Grants:PEG/User:Gretchenmcc/Linguistics Editathon series: Improving female linguists' participation and representation on Wikipedia/Report

From Meta, a Wikimedia project coordination wiki


Report accepted
This report for a Project and Event grant approved in FY Pending has been reviewed and accepted by the Wikimedia Foundation.
  • You may still comment on this report on its discussion page, or visit the discussion page to read the discussion about this report.
  • You are welcome to Email grants at wikimedia dot org at any time if you have questions or concerns about this report.


Note: I guess I was supposed to put User:Gretchenmcc rather than Gretchenmcc in the "NAME" field, so for reference the actual grant is here. Let me know if I should remake this report with a matching name, I guess.

Project status[edit]

Did you comply with the requirements specified by WMF in the grant agreement?
YES
Is your project completed?
YES

Activities and lessons learned[edit]

Activities[edit]

Timeline:[edit]

  • Pre-grant - preliminary planning with organizers of all three conferences to agree to hold an editathon and waive registration fee of facilitator
  • May 2015 - planning for May editathon, including creation of French version of how to edit Wikipedia for linguists slides
  • 31 May 2015 - May editathon at Canadian Linguistics Association annual meeting in Ottawa
  • June 2015 - planning for July editathons, wrote interim report for May editathon
  • 8, 15, 22, 29 July 2015 - four editathons at the Linguistic Summer Institute in Chicago
  • August-September 2015 - planning for October editathon, wrote interim report for July editathons
  • 16 October 2015 - October editathon at North East Linguistics Society annual conference in Montreal
  • October 2015 - wrote interim report for October editathons (including activity not funded by grant).

Ancillary outcomes:[edit]

  • 1 workshop about using Wikipedia in the classroom (July, Chicago) and many informal conversations at conferences: 19 linguistics professors who expressed interest were put in contact with Jami Mathewson at WikiEd. I also provided support to the Linguistic Society of America's new partnership with WikiEd, which was announced in this blog post.
  • French version of instructions created with the support of the Canadian Linguistics Association, who wanted the event to be bilingual. (English and Spanish versions already existed at bit.ly/lingwiki and bit.ly/lingwikies, respectively. French version now at bit.ly/lingwikifr )
  • Created fun linguistics stub-sorting guide with User:Keilana which was used at an editathon that I ran outside the grant itself (linguistics undergrads sorted 312 unclassified linguistics articles).
  • Long-term lingwiki participant User:loztron successfully applied for her own "pizza grant" from Wikimedia UK to fund food for a series of editathons in London, where participants put information from academic grammars into language articles on Wikipedia

Lessons learned[edit]

