Regaring measuerment of editor retention - this is
tricky - as in fact many
participant created new accounts only to join the contest. Some of them had
accounts on Wikipedia (but different) - some others - abandoned their
accounts and created a new ones for various reasons (the most trival - they
have forgoten passwords). There are also user who are active only during
contensts - also for various reasons - not only due to possibility to win
attractive prizes, but also because the normal upload process is too tricky
for them, or they don't know what to photograph if there is no easy to use
list of objects.
In fact measurement of editor retention is tricky even for workshops if it
is only based on list of nicknames. I saw this many times - that people
create the accounts during the workshop and then abandon them, but create
later a new ones. The only effective way to follow the retention of users
after workshop is to collect their e-mails and then survey them some time
after the workshop. It might produce completely different picture that
studies based on following the activity of accounts created during
workshops...
2015-05-07 11:34 GMT+02:00 Lodewijk <lodewijk(a)effeietsanders.org>rg>:
Hi Sam,
The main misconception (which is understandable, but also often pointed out
already) is that Wiki Loves Monuments can be fundamentally different
projects from a goals-and-outcomes point of view, based on the interests
and strenghts of the local organizers and the local situation. In some
countries, the main outcome of the competition is that it brings together
organizers for a first project, that can then move on, and leverage their
collaboration in other projects. In other countries it fosters
collaborations with other organizations.
In some countries, it is a very grassroots competition, with low budget and
big focus on getting a lot of photos. In other countries, there is a lot of
effort (and funding) going into catching editors, setting up structures or
overcoming the local challenges or making concepts better aware.
Aside from the fact that many of these outcomes are qualitative, which
seems to get no attention in the (summaries of the) reports, but do get
described in the reports of the individual contests, the local competitions
are too diverse to try and catch as one group.
This is a fundamental flaw (pointed out before) in the approach. The work
is appreciated of course, the numbers can be useful - the way they are
presented is however very sensitive for major misunderstandings.
Besides this, there are several very specific flaws in the number crunching
that have been pointed out, which are for example messing up the numbers on
editor retention.
I hope that at some point WLM organizers can be given the tools, enthusiasm
and support to create their own evaluation on a larger scale. That way I
hope that some of the flaws can be avoided thanks to a better understanding
of the collaborations, structures and the projects in general.
All in all it is good to have something 'to shoot at' but I would prefer
that these reports are produces more in concert with the stakeholders
involved and affected, rather than 'announced' and 'presented' to the
wide
community.
Best,
Lodewijk (effeietsanders)
member of the international coordinating team 2011-2013
On Wed, May 6, 2015 at 4:40 PM, Samuel Klein <meta.sj(a)gmail.com> wrote:
Claudia, I share your concerns about reducing
subtle things to a few
numbers. Data can also be used in context-sensitive ways. So I'm
wondering if there are any existing quantitative summaries that you find
useful? Or qualitative descriptions that draw from more than one
project?
Figuring out what ideas are repeatable, scalable,
or awesome but one-time
only, is complex. We probably need many different approaches, not one
central approach, to understand and compare.
I'm glad to see data being shared, and again it might help to have many
different datasets, to limit conceptual bias in what sort of data is
relevant.
On May 6, 2015 9:59 AM, "Claudia Garád" <claudia.garad(a)wikimedia.at>
wrote:
> Hi Sam,
>
> I am sure there are figures and stories that the various orgs collect
and
> publish. But they are spread across different
wikis and websites and/or
> languages. E.g. many of the FDC orgs are looking into ways to
demonstrate
these
more qualitative aspects of our work (e.g. by storytelling) in
their
reports.
But these information does not get the same attention and publicity in
the
wider community as the evaluation done by the
WMF. Many WMAT volunteers
and
I myself share the concerns expressed by Romaine
that these
unidimensional
> numbers and lack of context foster misconceptions or even prejudices
> especially in the parts of the community that are not closely involved
in
> the work of the respective groups and orgs.
>
> Best
> Claudia
>
>
>
> Am 06.05.2015 um 13:40 schrieb Sam Klein:
>
>> Hi Romaine,
>>
>> Are there other evals of WLM projects that capture the complexity you
>> want?
>>
>> Perhaps single-community evaluations done by the WLM organizers there?
>>
>> Sam
>>
>> On Wed, May 6, 2015 at 7:21 AM, Romaine Wiki <romaine.wiki(a)gmail.com>
>> wrote:
>>
>> Hi all,
>>> In the past months the Wikimedia Foundation has been writing an
>>> evaluation
>>> about Wiki Loves Monuments. [1]
>>>
>>> At such it is fine that WMF is writing an evaluation, however they
fail
>>> in
>>> actual understanding Wiki Loves Monuments, and that is shown in the
>>> evaluation report.
>>>
>>> As a result on the Wiki Loves Monuments mailing list a discussion
grows
>>> about the various problems the
evaluation has.
>>>
>>> As the Learning and Evaluation team at the Wikimedia Foundation
already
>>
had
>> released the first Programs Reports for Wiki Loves Monuments, we are
now
>>> put as fait accompli with this evaluation report.
>>>
>>> Therefore I am writing here so that the rest of the worldwide
Wikimedia
>>> community is informed that this is
not going right.
>>>
>>> Wiki Loves Monuments is not just a bunch of uploads done in
September,
>>> the
>>> report is too simplified without actual understanding how the
community
>>> is
>>> doing this project.
>>>
>>>
>>> Romaine
>>>
>>>
>>>
>>> [1]
>>>
>>>
>>>
https://meta.wikimedia.org/wiki/Grants:Evaluation/Evaluation_reports/2015/W…
>>>
_______________________________________________
>>> Wikimedia-l mailing list, guidelines at:
>>>
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
>>> Wikimedia-l(a)lists.wikimedia.org
>>> Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>>
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
>>
>
>
>
_______________________________________________
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l(a)lists.wikimedia.org
Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
_______________________________________________
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l(a)lists.wikimedia.org
Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>
_______________________________________________
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
Wikimedia-l(a)lists.wikimedia.org
Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>