By Sophie Mitra, January 24, 2018
Data based on the Washington Group (WG) questions can be used in different ways to get at the diversity among persons who report a range of difficulties in functioning. The difficulties can be conceptualized and measured on a continuum or spectrum of severity from ‘no difficulty’ to ‘unable to do’. In my work, it became apparent that it was important to explore and use this continuum.
The WG questions collect data on severity with a four point answer scale for each of the six domains: ‘1-no difficulty’; ‘2-some difficulty’; ‘3-a lot of difficulty’; and ‘4-unable to do’. To identify persons with disabilities, a threshold needs to bet set, which represents a social judgment to differentiate persons with and without disabilities. The WG recommendation for international reporting is to use “a lot of difficulty” as a threshold: persons who report “a lot of difficulty” or “unable to do” for at least one domain are considered to have a disability. Others are considered as not having a disability. Other cut points can be used for other purposes and, as will be shown, it is possible and useful to take fuller advantage of the information collected by the four-point answer scale.
In a recent study I initially planned to use exclusively the dichotomy disability vs no disability as per the WG recommendations. Because a pattern emerged that persons experiencing ‘some difficulty’ were worse off than persons with ‘no difficulty’ for several wellbeing dimensions, I ended up using two thresholds and an index based on the Washington Group questions to try to capture the continuum of difficulties and assess if it is correlated with deprivations.
First, using two thresholds, individuals were grouped into three mutually exclusive categories of difficulties:
- no difficulty in the six domains;
- ‘some difficulty’ in one domain (or more) and no difficulty in other domains;
- ‘a lot of difficulty’ or ‘unable to do’ in one domain (or more).
(2) and (3) were labeled as having moderate and severe difficulties respectively.
Second, the study also relied on a score, which is the normalized Sum of answers (each ranging from 1 to 4) to the six questions with a minimum of six (MinScore) and a maximum of 24 (MaxScore) as follows:
For example, if someone answers ‘1- no difficulty’ to the WG questions except ‘4- unable to do’ for seeing, then the sum of answers is nine and the functional score is:
The functional score has a minimum of 0 and a maximum of 1, and many possible values in between. For the household level analysis, the household functional score is the highest individual score among the adults in a household. With this score, every person or household is placed on a continuum.
In Ethiopia, Malawi, Tanzania and Uganda, the study above found that individuals experienced both moderate and severe difficulties, Table 1 [pdf]. Moderate difficulties are a lot more common than severe ones. For instance, in Ethiopia, 12.85% of adults aged 15 or older experienced difficulties of any degree, with 3.46% experiencing severe difficulties.
In addition, the study found that there is a severity gradient in the association between socio-economic status and disability. Figure 1 [pdf] shows that the household functional score is consistently higher in the poorest asset quintile (quintile 1) than in the richest (quintile 5).
Finally, the association between difficulties and deprivations was found for both moderate and severe difficulties, although it was typically larger and more often significant for the latter. It is illustrated in the spider chart in Figure 2, which gives the deprivation rates for persons with moderate, severe or no difficulty for five dimensions of wellbeing (less than primary school, sick or injured, not working, economically insecure, materially deprived). The three lines, from dark to light green, connect the deprivation rates for persons with severe, moderate or no difficulty respectively. Perhaps unsurprisingly, the darker lines are often on the outskirts of the lighter line for each country, showing higher deprivation rates for persons with moderate and severe difficulties. The gap between persons with moderate and no difficulty is smaller than the gap between persons with severe and no difficulty.
Figure 2: Rates of deprivations by dimension across severe, moderate, no difficulty status
All in all, results from this study using WG data for Ethiopia, Malawi, Tanzania and Uganda suggest that researchers and data analysts should consider using more than one threshold or a score for disability. For some analyses, the finer resolution that this provides may give additional useful insights.
- Mitra, S. (2018) Disability, Health and Human Development. – Open access e-book published by Palgrave Macmillan.
Dr. Sophie Mitra, PhD, is Professor in the Department of Economics at Fordham University with research interests in development economics, disability, health and applied microeconomics.