What worked well?
Running editathons with combination of rotating facilitator & an online guide custom-made for this audience at a short url (bit.ly/lingwiki) worked well. Having the how-to online meant that people could click directly through to links rather than typing them in from a paper handout (and bonus, no printing costs) and it was easy to refer to and forward to others later. The guide + rotating facilitator model was very hands-on and even new editors actually made edits. There were enough attendees who had some experience with Wikipedia or were quick at picking up tech skills that people could help their neighbours if necessary in the larger groups. I also talked specifically with attendees who I knew had Wikipedia experience in advance, so that if the group was too big they were prepared to get up from their laptop and help facilitate instead, but if it was small they could edit rather than having too many facilitators standing around talking to each other (thanks User:RoboKyun!).
Having a "captive audiences" at conferences worked well. Both in terms of promotion on official conference schedule and announcements, and being in times/spaces where attendees are already. Huge attendance at may editathon because it came with free lunch and was in a room that conference talks were also being held, so attendees knew exactly where to go and it was easy to stumble in (also since attendees are going to stick around for afternoon talks anyway and would have had to pay for own lunch otherwise, they have an incentive to come).
Using editathons as an excuse to talk about Wikipedia with conference attendees, because classic conference small talk is "so are you presenting something?" So even those who didn't attend the editathons itself learned about Wikipedia. (I sent 19 names of professors potentially interested in using Wikipedia in the classroom to Jami at WikiEd as a result of confs.)
According to a few other grantees I talked to, I seem to have gotten a very good survey response rate (67%) although it still wasn't everyone, so I also did a manual headcount so at least we'd have accurate participant numbers. Factors that helped: having the survey at a short, easy to type url (bit.ly/lingwikisurvey) and stopping the editathon 10 min before the end and asking participants to finish up their last few edits and do the survey seems to have helped. As did promising them that they'd get to see a list of all the articles created and other stats soon after the editathon. I also asked participants individually if they'd done the survey yet as they seemed about to close their computers, and pointed out that doing the survey would help justify the free food they'd gotten.
It was worth using a blank write-in field for "gender" in the survey as several participants wrote in genders that were not male or female.
We circumvented the 6 account per IP address in 24 hours creation limit using three strategies: encouraging participants to create an account in advance, encouraging participants to revive long-dormant Wikipedia accounts using email password recovery if necessary (this was surprisingly common), and by encouraging participants who had mobile data connections on their phones to make accounts using their phone's IP address rather than the university network's, if possible. I know I could also have requested bypass permission, but with this combination of strategies, the 6 account max was not an issue.
There was little correlation between the number of attendees at a conference and the number of participants at the editathon held at that conference: the smallest conference, in May, had the largest number of attendees, and the medium-sized conference, in October, had the smallest number of attendees. It's worth having editathons even at smaller conferences, since these often have a less jam-packed schedule that more easily accommodates an editathon. The first day of a conference is definitely better than a later day as participants do get tired.
What didn't work?
The editathon at the end of a conference day when attendees were tired had fewer attendees than the lunchtime or midafternoon editathons (I thought dinnertime might work better than lunch because we had more time and could go for a drink after, but this turned out not to be the case. Learning!) At any rate, the October editathon couldn't have been at lunch because the poster session was over lunch.
4-hour editathons in July were really too long, next time I'd go for 2-3 hours. (The May/Oct confs had tighter schedules and let me have about an hour and a half, which was a bit on the short side but better than nothing, also the easiest to fit into the schedule.) I am piloting in January 2016 an editathon that runs for 3 hours parallel to afternoon conference talks, and we'll see how that goes.
Asking for signups as part of conference registration had a null or even negative effect (38 people signed up for the October editathon but only 12 attended (which I why I went slightly over budget for pizza, because i was expecting this huge turnout, but then I had to encourage people to still come even if they hadn't signed up in advance). I think this demonstrates interest in editing Wikipedia on the part of linguists but poor evening timing given the packed conference schedule. (I heard that the plenary talk immediately preceding the editathon was also poorly attended, as was the other workshop that ran at the same time as the editathon.)
What would you do differently if you planned a similar project?
The metrics of "how many participants" and "how many articles" are weird when there are recurring participants who may or may not work on the same articles from one editathon to another, so I'd like to think more about how to track that accurately.
I'm exploring other options for how to fit an editathon into a conference schedule where it doesn't work to have it over lunch — another (non-WMF funded) editathon I'm running in January 2016 at a conference will have it running as a drop-in in the afternoon concurrently with talks, and we'll see if that works better.
Be even more clear about the need to bring laptops — I did mention it, but a few people still showed up only with smartphones and didn't have a great editing experience. The people who brought tablets with external keyboards seemed to do okay, kinda, although I wouldn't particularly recommend it especially because many participants were adding citations which require using multiple windows/tabs at the same time.

Learning patterns[edit]

I endorse the learning patterns Fostering affinity groups and Six-account limit.

Outcomes and impact[edit]

Outcomes[edit]

Provide the original project goal here.
I expect that 6 further editathons with 15-30 participants each will result in the creation/improvement of 130-175 articles, the involvement of at least 75 unique participants and quite a few returning ones, and hopefully further engagement such as future satellite editathons and continued editing with the support of the #lingwiki hashtag.
Did you achieve your project goal? How do you know your goal was achieved? Please answer in 1 - 2 short paragraphs.
The 6 editathons happened, with a total of 120 participants and 103 articles created or improved, according to the survey that I conducted (see detailed stats below). This is a slightly higher number of participants and a slightly lower number of articles than anticipated, but certainly within the ballpark of my projection.
This project also succeeded in raising awareness among linguists about Wikipedia editing as a form of outreach, which I should have included as part of the official goal since it was definitely an unofficial one! One professor that I talked to at the May editathon used Wikipedia with her linguistics class in the fall; I also put 19 linguistics professors that I'd talked with individually at the conferences in touch with Jami Mathewson at WikiEd after the Memorandum of Understanding between the LSA and WikiEd was signed in November 2015. This grant also led to other spin-off editathons such as the stub sorting one using the guide I created with User:Keilana and the series organized by User:loztron.

Progress towards targets and goals[edit]

Project metrics

Project metrics Target outcome Achieved outcome Explanation
At least 75 participants 120 participants Participants were counted by a headcount in case they didn't all fill out the survey
At least 130 articles created or improved 103 articles created or improved The primary reason why I think I over-estimated here was that many participants kept working on the same article for several weeks and/or worked jointly on the same article as part of the four editathons in July. While this resulted in substantial improvements to those articles, it makes the simple number of articles look lower.

