Achieving the Vision Geo-statistical integration addressing South Africa s Developmental Agenda geospatial + statistics The Data Revolution humble beginnings, present & future - South Africa UN World Data Forum Theme (Innovations and synergies across different data ecosystems) Sub theme (Integrating geospatial, statistical data and other information) Cape Town, South Africa, January 2017 Sharthi Laldaparsad Statistics South Africa Policy Research & Analysis
From paper maps to geo-statistical frames
Statistical geography a key strategic direction for Stats SA Standard Geographic Frame (consistent & stable over time) Geo-statistical Building Block against which information can be collected & rolled up for dissemination Updated & reliable fundamental sampling/ estimation frames: Dwelling/ Address Frames; the Business Frame; the National Population Register Built a dependency on other data suppliers/ producers. We want quality data from them. Dependency on a functioning SASDI. Therefore statistics supports the SASDI development. SDI Act (policies, standards, data, technologies, capacity building, etc.)
Agenda 2063-The Africa We Want The Policy Drivers : Global Need for Spatially-Enabled Complex Information Clear articulation of vision, goals and aspirations. Impetus to align and build capacity. Recognised the need for quality statistical and locational management of data. Need for disaggregated data (people, gender, age, by place; urban/ rural; economic sectors). Place & space very important for vertical and horizontal analysis. 2020 Round of Censuses UN Principles and Recommendations for Population and Housing Censuses, Rev.3, to ensure complete integration of statistical and geospatial information (para. 349, UNSD, 2015)
South Africa s National Development Plan (NDP) - Vision 2030 Evidence-based. Spatial & Statistical information basis for planning, monitoring and evaluation South Africa s Developmental Agenda Clear statement of Government priorities. Alignment of short-term, medium-term & long-term planning Not all data used for planning is produced by Stats SA. Will involve collecting, standardizing, utilizing important information from other sources. Creating accountability. Appropriate legislation. Dependency on a functioning NSS.
The Global Statistical Geospatial Framework (abbreviated as GSGFramework) General comments: High-level framework Not a one-size-fits-all Start anywhere Demand for small geography data. More frequent data. Policy impact - SDGs, Africa 2063, NDP. Census 2020. UN Expert Group on the Integration of Statistical and Geospatial Information Principle 5 (Analysis, policy impact) Data Ecosystem ISO & OGC Standards. Guiding Principle 4 Principles (Technical standards, become data sharing) important Principles of Official Statistics (trust, concepts, classifications, consistent, etc.) wards, etc.). Principles Principle of the 3 (Output GSGFramework geographies ) Changing (SDI, boundaries. standards, classifications, data sharing, etc.) Principle 2 (Unique identifiers) Standard geographic administrative boundaries (province, municipality, electoral No PIN. Link mainly thro the geography. Geography not always standardised. Principle 1 (NSDI, Seamless integration ) SDI Act 2003 Stats Act - Legislative reform More institutional/ political leadership/ support
Policy Analysis = Geo + Stats Identifies constraints. Feeds back into the geospatial and statistical processes. For quality data. Testing of the integration. Will it deliver where are we in the developmental agenda? What are the challenges? What are the priorities?
Census 2011 Population (Census 2011, CS 2016) Population share a measure of primacy CS 2016 Population itself an indicator. Geo(graphy) itself an indicator.
2006 Number of buildings completed How is growth promoted in all regions? Is there a minimal set of core economic activities for regional economic sustenance? 2013 Vulnerability and sustainability of regions. Resilience In both 2006 differs and 2013, from Cape region to Town and Tswhane completed the region. most number of buildings Uneven geographical development. Grow the economy. Create jobs.
2010/11 New VAT Registrations (2010/11, 2014/15) Boundaries blurred 2014/15 Data supplied by South African Revenue Services (SARS) Grow the economy. Create jobs.
Municipal Capital Investment Framework & Spatial Development Frameworks Census 2001 Population Census 2011 Population Metros public capital spending, for 5 years i.e. 2009/10 2013/14
Disaggregating - Where people are? Characteristics of the area they live in? Problems of people, livestock, draught, fires, land degradation,...? Census 2011: Population DU Count Listing Capture 2011 DU Count 2014 Trends in population distribution over space & time City of Jo Burg SAE using census-survey-admin-other data
Data sources: Socio-demographic (Census 2011); Environmental (Dwelling Frame & satellite imagery) Census 2011 Population density %Females %0 to 6 year old %7 to 18 year old %19 to 34 year old %35 to 64 year old %65 plus %No or some schooling %Unemployed %No income %Working in the formal sector %In-migration %Basic services (water, electricity, sanitation, refuse) %Traditional dwellings %Dwelling owned, not paid up %Rented Etc. Dwelling Frame %Residential Dwelling points %Commercial points Methodology: LLR, PCA, GWR, AMOEBA algorithm Population Density was modeled. Classified satellite imagery %Built-up dense settlements %Roads & rail %Natural water & wetlands Mixed data sources Sound methods Source: Aldstadt and Getis, 2006
Stats SA Geography-types LLR Predictions AMOEBA Clusters GWR Predictions
GWR detail analysis Local R 2 Standardized residuals Standardized residuals: Anselin Local Moran s I
Achieving the Vision Geo-statistical integration addressing South Africa s Developmental Agenda Have we leveraged? What are the levers moving forward? Past and present very successful at linking censuses and surveys to spatial frames and visualising trends. Well supported by the international community (Handbooks, Principles, Guidelines, etc.) Goals and targets for sustainable development and its measurements are well articulated. Problem Data Demand (geographic detail, more frequent data), reducing costs, sustaining trust. Capacity for geospatial analysis and regional science expertise. Stats SA CRUISE programme. Lets discuss. use of admin data - it s the one legacy we can leave behind (DG-Statistics Denmark) Future linking censuses and surveys to administrative and other data sources. Embedding the geography. Institutionalising. Coordinating. Standardising. Sound methodologies. Capacity building. Innovating. Re-organising, restructuring and managing the data ecosystem. Lets discuss. Thank you sharthil@statssa.gov.za