GSBPM: General feedback about non-survey data sources

[Chris-ECE]

Comments relevant to non-survey data sources are listed below, in some cases with specific responses. In general, various changes made to the new version of GSBPM, including the renaming of phase 4 from "Collect" to Acquire" should reflect the desire to make it clearer that GSBPM is not specific to survey data.

Lithuania
GSBPM and GAMSO have already been implemented in Statistics Lithuania. The models are quite conveniently adapted and work when it comes to statistical information preparation processes. Now SL works on application and adaptation of the standards for the State Data Governance.
Though GSBPM describes and defines the set of business processes needed to produce official statistics, appears it can be applied to state data governance as well. In the revision process of GSBPM, we suggest paying more attention to processes of state data management and adapting the descriptions of phases and sub-processes to these needs. Data management I think this is more about non-survey sources of data (admin data)

Belarus
As GSBPM acts as the benchmark for describing and defining the set of business processes necessary for the production of official statistics, we think that it should include the features of organisation of production not only with the use of geospatial data, but also with the use of Big Data.

France
• Make sure all sub-processes of the GSBPM are adapted to data sources other than surveys (administrative data); Non-survey sources
• Text of the sub-processes are sometimes not easy to read with the addtion of information on survey, administrative data…
May be a first general description, independent of the source used. Then a text specifying the particularities according to the type of data processed.

Mexico
• We suggest emphasizing the use of non-statistical sources (i.e., big data). Non-survey sources

Ecuador
The GSBPM 5.1 version provides an exhaustive description about how to implement the standards for statistics based on administrative registers, unlike the version 5.0 which focused more in survey and census; nonetheless, it is important to analyze the pertinence of including in the document how to implement the model for each kind of data source (e.g. surveys, censuses, statistics based on administrative records), furthermore changing the document to make it easier to read and include less theoretical content. Not clear?
UNECE response: Hopefully, we have made changes that make this version clearer that are relevant to other sources of data. - Because the model is generic and conceptual/theoretical, documenting how each type of data source is handled is beyond the scope of this revision, especially as there are a number of extensions/specialisations of GSBPM already in existence..

From past GeoGSBPM Task team (2021)

1. (Meeting 10 Feb, 2021) Linked open data standard is important not only for geospatial data, but for all data that statistical organisations produce, this issue can be submitted for next GSBPM revision. Linked Open Data
   UNECE response: Linked Open Data is mentioned in 3.3, 7.1 and 7.2