[OSM-talk] Report on the OSMF 2021 Survey after One Week

[Joseph Eisenberg]

Mappers have an email address associated with their mapper account. It
should be possible to randomly select 1% of users who submit a changeset
each month during the time of the survey, by sending a link to their
account. For iD users the link could be shown directly on the website, if
desired - this is already being done for everyone, but it isn't randomized
so self-selection bias is the big problem.

Database users download data at some point. You could automatically send a
link to the survey to 1% of people who download a planet file or extract
via openstreetmap.org - though getting a more representative sample would
require help from database mirrors such as geofabrik and openstreetmap.fr,
or sources such as overpass-turbo.

You could adjust the percentage who get sent links in each country to help
get demographic balance. For example, if 50% of active mappers with
changesets in Japan respond to the survey but only 10% of in Russia, you
could send out more links until the total number is representative of the
active mapper community in each area. And if a mapper doesn't respond to
the first message, you could send out a few reminders to those who were
randomly selected before.

More important would be making sure that the survey questions are actually
addressing the true concerns of the community in different places, and that
the survey is available in the language of most users in each area where
you hope to get useful responses. I appreciate that the OSMF has worked on
the second issue by trying to get a good amount of translations done. But I
don't think the questions were developed with global community input.

– Joseph Eisenberg

On Sun, Jan 24, 2021 at 2:14 PM Allan Mustard <
allan.mustard at osmfoundation.org> wrote:

> That would of course be a wonderful option.  Unfortunately, the Foundation
> does not presently have the ability to generate a randomized statistical
> sample of the larger community (which includes mappers, data users,
> software developers, systems/hardware admins, communicators, and event
> organizers).  If you have a specific proposal for how to accomplish this,
> one that is practical and would not break the bank, please PM me and we'll
> discuss.
>
> cheers,
> apm
> On 1/24/2021 4:53 PM, Joseph Eisenberg wrote:
>
> The better option would be to:
> 1) identify the community
> 2) create a randomized statistical sample of this larger community
> 3) survey just this randomly-selected subset.
>
> --
> -------
> *Allan Mustard, Chairperson*
> *Board of Directors*
> *OpenStreetMap Foundation*
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20210124/54918a99/attachment.htm>