Merge pull request #265 from OCHA-DAP/poverty_rate_changes

HDXDSYS-1384 Update HAPI poverty rate docs with info about trends falling back on mpi data

docs/data_usage_guides/affected_people.md

@@ -177,4 +177,4 @@ For available query parameters, please see the
 * The PIN should **not** be summed across sectors or population statuses,
   as the same people can be present across multiple groups
 * For the number of people affected across all
-  sectors, please use the PIN value where sector=Intersectoral.
+  sectors, please use the PIN value where sector=Intersectoral

docs/data_usage_guides/food_security_and_nutrition.md

@@ -31,10 +31,15 @@ For available query parameters, please see the
   and applying the algorithm from
   [`hdx-python-country`](https://hdx-python-country.readthedocs.io/en/latest/),
   which uses phonetic name matching and manual overrides
-* Any rows whose p-codes are unmatched are p-coded at the national level.
+* Where admin 1 names could not be p-coded, the provided p-codes from the 
+  source data are at national level
+* Where admin 2 names could not be p-coded, the provided p-codes from the 
+  source data are at admin 1 level if possible or national level if not
 
 ### Usage Notes
 
+* The data is available at national, admin 1 and admin 2 with admin names 
+  supplied in the returned data along with p-codes where available
 * The total population (`ipc_phase`="all") is not necessarily equal to the sum of
   the populations in phases 1-5. The differences are usually small (due to
   rounding errors), or because there is no IPC phase data

docs/data_usage_guides/population_and_socio-economy.md

@@ -47,9 +47,11 @@ The global [Oxford Multidimensional Poverty Index](https://ophi.org.uk/global-mp
 (MPI) measures multidimensional poverty in over 100 developing countries,
 using internationally comparable datasets. The MPI assesses poverty through
 three main dimensions: health, education, and living standards, each of which
-is represented by specific indicators. Please see the
-[OPHI methodological note](https://ophi.org.uk/publications/MN-54) for more
-details.
+is represented by specific indicators. For each country, MPI trends over time 
+are supplied if available. Relevant OPHI methodological notes are 
+[58](https://ophi.org.uk/publications/MN-58), 
+[59](https://ophi.org.uk/publications/MN-59) and 
+[60](https://ophi.org.uk/publications/MN-60).
 
 ### Summary
 
@@ -73,6 +75,14 @@ For available query parameters, please see the
 
 ### Usage Notes
 
-* The data is disaggregated to admin 1, but not p-coded. We have kept the
-  admin 1 names in the data, but link only to national level p-codes.
-  We plan to p-code this data in a future release.
+* The data is available at admin 0 and admin 1 with admin names supplied in the 
+  returned data
+* We use p-codes from the source data which was p-coded by taking the admin 1 
+  names, and applying the algorithm from [`hdx-python-country`](https://hdx-python-country.readthedocs.io/en/latest/)
+* Where admin 1 names could not be p-coded, the provided p-codes from the 
+  source data are at national level
+* Trends are estimated using indicators in the global MPI that are harmonised 
+  across the time periods and are used where data are available for a country
+* For any country where trends are unavailable in the source, the latest data
+  (which is not harmonised across time) are used instead
+