Furthermore, only 81 out of 120 participants filled out the survey asking them to list the articles they worked on. By my estimates, almost everyone who attended worked on at least one article and several worked on more than one (except a few people who just dropped in for a few minutes seeking information), so this number is an under-count.

Over 50% female participants 59% female +1% agender participants Should have phrased goal "over 50% non-male participants" since I doubt that nonbinary people are overrepresented on Wikipedia.


Global Metrics[edit]

We are trying to understand the overall outcomes of the work being funded across our grantees. In addition to the measures of success for your specific program (in above section), please use the table below to let us know how your project contributed to the Global Metrics. We know that not all projects will have results for each type of metric, so feel free to put "0" where necessary.

  1. Next to each required metric, list the actual outcome achieved through this project.
  2. Where necessary, explain the context behind your outcome. For example, if you were funded for an edit-a-thon which resulted in 0 new images, your explanation might be "This project focused solely on participation and articles written/improved, the goal was not to collect images."

For more information and a sample, see Global Metrics.

Metric Achieved outcome Explanation
1. # of active editors involved 27 This number only reflects attendees who filled out the survey and included their username, so it's almost certainly an undercount.
2. # of new editors 40 This number only reflects attendees who filled out the survey and included their username, so it's almost certainly an undercount.
3. # of individuals involved 120 These numbers are based on informal headcounts that I did at the events, not how many participants filled out the survey, so they are the most accurate of the figures (although it's possible that I may have overlooked a couple people arriving or leaving partway through)
4a. # of new images/media added to Wikimedia articles/pages 0 Not the goal here
4b. # of new images/media uploaded to Wikimedia Commons (Optional) 0 Not the goal here
5. # of articles added or improved on Wikimedia projects 103 This is almost certainly an undercount as by observation, almost all participants edited at least something, but not all filled out the survey (although of those who did, some of those who declined to give their username still gave urls of articles they worked on, so it's less of an undercount than the stats that rely exclusively on wikimetrics)
6. Absolute value of bytes added to or deleted from Wikimedia projects 263513 This is almost certainly an undercount as by observation, almost all participants edited at least something, but not all filled out the survey (and of those who did, some of them declined to give their username so could not be included in wikimetrics)
Learning question
Did your work increase the motivation of contributors, and how do you know?
Participants responded favourably to several questions about motivation asked as part of the survey at the end. To the question "Do you think you'll keep editing Wikipedia in the future, even without an editathon?" 60% of participants responded "yes" and 40% responded "maybe" (0 responded "no"). To "Would you participate in a future Wikipedia editathon?" 79% of participants responded "yes" and 21% responded "maybe" (0 responded "no"). To "Would you try to run an editathon yourself or have students edit Wikipedia?" 31% responded "yes", 46% responded "maybe", 16% responded "no" and 6% responded with a write-in comment to the effect of being willing in principle but wanting more training.
In a blank text field where participants were asked for comments and suggestions for improvement, while most just left this field blank, several participants made positive comments:
    • I like having all the slides available and letting folks go at their own pace. That was a great idea.
    • Thank You! So much fun!!
    • Learned about it at LSA 2015; was unable to attend then and was glad to see it at NWAV.
    • Gretchen (our Wikipedia Editathon leader) is great---enthusiastic, knowledgeable, patient, all the qualities required for such an effort
    • Thanks for organizing the editathon.
    • This is fun. I think I might continue working on the same article. It needs more help. Or I could try something new. Thanks!
    • No comments but I just want to express my support for the editathons!
"Negative" comments were about minor technical/organizational fixes, such as linking to the language article template (now done) and sharing the how-to slides ahead of time, which clearly demonstrated interest in making the event as good as possible.


Impact[edit]

What impact did this project have on WMF's mission and the strategic priorities?

Option A: How did you increase participation in one or more Wikimedia projects?

Option B: How did you improve quality on one or more Wikimedia projects?

Option C: How did you increase the reach (readership) of one or more Wikimedia projects?

Option A: This project increased participation among two groups identified as strategic priorities by Wikimedia: women and academic experts.

In a write-in text field, 59% of participants identified as "female" or "F", 40% of participants identified as "male" or "M", and 1% identified as "agender". This is similar to the gender ratio of linguistics as a field and considerably better than the gender ratio of Wikipedia. (The presence of participants who did not identify as male or female also underlines the importance of allowing people to choose from more than just binary gender options.)

According to the survey, 14% of participants were linguistics profs, 5% were linguists working outside academia, 9% were linguistics post-docs, 62% linguistics grad students, 6% linguistics undergrads, 3% other categories of linguist, and 1% not a linguist, for a total of 99% linguist participants, 90% of which had some grad school or higher.

