Definitions matter – part two

How is it possible to monitor progress, or otherwise, towards the UN’s Sustainable Development Goals (SDGs) for water and sanitation without clear definitions?

David Satterthwaite's picture
David Satterthwaite is senior associate with IIED's Human Settlements research group
08 December 2021
The transition to a predominantly urban world
A series of blogs and interviews exploring what social, political and environmental factors cause cities to thrive or decline
A man lines up six bright water bottles under a series of outside taps

Filling bottles with water in Sanaía, Yemen (Photo: Foad Al Harazi/World Bank, via Flickr, CC BY-NC-ND 2.0)

This continues the discussion from a previous blog on definitions that matter because they can misrepresent, distort or obscure key issues. Among examples given were definitions for poverty lines and toilet standards that were set far too low.

Added to this is insufficient official data in many nations, including a lack of censuses or inaccessibility of their data. And for many issues and nations, the data needed for instance whether available water supplies are safe/free from contamination is not collected.

UNICEF and the World Health Organization sought to address these deficiencies through a Joint Monitoring Program (JMP) that collects, classifies and analyses existing data on water and sanitation.

It seeks to accurately monitor progress towards WASH (water, sanitation and hygiene) related Sustainable Development Goals (SDGs)  adding hygiene and menstrual health to its concern for water and sanitation. This blog considers the data and definitions used by the JMP in its 2021 report on progress on household drinking water, sanitation and hygiene 2000-20.

This is the latest in a series of blogs and interviews, curated by IIED senior fellow David Satterthwaite, examining different aspects of global urban change.

The scale of the problem

Billions of urban dwellers lack good quality water and sanitation in their homes. In many nations, deficits are still growing. But this fact can be hidden in the definitions used. According to what the UN categorises as ‘improved’ water, only 2% of the world’s urban population lacks this provision.

Even in low-income and the least developed nations, 95% of their urban dwellers have ‘improved’ provision. But the definition of ‘improved’ sets the bar so low that its use understates the problem. For instance, improved provision for water does not necessarily mean a supply that is free from faecal contamination or accessible, regular or affordable.

The definitions and data dilemma

The JMP is very careful about definitions. It has done much to improve the quality and range of available statistics on water and sanitation. But it has to rely on data provided (or not) by national governments.

It has managed to disaggregate this data, so it can report on progress for nine different kinds of provision for water that are within what it terms ‘improved’, and six different kinds of ‘improved sanitation’ facilities. For each of these, there is a clear definition.

But the JMP then combines them in aggregated statistics that underplay the scale of deficits because the definitions are so broad.

‘Improved’ water provision includes:

  • Piped supplies: tap water in the dwelling, yard or plot (including piped to a neighbour), public taps or standpipes, and
  • Non-piped supplies: boreholes/tube wells, protected wells and springs, rainwater, packaged (including bottled and sachet), delivered (including tanker trucks and small carts/tanks/drums), and water kiosks.

Unimproved water provision is surface water.

The tables below show how the proportion of the urban population without provision for water decreases dramatically if ‘improved’ provision means piped. For instance, 83% of Nigeria’s urban population has ‘improved and not piped’ and for India the figure is 31%.

‘Improved’ sanitation includes: flush or pour-flush toilets or latrines connected to sewers, septic tanks or pits; ventilated improved pit latrines (VIPs); pit latrines with slabs; or composting toilets.

The next tables show dramatic differences between urban populations with ‘improved’ provision (98%) and with sewer connections (64%).

For the least developed nations, 76% of their urban population had ‘improved’ provision but only 11% had sewer connections. India claimed that 99% of its urban population had ‘improved’ provision but only 34% had sewer connections.

Of course, sewers are not the only system that can provide high quality sanitation – and cities often lack sewage treatment plants – but it is a telling figure.

Local context is key

Whether or not the different kinds of ‘improved’ provision deliver on SDG 6 depends so much on local contexts. Much of the WASH urban deficit is concentrated in the large, dense informal settlements where there is a lack of space for pit latrines and other on-site systems – as is also the case for multi-storey tenement dwellings.

This and earlier JMP reports stress how WASH conditions are usually worse in rural areas – but no allowance is made for differences in context. Some of the forms of ‘improved’ provision such as pit latrines with slabs may work well in rural areas but not in most large, dense urban settlements.

For low-income nations, the proportion of their urban population provided with sanitation can be:

  • 46% (based on what the UN categorises as basic)
  • 20% (based on what the UN categorises as safely managed), and
  • 15% (having a toilet connected to a sewer).

And for water, the proportions can be:

  • 94% (improved)
  • 67% (improved and piped), and
  • 58% (safely managed).

Shift from ‘improved’ to ‘safely managed’

Recognising that 'improved’ provision for sanitation had given too little attention to the management of excreta – for instance managing faecal sludge from pit latrines – the JMP established a new category ‘safely managed.’

This is ‘improved’ provision where in addition excreta are safely disposed of in situ or removed and treated off-site, and where wastewater is also treated. The JMP also divided ‘improved’ provision into basic (not shared) and limited (shared).

Safely managed drinking water was also added. This is water from an improved source that is also accessible on premises, available when needed and free from faecal and priority chemical contamination.

‘Safely managed’ water and sanitation are more accurate measures for assessing progress towards the sixth SDG.

No easy task…

The JMP has a difficult task. It is called on to provide relevant statistics to monitor national governments’ progress towards the SDGs. But it has to rely mostly on data provided by censuses and national sample surveys.

It would be easy to produce a more comprehensive list of indicators that would include better coverage of health and affordability issues, but it is constrained by the limitations in the data that national governments provide.

There is also the gap between ambitious official goals, including the SDGs and the data available for measuring these. For instance, the 2030 Agenda states that "SDG indicators should be disaggregated where relevant by income, sex, age, race, ethnicity, migratory status, disability and geographic location or other characteristics". This is far beyond what is possible with existing data in most nations.

Finally, there is the fact that city governments have major WASH roles and responsibilities. Part of the commitment to more disaggregated data should include figures for each city and city district, recognising that these may be limited as much government data is from household surveys with sample sizes too small to report on individual cities.

One valuable role for the JMP would be to see how much WASH data is available for existing cities and city districts. For nations with recent censuses, census data should be able to provide some of this.

About the author

David Satterthwaite ( is senior associate with IIED's Human Settlements research group

David Satterthwaite's picture