Last year IATI's member’s assembly discussed and approved a new vision and mission document (see Strategic direction for IATI) outlining the strategic direction IATI should take in the years to come.
In the initial years, IATI has focused on persuading organizations to become publishers in IATI. In that light, IATI could be very liberal and accommodating in facilitating proposals for change of the standard. As of last year the vision and mission has changed in the direction of increasing the relevance and stimulating the use of our data. Therefore, proposals for change have to be viewed in that context: does the proposed change contribute to the relevance and use of IATI data.
That is easier said than done. The challenge is to translate the rather abstract concepts of data quality and data use to the operational level. When failing to do so we will risk that the standard develops in other directions, further increasing in complexity and scope, without contributing to the relevance and use of IATI data. Concerns about the complexity and scope of the standard and its negative impact on data use, have also been expressed on the 2017 TAG where John Adams hosted a session about the future of IATI.
So how to bridge to gap between the abstract concepts data use and quality to the operational level used to discuss, define and assess IATI upgrade proposals? What seems to be lacking is an intermediate, more fine-grained level of criteria, making it possible to assess the alignment of change proposals. Having participated in quite some discussions on the forum, I missed these criteria. Therefore I propose to develop a set of criteria, which reflect the essence of IATI’s vision and mission and which can be used in practice.
To clarify what I mean I will present some example criteria, their rationale and what it would mean if the criteria are applied to a change proposal:
Example 1: let’s assume that we have a criterion ‘Changes to the standard should promote that IATI stakeholders speak a common language’ with as rationale that speaking a common language facilitates data use among different stakeholders. If I were to propose to add a new sector vocabulary with the Netherlands MFA internal sector classifications, this change would not be approved since it would make it more difficult to do sector analysis, using data of multiple publishers.
Example 2: let’s say we have a ‘consistency’ criterion Changes to the IATI standard should increase and not decrease the internal consistency of the standard with as rationale that consistency improves the ease of use of data. Having this criterion in the past could have prevented introducing the inconsistencies in the results standard, now fixed in 2.03
Example 3: let’s say we have a ‘relevancy’ criterion 'Changes to the IATI standard should be relevant for a majority of IATI users in at least one constituency. The rational of this criterion could be that changes that are only relevant for a small part of the IATI community, will increase the complexity of the standard and make data use harder, but not benefit many users. To illustrate this point: currently 31% of all IATI elements are published by less than 10 publishers (source IATI dashboard elements )!
The real value of developing these criteria is i.m.o. that we have a common understanding of what we mean with data usability and data quality on a level that can be used in practice. I would like to hear what you think about this, maybe have some other examples and add some thoughts.