Can I use tap water here? In many cities across the world visitors even fear brushing their teeth with what comes out of the pipe. People spend huge amounts of money on buying bottled water, paying in some countries a 2000 fold premium even though bottled water may come with safety problems too. On the other hand, according to many official statistics, piped water equals drinking water. Consider the Millenium Development Goals: Their Target 7c is to “Halve, by 2015, the proportion of the population without sustainable access to safe drinking water” and the UN proudly announce that this goal has been achieved already!
Only in the smallprint we learn that the UN do not actually have information about water quality. They recur to a so-called proxy indicator instead, equating drinking water with (estimations about the percentage of households with) access to “improved sources” such as pipe systems. The problem of course is that piped water, especially in developing countries, is often contaminated. Just taking bacterial contamination (one of many possibilities) into account, technical studies suggest that the the current UN calculation “underestimated the progress required to meet the drinking-water component of MDG Target 7c by 10% of the global population” – this represents hundreds of millons of people!
One ought to be aware of the risks of “too aproximate” proxy indicators. Summary statements based on poor equations may mislead decision makers (who will see no need to invest in water quality testing if the UN tells them that improved sources are fine) and the public alike.
In a way, defining the level of approximation is a challenge for virtually all indicators that are prevalent in public policy discussions and the media. This applies to politically more contested topics (is GDP an appropriate measure of economical progress; what do PISA test scores say about the performance of education systems?) but also to quite technical ones. For example, in the case of drinking water the testing for E.coli bacteria is interestingly the application of another proxy since E.coli itself is generally harmless but a good predictor for the presence of other pathogens. The underlying issue for all of this is “quality criteria for indicators” which I feel like writing about in another post!
What I´d like to put up for discussion now are some general thoughts derived from the frankly “scandalous” MDG drinking water proxy indicator:
1) The more global comparisons one wants to make, the more one needs to deal with issues of poor data quality and limited data management capacity. In the case of the MDGs universal scope implies using an indicator that is manegeable by the weakest country. More local indicators can be fine-tuned to local capacities.
2) The more global comparisons one wants to make, the greater the diversity in settings that indicators need to capture and standardise. For example, MDG indicator definitions need to consider any type of conceivable water supply from desalination in the desert to snowmelt in polar regions. This is also worthwhile considering for any type of indicator that is strongly influenced by culture, e.g. subjective wellbeing which is currently much promoted as a global alternative to GDP but difficult to standardise cross-culturally. More local indicators (incl. definitions and methodologies) can be finetuned to local circumstances.
What to take a away? Given the evident benefits from international comparisons and frameworks such as the MDG I woudn´t suggest to stop working on them. To the contrary, there´s so much to learn from international and cross-cultural definitions, “indicator banks” and associated data collection methodologies. On the other hand, it would be foolish to only use poor international proxies and neglect the potential of better, locally more appropriate indicators and monitoring systems, would´t it? In the case of drinking water, plenty of countries and cities can surely do better than just classify types of sources but publish actual quality data so inhabitants and visitors know whether tap water is safe. How to get this going?