This observation has been made by a few people (some of them involved in the scholarship decision-making process) is that past recipients often continue to out-perform others in terms of the criteria in subsequent years. What hasn’t been commented on is why this is so?

If we believe that an attendee to Wikimania benefits in terms of learning new skills, hearing new ideas, making new contacts, then we should hardly be surprised if an attendee is then in a position to “grow” as a Wikimedian and hence be more able to “out-compete” others who didn’t have the benefit of attending. (And If we don’t believe that attendees benefit or grow from Wikimania attendance, then we should stop running Wikimania). Also the scholarship recipient has an expectation to share with their community what they have learned, even that process of sharing adds to their list of activities that they can use as evidence as subsequent scholarship applications.

Aside. If you have read the book Freakonomics or followed their blog, you will be aware

https://en.wikipedia.org/wiki/Freakonomics

of their study of how professional footballers tend to have their birthdays clustered in a few months of the year and how this phenomenon has its roots in spotting football talent in very young players and then training them. Because junior sport is usually based around age limits with a specific cut-off day, the children who just exceed the age limit by a month or two will usually be less physically developed than those who exceed the age limit by 10 or 11 months. Thus, the older children in the cohort are more likely to be selected for the team and receive coaching. Next year (still with a relative age developmental advantage AND with one year of extra coaching) these older children in the cohort are again appear the most able and again selected for the team (giving them yet another year of coaching benefit over those not selected). This cycle repeats throughout their childhood ensuring the older ones within the “age year” are disproportionate represented in both junior sport and then into college and professional sport, giving rise to the observed clustering of birthdays in professional footballers.

This is exactly the same phenomenon as we are seeing with Wikimania scholarships.

How can the playing field of Wikimania scholarships be made a little fairer? I don’t think the answer lies in deducting some points from those who have had a scholarship before. I think the solution lies in having two streams of scholarships, one for the first timers who compete among themselves on criteria that assesses their *potential* to “grow” through the Wikimania experience and a second set of scholarships for those who are applying to come for a second/third/… time with criteria more appropriate  to that group, how much did they “grow” and how much did they “share” relative to the number of Wikimania opportunities they have had (note one might also want to include attendance at Wikimedia Conference and other similar movement events in this regard)?

Note in both streams it is still possible to include factors like the Global North/South issue, minority groups, etc in the criteria as consistent with the movement’s strategic goals. The key difference is whether you are assessing only potential for growth from attending for the first-timers as opposed to observed growth from past attending and likely potential for further growth from additional attendance for the repeaters.

If that approach is taken, then the only question that remains is the relative number of scholarships (or amount of funds) available in each of the two streams. Obviously there’s a range of possibilities, but I would be tempted to operate on a simple pro-rata principle at least in the first year of operation. After the weeding out of the ineligible or people who show poorly against the criteria (however many phases there are to do that), look at the size of the two remaining groups and go pro-rata. That is, if after the preliminary cull(s), there are 200 potential first-timers and 100 potential repeaters, then allocated twice as many scholarship (or twice as much funding) to the first-time group as to the repeater group. If that does not seem to produce a good mix of attendees, then tweak it whichever way seems appropriate the next year.

My key point is to stop comparing a basket of mixed apples and oranges and start comparing apples with apples and oranges with oranges. That should give you mix of  the best apples and the best oranges.

Kerry