[Osmf-talk] Surveys

Haklay, Muki m.haklay at ucl.ac.uk
Wed Dec 6 21:39:43 UTC 2017

I’ll join also the support of a sample, but I wouldn’t go for random. Going back to the study that Nama did (with bits of help from me), we started from the analysis of the planet file to have classifications of participants, then used the username to invite people through mapper-to-mapper communication. The reason that you need to follow a more stratified sampling is participation inequality (that’s why I’m going on and on about it - https://www.ubiquitypress.com/site/chapters/10.5334/bax.c/ ) – because there is such a tiny group of heavy contributors, without a targeted sampling, it’s very likely to miss them…

From: Tim Elrick [mailto:osm at elrick.de]
Sent: 06 December 2017 23:23
To: osmf-talk at openstreetmap.org
Subject: Re: [Osmf-talk] Surveys

I want to second Thomas here. To avoid unwanted biases the best would be to send out an e-mail to a random sample of all registered mappers. In case, we are only interested in active mappers, we could limit this to a somehow defined active mode (e.g. 'has mapped at least once last year'), but still should approach them by e-mail. I cannot tell, however, whether this would conflict with OSM's intention or some national laws not to send out unsolicited e-mails.
I want to offer to volunteer drafting a survey as well, if needed, as I already have some experience with academic surveys.


Am 06.12.2017 um 16:05 schrieb Thomas Gertin:

That is an interesting idea of putting notices for a survey into the

desktop editors. However, I don’t think that would be my top choice if

I were to design a survey for OSM.

I think that surveying a random sample of the population is extremely

important. It is important not to risk introducing unintentional or

intentional biases in the survey.

My naive assumption is that a survey advertised within the desktop

editors would result in a higher participation rate from advanced

mappers and from OSM users who maintain more active ties within the

community. It would difficult to measure and correct for these biases

(and potentially others). In addition, this could introduce issues

with the validity of the results.

We need to be wary of designing surveys where people self-select

themselves to participate.

-Tom G

On Wed, Dec 6, 2017 at 2:40 PM, Michael Reichert <nakaner at gmx.net><mailto:nakaner at gmx.net> wrote:


Am 06.12.2017 um 16:16 schrieb Simon Poole:

The main problem with this (again broken record mode) and really ages

old issue, is that we do not ask sign ups for permission to send them

unsolicited mail in any form.

The problem is not terribly difficult to fix, just nobody that has

proposed to do a survey (and there have been dozens and dozens over the

years) , has actually had enough interest to either fix the issue

themselves or pay somebody to do so.

If a survey should be answered by active mappers, notifications in their

editor would be a way to reach even more than using third-party news

services like Wochennotiz and WeeklyOSM. JOSM has a message of the day

feature as long as I can remember. iD is usually used from osm.org, i.e.

a message (top left corner) on osm.org for logged-in users could reach

mappers using iD. This would also reach users who use Potlatch 2. Only

mobile editors are not reachable. I myself think that popus with the

latest news are more annoying than helpful in mobile apps.

To summarize it, we can reach mappers using desktop editors already.

There are only two groups which are unreachable: mobile-only mappers and

inactive mappers. If I were an inactive mapper, I would find it annoying

to get an email asking to contribute to a survey.

Best regards



Per E-Mail kommuniziere ich bevorzugt GPG-verschlüsselt. (Mailinglisten


I prefer GPG encryption of emails. (does not apply on mailing lists)


osmf-talk mailing list

osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>



osmf-talk mailing list

osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20171206/0f88a2f4/attachment.html>

More information about the osmf-talk mailing list