User:Jeblad/stats 2016-01-27

The number of wikipedians and edits per capita in Norway is still high (77+7 per million speakers), only comparable to countries like Estonia (88), and Iceland (56). Israel (126) is well above us, but we are far above that of the large language communities like English (20), French (20) and German (38). Every month, there are 1500-1700 editors making changes to the Wikipedia in Norway, with 10-15 % of these wikipedians at the Nynorsk version and the rest at Bokmål. As of this writing (6. Feb. 2016) there are 1697 active users at Norwegian Bokmål, up from 1630 active users at last report (28. Sept. 2015), 227 at Norwegian Nynorsk (up from 176), and 23 at the Sami Wikipedia which has virtually no edits anymore (up from 18). Some of the deviation from previous numbers can be season variations.

In 2015 the Wikipedia versions in Norway had a continued drop for new users, a stagnation in the drop for less active users, and a slight increase for the important core cohort with more than 100 edits. In December 2015 there were 360 contributors with more than 5 edits at Norwegian Bokmål [1] and 33 contributors with the same amount of contributions at Norwegian Nynorsk [2]. One year earlier the same numbers were 348 and 34, that is we have slightly better situation now. Two years earlier the numbers were 409 and 39, that is slightly better than today. The continuous fall from previous years has at least slowed down for the core cohort, and it has perhaps turned into a small increase.

The provided graphs shows that the drop for contributors with more than 100 edits for Norwegian Bokmål is about to gain a small growth, yet it is still too early to say anything with confidence. We would need statistics for a couple of years to say something with real confidence.

One possible explanation could be that a drive to add sources to the articles have triggered some interest from previous contributors, and that we now observe this additional activity. The bot adding the template about missing sources triggered emails sent to the contributors, and the previous contributors were then adding edits which we now observe. This is similar to the attempts by user Netrom to trigger contributions with the SuggestBot (AnbefalingsBot at nowiki).

New users with edits drops more than the contributors with more than 5 edits, and this indicates that we are hit by some external effect. We guess that it might be an increased amount of contributions on social media like Facebook. That use a considerable amount of peoples spare time, creating a ripple effect into Wikipedia.

If we reads the graphs a bit speculative it seems like the cohort with 5 edits or more is about to flatten the same way the graph for cohort with 100 edits or more did. That cohort has an influx of contributors from the new ones, and as long as that falls the cohort with 5 edits will probably be low, and the fall does not slow down at the moment.

Other interesting numbers are the amount of editing at Wikidata that impacts the Norwegian Wikipedias. Over 5000 edits impacting the article subjects as of this writing,[3] 2335 edits (46.7%) at Bokmål and 4359 edits (87.2%) at Nynorsk emerged from Wikidata. Unfortunately we are not able to utilize all edits on Wikidata, as the number of proficient programmers in the Norwegian community is to low. In comparison the number of categorizations at the Bokmål Wikipedia is 932 edits (35.0%) of the changes emerging on the wiki itself. The same number on Nynorsk Wikipedia is 303 edits (47.3%) of the changes emerging on the wiki itself. Categorization is somewhat similar to the edits on Wikidata, and can be viewd as an indication of how much impact Wikidata should have given that there were no other users editing Wikidata then the community from the respective Wikipedias.

We don't know what the expected number should be, but at least at Nynorsk Wikipedia the impact from Wikidata is very large. That is probably to be expected as the international community is comparably much bigger as seen from nnwiki than nowiki. As a corollary to this it could be said that it is even more important for small wikis to have templates and modules to reuse data from Wikidata.

2003-2015: Edits on Nynorsk, Bokmål and Nordsamisk Wikipedia The following graphs shows the statistics, but processed and visualized to clarify the trends. The blue lines shows the numbers "as is" from from the stats. The red line shows the 12 month moving average, taken from the previous 12 months. The yellow line shows a 12 month difference over the 12 month moving average. Because we use a moving average taken from the previous 12 months, the effects seems to be delayed and smoothed. This is as expected, and can be interpreted as an increased confidence in the numbers.

Norwegian Nynorsk Wikipedia the smaller of the two Norwegian projects. Important thing to note is the drop for contributors with more than 100 edits that starts in the spring 2013. We think this can be attributed to Wikidata, and the IW-bots stopping maintenance of the IW-links.

Norwegian Bokmål Wikipedia the larger of the two Norwegian projects. An important thing to note is that the drop seems to have stopped for users with more than 100 edits, and the cohort is again increasing in size. The strong peaks at the number of monthly edits are cleanups done by a bot.

Northern Sami Wikipedia which is the smallest project within our area. As seen from the daily contributions the edits have died away the last two years. It is not obvious how this project can be revitalized.

Notes and references edit

  1. Wikipedia Statistics Norwegian
  2. Wikipedia Statistics Nynorsk
  3. Numbers extracted from "Recent changes" slightly before 11th February 2016 12:23 (UTC)