1. In the language of statistics, uncertainty is a latent variable*(N. Bloom) …particularly when it is mainly statistics that lead to policy-induced uncertainty. In policy making, it too often boils down to ‘2b or not 2b’.** (M. McGrath) One can ask: Where then are human rights (HR) policy considerations relegated-to?
*: We walk many roads, all paved with uncertainty; so, choose: But do so wisely as you walk (JeromeKoenig) and be aware that the omnipresent ‘omitted variable bias’ is at the very center of much empirical social sciences research. The fact is that nothing plus nothing does not give nothing, but sometimes gives a little something. (Julio Cortazar)
**: Many researchers who have been prominent in identifying a certain problem say that getting involved in the solution of the same is none of their business. Deplorable. (FernandoMonckeberg)
2. You have to agree with me thatthe ‘p values’ of biostatistics are no substitute for human values, especially HR values.(Jonathan Mann) Why?, because wisdom is being reduced to knowledge which in turn has become to mean reducing knowledge to information*** –or, as we are now supposed to say, data. I posit that statistics are deliberately positioned as objective by the forces of status-quo, i.e., in the sense that statistics are ‘free from context’. But such a position is utterly reductionistand even naive. (GeoffreyCannon)
***: Where is the word we lost in words and where is the knowledge we lost in information. (T. S. Elliot)
3. On their own, statistical indicators are inconclusive. They say nothing without clear reference points against which to judge performance and to assess the adequacy of achievements or progress over time …and the reference point, in our case, I dare say never is human rights.
4. Together with others, I am especially weary of the prevailing tendency to ‘fetishize’ particular techniques, especially the quantitative techniques that are so tempting to many. I see in them a number of unintended risks and consequences. There is, for instance, a risk that ‘perfecting’ the particular tool or technique becomes an end unto itself –the danger being that the tool becomes overly complicated and inaccessible to the intended user.Another risk of fetishizing quantitative tools and techniques is that these tools can and do narrow the lens of analysis, reducing a complex reality to simple, verifiable numbers, and thereby making invisible otherwise relevant factors. As a result, ‘the multi-dimensional can be easily confused with the two-dimensional’. So, beware of the risks of reducing assessments to a technocratic exercise that analyzes the trees while missing the forest –and masks the value judgments that are inherent to choosing particular indicators and collecting specific data –HR always being the looser here. (CESR’s OPERA approach)
5. But it does get worse: Observations derived from correlating statistical data are all too often mistaken for causal relationships. The effects of single factors are easier to pinpoint and trace than the interaction of several factors, yet these interactions are often the key that unlocks outcomes. Selecting manageable indicators to capture concepts such as empowerment, HR violations or greater democratic engagement is a struggle that HR workers often lose (the preparation of the post-2015 development agenda being only one current example).**** One key factor here is the capability of citizens and their organizations to influence the information to be gathered and making it available in a transparent manner. The latter requires the mobilization of rather vast masses of citizens who must become involved in some type of HR learning.
****: In the context of HR, we should keep in mind three types of indicators: (i) structural indicatorsthat capture the acceptance, intent and commitment of states to undertake measures in keeping with their HR obligations; (ii) process indicatorsthat assess states’ efforts, through its implementation of policy measures and programs of action, to transform its HR commitments into desired results and (iii) outcome indicatorsthat assess the result of states efforts in furthering the fulfillment of HR.(hrindicators@ohchr.org ; http://www2.ohchr.org/english/issues/indicators/index.html)
6. The rather mechanical correlation of data –and the use of many other old and new statistical techniques– fits the current obsession with counting. Numbers may be fine as far as they go, but they can neither explain our behavior nor how we experience the motivating power in our search for HR and dignity for all. In the words of the note that Albert Einstein kept on his wall: “Not everything that counts can be measured, not everything that can be measured counts”. (The Broker, Issue 25, June 2011)
7. In general, development practitioners, as it is now conventionally posited, seem to assume that the inductive method, which is to say the accumulation and ordering of facts, i.e., data, into evidence, using increasingly sophisticated statistical techniques, will of itself generate objective findings. This is just plain wrong. For a start, the very ordering of facts, the equivalent of brick-laying, implies a plan, an idea. But in any case, the purpose of evidence is to support or refute a theory. The theory comes first. This is the deductive method. True, the theory may emerge or be refined in the course of the organization of facts, just as a master mason may change the architect’s mind. Induction is comfortable. It implies that information will generate objective measured quantified conclusions: Job done! These days the inductive method is mostly worked out by computers and is becoming increasingly detached from thought. By contrast, deductive processes require constant thought. By their conformist, middle-of-the-road nature, old theories and indeed stale ideologies and systems of ideas can be and indeed should be challenged, as needs arise and, in our case, HR circumstances require. This may feel dangerous to some ‘with an axe to grind’. Indeed, but this is the way of the real world and we see no more room for HR ideas to be resisted. (G. Cannon)
8. Bottom line: We should be careful to avoid interpreting the idea of a ‘data revolution’ too narrowly, i.e., as only being about statistics. Data, as defined by the Oxford Dictionary, is “facts and statistics collected together for reference or analysis.” But facts, especially HR facts, can also come from testimonials, videos, story-telling and other participatory research approaches. Furthermore, these methods can give an important and often critical context to an issue far beyond what the numbers alone may or may not reveal. Numbers are useful, but only as long as they do not master us.
9. If we are truly serious about effective means of monitoring progress on the next set of post 2015 development goals, these approaches must be part of the ‘revolution’.
10. Keep in mind that, in an effort to bridge the digital divide between the knows and the know-nots(AnwarFazal), innovations in technology can assist greatly (e.g., SMS-based polling, low cost video capturing) alongside more traditional, ‘low tech’ solutions (e.g., particularly open town hall meetings). (S. O’Shea) The ultimate effort I guess I am actually calling for is for us to increasingly step out of the exclusive ordinary range of statistics (S. Spender) and complement the use of data with the use of testimonies. [Am I talking of statistimonies or testictics?].
Claudio Schuftan, Ho Chi Minh City
cschuftan@phmovement.org