ActivityStatus codes - mixup of descriptions for codes 3 & 4?


(Steven Flower) #1

I few people have mentioned that the codelist descriptions for codes 3 (Completion) and 4 (Post Completion) ActivityStatus seem to be mixed up:

http://iatistandard.org/201/codelists/ActivityStatus/

3 - Completion: Physical activity is complete or the final disbursement has been made.

4 - Post Completion: Physical activity is complete or the final disbursement has been made,
but the activity remains open pending financial sign off or M&E

The description for 4 should be with code 3, and vice versa!

These descriptions were added in the 2.01 process, as they do not exist in 1.05: http://iatistandard.org/105/codelists/ActivityStatus/

Do others agree?


IATI Website is LIVE
What ideas do you have for TAG2018?
(Siem Vaessen) #2

Yes, I agree. Seems like a glitch and should be fixed.


(Dale Potter) #3

Thanks both, we’ve added this as one to discuss in the IATI tech team issue call.


(Yohanna Loucheur) #4

Agree that they seem mixed.


(Wendy Rogers) #5

Just an update to say that we are still looking at this one and hope to have a response or proposal on it soon.


(Yohanna Loucheur) #6

Was this ever resolved?


(Steven Flower) #7

@dalepotter @Wendy will know…


(Wendy Rogers) #8

We have been looking at this issue in some depth to try to resolve it in the easiest way possible. As a result we have been looking at how publishers are currently using these values and most do appear to be using them in the order 1,2,3 or 1,2,4,3 if the project does have post delivery activities. Whilst the numbers are not concurrent as might be preferred it is not necessarily a problem that the status values are not consecutive if it is clear when each value should be used? Therefore we would ultimately prefer to make the descriptions clearer with regard to how and when they should be used?

We did also discuss the possibility of re-ordering the values but that of course could only be done at an integer update of the IATI Standard. It would of course also mean that almost every publisher would have to amend their existing published datasets which needless to say is something we would want to avoid due to the great inconvenience it would cause.


(Andy Lulham) #9

Sorry to revive this thread – I’ve just run into exactly the same issue, and was perplexed by the documentation!

That’s part of the issue… But there’s also the codelist item names. It’s reasonable to assume the status with the English name ‘Post Completion’ comes after ‘Completion’. Similarly, the French name ‘Fermé’ sounds like it comes after ‘Finalisation’. So the names don’t really match the descriptions, either.

In that case, perhaps it would be useful to:

  1. Change the names of the codelist items to better reflect the descriptions, and
  2. Consider reordering the codelist items in the documentation? I.e. list item 3 after item 4. Like this:

(Wendy Rogers) #10

Thanks for the comments @andylolz and it would be good to know what option publishers and others would prefer. If we were to re-order the codelist (and make the change as part of an integer upgrade) would publishers be happy to amend any existing published data or would the effort required outweigh any benefit?


(Andy Lulham) #11

Thanks @Wendy! I’ve edited my previous comment to make it clearer that I’m after a solution that wouldn’t require an integer upgrade. I’m fast realising that is tricky!

So just to state: I think it’s okay that the codelist codes don’t proceed in numerically ascending order (according to descriptions and usage). For me, the only problem is that the names and descriptions don’t match, so the documentation is ambiguous. If this could somehow be clarified outside of an integer upgrade (e.g. by reordering the codes in the documentation only, or renaming the codes, or even adding an explanatory note in the documentation) then that would be great.


(Steven Flower) #12

I don’t think this has made its way to list on Standards Day?


(Yohanna Loucheur) #13

Not that I recall, despite having 120-odd items on Bill’s list.


(Steven Flower) #14

I don’t think this made it to 2.03 - as it was not on the list

Is this still in need of fixing though? In the same way we’re considering the commitments/budget description to be a “bug”?

–> @bill_anderson @Herman ?


(Herman van Loon) #15

@stevieflow

Agree, the current definitions are confusing. The problem though is i.m.o. not the description, but the naming of the code. Post-completion suggests that this status comes after the completion status. This is not true though. The 'post-completion’ status comes before the completion status.

This would suggest renaming ‘post-completion’ to ‘pre-completion’, leaving the descriptions unchanged.


(Andy Lulham) #16

Well, note that the descriptions weren’t there at 1.0x: http://iatistandard.org/105/codelists/ActivityStatus/

As I mentioned above, “the names and descriptions don’t match, so the documentation is ambiguous.” So as I see it, the options are:

  • Change the name (e.g. to pre-completion)
  • Change the description(s)
  • Add an explanatory note in the documentation

Of those, I’m most hopeful about option 3, given I suspect the other options will likely be viewed by @IATI-techteam as breaking changes.


(Yohanna Loucheur) #17

I have just been reminded of this problem as I closely review (almost) all IATI codelists for translation. Codes 3 & 4 names remain mismatched with their descriptions.

@IATI-techteam , any progress towards a solution? What did you think of the options outlined by @andylolz ?


(Steven Flower) #18

@YohannaLoucheur are you in agreement with this, too?


(Yohanna Loucheur) #19

Absolutely! The code names are in the wrong order (whereas the descriptions are in the right order).

In addition, FWIW, I think the French terms are slightly clearer. “Finalisation” implies that the closing process has started, but it doesn’t say that it’s completed. “Fermé” is unambiguously completed, done, over with. Is it worth trying to clarify the English by replacing “Post-completion” with “Closed”?

BTW, in line with the 1.04 upgrade that replaced all names with codes, the actual codelist is 1, 2, 3, 4, 5, 6. Names in various languages are simply attached to the codes; changing a name in one language should not be seen as a change to the standard.


(Steven Flower) #20

I just noticed that this post has been filed into “3.01 Integer Upgrade Proposals”. I dont think I did that originally (the first post was July 2015!) so wanted to check @IATI-techteam