptive) ?

    I wonder if others on this network would like to share ideas on this issue ?


    Robert Serpell PhD
    Professor of Psychology, University of Zambia


    Hello Dr. Serpell,

    I think large scale retrospective studies of adults might prove useful towards the goal of cross-culturally understanding these core attributes. They will allow you to not only ask the adults what was determined as important values to hold in their childhood, but also what it means to them today (taking advantage of their maturity and possible utility of what they’ve learnt). We will also then be able to learn who enforced which values – were certain ‘softer’ values imposed more by the mother (to be stereotypical for a minute)?

    Once the attributes are gained in quantity, they can be qualitatively and linguistically (through words and metaphor) compared cross-culturally for what meaning and importance they ascribe to these attributes.

    Going further in this direction might help toe the line between universal and culturally appropriate attributes.

    I hope this is in line with the kind of discussion you were hoping to have.

    Nivida Chandra
    Doctoral Scholar (emotional neglect in children)
    Indian Institute of Technology Delhi, India


    Elizabeth Prado

    The distinction between descriptive norms and prescriptive norms is very helpful. Traditionally, what children “should” be able to do has been determined by norming studies, where the norm sample is a representative sample of a healthy population. In LMICs, it can be difficult to select a sample that is both healthy and representative. Are stunted children healthy? What about anemic children? What if 70% of a population of children is anemic and/or stunted? Should they all be excluded from the norm sample? In this case, it seems impossible to have a sample that is both healthy and representative.

    The reason we want to know what children “should” be able to do is that we want to know if a child or a group of children is behind where they should be, because we think that will matter for their future. We know that there is a wide range of variance in what normal, healthy development looks like. Between two 5-year-old children who have equal cognitive and language scores, one might have said their first word at 12 months and the other at 20 months. At an individual level, this type of variance doesn’t seem to “matter” for future outcomes.

    At a population level, these types of differences are more likely to “matter.” If an entire population said their first word, on average, 3 months later than another population, I would also expect to find a difference in language and cognitive scores at age 5 years. However, these two populations would probably differ on many many factors. We need well-controlled longitudinal studies to answer these types of questions. I think studies of predictive validity may be a good way to develop prescriptive norms. What “should” children be able to do in a way that matters for later outcomes?

Viewing 3 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.