Another survey, another botched sampling methodology

Discussions on Wikimedia governance
User avatar
thekohser
Majordomo
Posts: 13406
kołdry
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Another survey, another botched sampling methodology

Unread post by thekohser » Wed Jun 08, 2016 3:01 pm

You'll see that the Wikimedia Foundation wants our help with finding a new Executive Director!

As usual, the WMF has really stepped up its careful sampling methodology:
Our survey is currently open in the top 10 wiki languages. A sample of
editors from various languages have been invited to participate and we are
also sending an invitations to anyone here and through our networks.
So, anyone can participate, and if you want to participate about 12 or 13 times, that's okay, too! (Just use different IP addresses.)

Time is of the essence, though -- the survey is only going to be open another day, unless they extended the one-week plan.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
Kingsindian
Habitué
Posts: 2593
Joined: Fri Nov 20, 2015 10:07 am
Wikipedia User: Kingsindian

Re: Another survey, another botched sampling methodology

Unread post by Kingsindian » Wed Jun 08, 2016 3:17 pm

I wonder why they even bother with this kind of stuff. It's clear that they have no intention of looking at the data. Another example comes from the Inspire campaign "Leaderboard":
An idea's placement in this list does not signify that it is better or worse than any other idea, and the number of endorsements an idea receives does not determine whether that idea will be selected for funding..
Then perhaps they shouldn't call it "Leaderboard", with ideas sorted according to number of endorsers? Just a thought.

User avatar
Poetlister
Genius
Posts: 25599
Joined: Wed Jan 02, 2013 8:15 pm
Nom de plume: Poetlister
Location: London, living in a similar way
Contact:

Re: Another survey, another botched sampling methodology

Unread post by Poetlister » Wed Jun 08, 2016 8:48 pm

thekohser wrote:So, anyone can participate, and if you want to participate about 12 or 13 times, that's okay, too! (Just use different IP addresses.)
For bonus marks, contribute in three different languages.
"The higher we soar the smaller we appear to those who cannot fly" - Nietzsche

Hex
Retired
Posts: 4130
Joined: Thu Nov 01, 2012 1:40 pm
Wikipedia User: Scott
Location: London
Contact:

Re: Another survey, another botched sampling methodology

Unread post by Hex » Wed Jun 08, 2016 9:04 pm

Poetlister wrote:
thekohser wrote:So, anyone can participate, and if you want to participate about 12 or 13 times, that's okay, too! (Just use different IP addresses.)
For bonus marks, contribute in three different languages.
Why stop there? You could put a photo of someone from the internet on your user page and pretend that it's you.
My question, to this esteemed Wiki community, is this: Do you think that a Wiki could successfully generate a useful encyclopedia? -- JimboWales
Yes, but in the end it wouldn't be an encyclopedia. It would be a wiki. -- WardCunningham (Jan 2001)

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: Another survey, another botched sampling methodology

Unread post by thekohser » Thu Jun 09, 2016 10:18 am

Hex wrote:
Poetlister wrote:
thekohser wrote:So, anyone can participate, and if you want to participate about 12 or 13 times, that's okay, too! (Just use different IP addresses.)
For bonus marks, contribute in three different languages.
Why stop there? You could put a photo of someone from the internet on your user page and pretend that it's you.
The survey never asks for the respondent's User page.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

Post Reply