TIM'S COMMENTS REGARDING HADI'S PRESENTATION

[Barbara Bugosh <bbugosh@andrew.cmu.edu>]

(i) Hadi discussed what he saw as the desirable properties of an index:
monotonic, orthogonal to other objectives, and fair. These points can be
related to the properties of fundamental objectives for any decision. An
objective that is a fundamental end of importance for a given decision,
rather than a means to another objective, will almost always be monotonic.
Objectives with peaks seem to inevitably be means to other things.
Fundamental objectives will also be non-redundant to others. You can't
guarantee they are orthogonal, but they will at least not overlap. Finally,
as Baruch has shown us, picking measures for anything involves value
judgments. Hence its pretty difficult to guarantee that a measure is
universally seen as fair, since fairness is often defined in contradictory
ways (e.g., treat everyone the same, and help those who need help).


(ii) This is a national level index that is intended for marshalling
national political support or help from international agencies. It is not
relevant as a direct measure to be traded off against other objectives in a
specific policy decision. On the other hand, if the data were available on
a sub-national basis, it could be used as a water/human development
risk-ranking indicator, and serve as a basis for comparing different
regions or even communities within a country.

(iii) Hadi is asserting that human development is equal to avoiding
malnutrition. It is actually a proxy measure for several more fundamental
objectives. It is a good heuristic measure, for political purposes. It
captures a lot. It is not a proxy for water stewardship broadly.

It is a good aggregate proxy for all the following:

-child survival to adulthood
-child quality of life
-equity in food distribution
-equity in drinking water distribution
-institutional (and family) attention to individual child well being


None of these could be seen as water stewardship. It's a reasonable
aggregate  index of "humanity development".

(iv) I would suggest that Hadi prepare an influence diagram that includes
the objectives noted above in (iii) as well as factors such as the natural
endowment of water, food productivity, government structure, public health
initiatives, etc, to characterize the factors that influence the index he
has developed, as well as how the index contributes to other objectives. He
could even estimate these relationships with multivariate statistical
techniques if the appropriate data were at hand.

Regards,
tim