On Mon, Mar 11, 2019 at 12:16 PM Ilario valdelli <valdelli(a)gmail.com> wrote:
Any study is interesting, but if it could be country-based, it would be
We agree with you that the country component is quite important. There
is some ongoing engineering work to make the feature available in
QuickSurvey  which is the extension we use. If it gets ready, we
will include at least some sampling by country.
Let me give you a couple of reasons why from our perspective sampling
by country is key:
* Understanding readers from countries such as Nigeria is hard through
English Wikipedia as the traffic in enwiki is dominated by other
countries. In order to get enough responses from Nigeria, we have to
ask many more questions from the rest of the world which is not
something we want to do.
* Many chapters are organized by geographical regions and learning
about the readers in their geography can empower them in new ways.
* If we have enough responses from different countries, we can have
more accurate debiasing steps. For example, if we have enough data
from country x, we can look at the age distribution of respondents and
see if that age distribution matches the age distribution from that
country based on external databases available. If not, we can try to
correct for the differences, or at least be aware of the caveats.
On 06/03/2019 22:12, Leila Zia wrote:
> Hi all,
> As I mentioned in an earlier thread , we will be running reader
> surveys across a number of Wikipedia languages to learn about the
> reader needs and motivations in these languages as well as some of
> their demographic information (and perhaps the correlations between
> demographics and user motivations and characteristics).
> If your language community is interested to have statistics on the
> distribution of reader gender, age, education, native language, and
> geographic region (rural/urban) in your language (and depending on how
> much data we collect in your language, perhaps more insights), this is
> your chance to indicate interest at:
> I initially communicated 2019-02-15 as the deadline to sign up. Since
> then, we have run a pilot test on enwiki and we are investigating some
> of the results to see if any changes in the survey questions are
> needed. You have now time until 2019-03-15 to indicate interest.
> As always: this call is primarily a service to your language
> community. If you like it, take action on it. If you don't, no action
> is needed. :)
>  https://lists.wikimedia.org/pipermail/wikimedia-l/2019-February/091762.html
> Wikimedia-l mailing list, guidelines at:
> New messages to: Wikimedia-l(a)lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l