(Feedback from Statistics Canada, 27 September, 2017)

GSBPM is too survey collection oriented; and data acquisition / data exploration needs to be considered.

  • Suggest modifying the terminology to make it more relevant.  
  • Consider to change the word “collect” because this is very survey oriented.
  • Consider expanding the scope of Phase 4 to all forms of collection/acquiring data
  • Change to one of the following, ‘acquisition’, ‘acquire’, or ‘obtain’. In the definition, it could be expanded to include web scraping and acquisition of data. This would indicate that the model does not have to be linear, and that data integration is done at many stages for multiple purposes.
  • Add “data exploration” to 1.5 Check data availability

3 Comments

  1. InKyung Choi

    (Feedback from Statistics New Zealand, 28 September, 2017)

    Removing the word Collect and replacing it with something less survey focussed

    Recommended that the word collect is replaced by something else across the GSBPM.  The word Acquire is put forward as an option.

  2. InKyung Choi

    (Feedback from ONS, 2 October, 2017)

    Use of non-survey / admin data in our statistical outputs requires identifying data suppliers, communicating and negotiating with data suppliers, having data and documentation to be collected from data suppliers. Moving from surveys to non-survey / admin data means we are no longer in complete control of how the data is collected and (sometimes) processed. Should GSBPM be updated to reflect various stages of information assurance, online collection processes?

  3. InKyung Choi

    GSBPM Revision Group agreed NOT to change the name of phase 4 from "Collect" to "Acquire" (meeting on 3 Oct., 2017): “Collect” should not necessarily be interpreted in a restrictive way as interpreted in survey-based methods, it could mean a general way of obtaining data; explanatory text or example under the phase can be improved to better reflect the other sources of data

    GSBPM Revision Group will continue review other phases to better reflect other sources of data (meeting on 3 Oct., 2017)