As discussed in another thread we applied the dataset definitions in a more rigorous way to make the ranking more reliable and to compare countries against the same standard. This year we rejected all countries for evaluation that did not meet our granularity criteria for election results (polling station level).
Last year we let our reviewers decide as to whether they should reject or accept a submission. We saw however that this approach was sometimes inconsistent and could lead to unfair assessment.
As we discussed in another thread, we will discuss this point with the NDI, and will get back with an answer how we will treat this case in the future.