Page MenuHomePhabricator
Create Task
Maniphest T411898

Community Insights 2026
Open, Needs TriagePublic
Actions

Description

We aim to distribute the biennial Community Insights survey of active editors across Wikimedia projects in April 2026. This task will serve as the Epic for the project to track various task components and the overall scope.

Substantial changes:

  • We will be using LimeSurvey instead of Qualtrics
  • We will be using a single-email opt-in rather than double-email opt-in
Set-up & Distribution
  • Review and revise questionnaire (Final)
    • Request stakeholder input for new questions - we do not want to substantially increase the length of the survey, but we should ask questions relevant to strategic or product initiatives which are likely to gain reliable and actionable data from the survey
    • Remove questions that are no longer relevant
  • Finalize questionnaire text, including email invitation copy (Final email copy)
  • Send new questions and copy for translation
  • Scope multilingual and stratified sample process for LimeSurvey - should we use separate surveys for all languages, or use one survey? Ensure that we are able to maintain data integrity (subsamples, bins) and question localization (e.g. demographics) for whichever route we choose.
    • (Decision: One survey with all languages, using a code system for strata weights)
  • Decide whether we want a method to remind participants to complete the survey
    • (Decision: no reminder method; response rates appear to be on par with past years)
  • Program the survey, QA
  • Update metawiki page for the 2026 survey
  • Request privacy statement
  • Draw sample of qualified participants to contact at release of newest mariadb dataset (April 2026)
  • Configure EmailUser API and survey links for invitations; test and QA
  • Send out invitations! (Target: mid-to-late April 2026, depending on editor dataset update)
    • (Sent 16 April 2026)
Data Cleaning & Documentation
  • Close survey (Target: 7 May EOD; may keep it open a bit longer as responses are still coming in at ~5-10 per day; ideally we want to drop to 0-2 daily)
  • Update metawiki pages to inform community members and potential respondents that the survey is closed
  • Clean responses (duplicates, straight-lining)
  • Calculate weights and response rates
  • Recode YoY questions to align with past questions/variables
  • Append to previous Community Insights datasets for YoY analyses
  • Update Community Insights codebook
  • Update gitlab for sampling and distribution code
Analysis

To be updated when we get to this stage.

Related Objects
Search...

Event Timeline

TAndic created this task.Dec 5 2025, 8:22 PM
TAndic updated the task description. (Show Details)Dec 5 2025, 8:34 PM
Miriam added a project: Epic.Dec 8 2025, 4:21 PM
Miriam subscribed.
Isaac added a parent task: T398239: [FY25-26 Research Team] Contributor Motivations.Jan 21 2026, 3:12 PM
Isaac mentioned this in T398239: [FY25-26 Research Team] Contributor Motivations.
Sj subscribed.Jan 31 2026, 9:51 AM

I. Could this become an annual survey? Having some process that runs every year would be helpful even if the annual core is simpler / fewer updates / lighter overhead and new Qs are only added every 2y

  • I have gotten tremendous value as a researcher (and for community planning) from past reader + editor surveys, and would love for this process to have slightly higher visibility including among groups that might be able to use a global survey vs running their own ad-hoc one.
  • Annual iteration helps reduce noise in the time series and also catches shocks closer to when they happen

II. Could this be done in combination with an annual reader survey? Also would be unbelievably helpful for planning; and having both done at the same time would help identify changes that affect only one of these groups, vs both. (vs internet-wide shifts if timed to align with other annual web surveys for more comparables!)

Sj added a subscriber: leila.Jan 31 2026, 9:52 AM

@leila excuse me if there's a better place to discuss reader surveys, I wasn't able to find a specific space for that.

Miriam added a subscriber: YLiou_WMF.Feb 25 2026, 12:24 PM

Hey @Sj I am very sorry for the late reply on this!

A few thoughts I'd like to share in response to your very valid points!

  • Community Insights (contributor survey) runs every other year - mainly because of effort vs impact considerations. While contributors' social indicators don't vary massively year over year, the effort of coordinating such a large survey across 24 languages is quite substantial for a team of our size. The team is however continuously looking at ways to improve efficiency in survey design/distribution/analysis, so we might be able to find ways to scale.
  • The Global Reader survey does run annually (T410695) albeit at a different time (December) than the Community Insights survey.
  • @YLiou_WMF is putting together a strategy brief (T391125) which will include the frequency and methodology of different survey measurements. So the cadence for the above reader/contributor survey might end up changing. I suggest following that task and sharing your very valuable suggestions there!
Sj added a comment.Mar 4 2026, 12:15 AM

Maybe intelligent automation can help with point 1 :) I'll take a look at the other brief.

TAndic updated the task description. (Show Details)May 5 2026, 8:46 PM
Content licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 unless otherwise noted; code licensed under GNU General Public License (GPL) 2.0 or later and other open source licenses. By using this site, you agree to the Terms of Use, Privacy Policy, and Code of Conduct. · Wikimedia Foundation · Privacy Policy · Code of Conduct · Terms of Use · Disclaimer · CC-BY-SA · GPL · Credits