Apples to oranges. The WaPo dataset is behind a paywall so I cannot verify, but you used the term ‘police killings,’ whereas WaPo uses the term ‘police shootings.’ From previous interaction with the dataset I believe the WaPo source excludes non-firearm ‘police killings.’

You said yourself, suicide correlates with firearm ownership; I propose ‘police shootings’ likely correlates with the presence of firearms in police encounters which likely correlates to availability of firearms.

Furthermore, as somebody else mentioned, the time domain differs between the sets. You have different time periods, different classification criteria, potentially overlapping datapoints treated as distinct events. I do not believe any valid conclusions or observations could really be made from this presentation.

Oh I agree that the connections are tenuous and this was just a thought to get the ball rolling. That being said, I ought to clarify.

I am not saying Wikipedia was a ‘college website,’ rather at the time I would argue that HS/College-associated individuals made up the majority of people who would visit Wikipedia.

Likewise, I am not saying that FB expanding to the general public introduced the general public to Wikipedia; rather, the expansion of FB to the general public was more organic in that not everyone picked it up right away (the early adopters more closely resembled the initial user base of near-college aged individuals). This is where FB would serve as a nexus between ‘groups,’ not individuals. Thus, FB facilitated the growth of Wikipedia at this time, particularly among students.

My words are not perfect, but an example may do better:

Students at College ‘A’ use Wikipedia for research. Students at College ‘B’ do not know about Wikipedia. Through FB, a student at College ‘B’ learns of Wikipedia from an old friend at College ‘A.’ The student at College ‘B’ then tells his friends about Wikipedia, and eventually most Students at College ‘B’ know of Wikipedia.

In short: FB created a network exclusively for college students -> Initial Wikipedia users were likely college/university students -> FB then expanded to public users (mostly other teens/young adults) -> The knowledge of Wikipedia spread through FB networks -> The rapid growth of FB therefore contributed to the rapid growth/popularity of Wikipedia at that time

The connection to ‘edits’ is implicit; In general, the more people are aware of Wikipedia the more contributions/edits, all else equal. The timing of the outliers in question appear to coincide with the US academic calendar, whereby term and final papers are generally due at the end of the academic year (spring semester). Students, needing citations for information/sources included in their papers, may have been more likely to edit/amend Wikipedia pages than other Wikipedia users at that time.

If I had to fancy a guess, I would say 2007-2008 looks like it does due to Facebook and students. Some may remember that FB was initially restricted to college students with a student e-mail address from select colleges. Around 2007, FB started allowing anybody with a public email address join; the early adopters were mostly students at other colleges who previously did not have ‘TheFacebook’ for their college. There were a significant number of new HS users at this time as well.

February to May is Spring Semester, and at US schools this is often when the ‘Big,’ important final research papers are due.

My guess is that the rising popularity of FB (along with other social media/tech of the time) was related to the rising popularity of Wikipedia among an age cohort that was disproportionately sensitive to Wikipedia ‘sources,’ and thus more interested in adding/editing information on Wikipedia pages.

My guess is that this trend dies off quickly as the amount of awareness about Wikipedia and the potential for ‘bad actors’ to post untrue information becomes more widely known, and the educational institutions reject Wikipedia as a valid primary source.

Let it be known that I have nothing to back any of this up; I just came up with a guess while pooping.