Option B: This project also increased coverage and quality among two areas identified as strategic priorities by Wikimedia: articles about notable women and academic articles.

Of the 103 articles created or improved as part of this series of editathons, 21 were biographies of linguists of which 15 were female linguists. A further 30 articles were about minority languages, dialects, and the people who speak them (according to WikiProject:Languages, over half of language articles are stubs), 45 were about theoretical topics in linguistics (often chosen from the Linguistics Stubs category), and the remainder about other linguistic topics, such as linguistic organizations and databases. Having linguists edit at linguistics conferences is highly effective in encouraging them to work on topics related to their expertise.

Slightly fewer biographical articles were created than anticipated, I think largely because the two shorter editathons weren't quite enough time for new editors to create a whole new article so they ended up editing stubs instead. Participants at the July editathon also mentioned a well-cited freely-accessible online database about Amazonian languages whose information could be added to Wikipedia - talks are in progress about this.

A full list of articles edited can be found at [1] these three posts.

Reporting and documentation of expenditures[edit]

This section describes the grant's use of funds

Documentation[edit]

Did you send documentation of all expenses paid with grant funds to grants at wikimedia dot org, according to the guidelines here? Answer "Yes" or "No".
YES

Expenses[edit]

Please list all project expenses in a table here, with descriptions and dates. Review the instructions here.

I've made an auto-calculating google spreadsheet here with all required info: https://docs.google.com/spreadsheets/d/1oBx331VW3undudjLt38okp0D4s67Tw8U3ER953qXHjw/edit#gid=0

Notes on currency and budget reporting: 1. The whole grant took place in two currencies, Canadian dollars and US dollars. I elected to request funds in USD because the bulk of my expenses were originally in USD, but as I'm Canadian and live in Canada, the USD expenses were ultimately all converted to CAD by my bank and/or credit card, whereas the CAD expenses were not converted. I have therefore elected to report all actual expenses in CAD as they appeared on my bank/credit card statement (as applicable) — and have included screencaps from my CAD credit card statements in addition to receipts where funds were originally charged in USD. It's confusing, but less so than any other options. This is especially relevant because the Canadian dollar has fallen since this grant was awarded.

2. In the spreadsheet, the "projected USD" column is the expenses as budgeted for in the grant. The "projected CAD" column is the expenses in Canadian dollars as calculated using the exchange rate that my bank gave me when it received the grant as a wire transfer from Wikimedia. The "actual CAD" column is the amount I paid in CAD, either when purchasing in CAD or the CAD amount that my credit card charged me when purchasing in USD. Sometimes the actual amount reflects a different exchange rate, sometimes it reflects costs different than projected, sometimes both - see the "comments" column.

(These notes are copied from my approved request for re-allocation of funds on 17 August 2015 as I was dealing in two currencies at multiple exchange rates, so it's more complicated than a usual table. I believe this remains the clearest way of doing things.)


Total project budget (from your approved grant submission)
3736 USD = CAD 4424.30 (as reported by my Canadian bank when I received the wire transfer from WMF)
Total amount requested from WMF (from your approved grant submission, this total will be the same as the total project budget if PEG is your only funding source)
3736 USD = CAD 4424.30 (as reported by my Canadian bank when I received the wire transfer from WMF)
Total amount spent on this project
CAD $4428.85 (as reported by my Canadian credit card, using rates current at time of each transaction)
Total amount of Project and Event grant funds spent on this project
3736 USD = CAD 4424.30 (as reported by my Canadian bank when I received the wire transfer from WMF)
Are there additional sources that funded any part of this project? List them here.
The Canadian Linguistic Association sponsored food (CAD $108.19) and registration (CAD $100.00) for the May event (this was not originally budgeted for but was included in the submission for reallocation during the course of the grant and approved on 17 August 2015).
All conferences provided free space, wifi, registration for editathon organizer (except for the CLA's registration, this was budgeted for in the original grant) but since no money exchanged hands I am not counting them here.

Remaining funds[edit]

Are there any grant funds remaining?
Answer YES or NO.
NO
Please list the total amount (specify currency) remaining here. (This is the amount you did not use, or the amount you still have after completing your grant.)
$0 (I ended up exceeding the budget by $4.55 CAD, which I'm expecting to just absorb out of pocket)
If funds are remaining they must be returned to WMF, reallocated to mission-aligned activities, or applied to another approved grant.
Please state here if you intend to return unused funds to WMF, submit a request for reallocation, or submit a new grant request, and then follow the instructions on your approved grant submission.
N/A (no funds remaining)