Wikipedia Statistical Survey

From Wikimedia Usability Initiative
Jump to: navigation, search

Metrics

The initial step is to define the metrics which are relevant to the usability of Wikipedia.

User Demographic Data

UNU-Merit and Wikimedia Foundation conducted Wikipedia Global Survey in October 2008. Among more than 130,000 Wikipedians who participated in the survey, the average age of survey participants was 25.8 years old and half of the respondents were between 18 and 30 years old. Average age of readers was 25.3 years old while contributors was 26.8 years old. Two-thirds of the respondents were male and one-third was female. During the usability study, we observed the difference how the study participants among different age group interacted with editing process of Wikipedia. The objective of the Wikipedia Usability Initiative is to remove the barrier from across all age groups and gender, and it will be insightful to see how the age distribution of users especially for editors change over time. Activity Types The data provided by UNU-Merit shows that 62.55% of Wikipedia users are readers. Occasional contributors account for 24.5% and active contributors account for 9.18%. Although there are many factors and motives for users to contribute, the usability improvement of MediaWiki should help to increase the share of contributors.

User Behavioral Metrics

Wikipedia Statistics has essential statistical data for the usability metrics. However the data dump of English is yet to be completed from October 2006. Once the data is available, the following metrics will help the usability team to evaluate the usability improvements of the software. It is important to track externalities, such as activity surge in specific topics, activity level of vandalism, activities by bots and etc.

  • Contributors Wikipedians who edited at least 10 times since they arrived.
  • New Wikipedians Increase in wikipedians who edited at least 10 times since they arrived.
10 edits seems high entry point to be considered as a new Wikipedian. Having a discussion with Erik Zachte in tracking granular frequency. (e.g. 3, 5 times)

Article Metrics

  • Edits per month Edits in past month (incl. redirects, incl. unregistered contributors, incl. bots)

Application Specific Metrics

Edit - As the first usability release (Acai) reorganizes the tabs, it will be good to capture if the click-through-rate of edit tabs go up or not.
Edit/Save - Usability study showed that some study participant just get lost in within “Edit” mode and never get to “Preview” or “Save” It is critical that we track the save rate.
Search
Location - Search location in the new skin “Vector” is moved from above the left navigation bar to right hand corner. The search usage need to be tracked.
Go vs Search - “Go” takes the inquiry directly to an article, and “Search
Toolbar usage
Left Navigation
Vector Opt-in rate

Quality of edits

It's not yet clear on how this may be measured yet. Please feel free to discuss this process with us.

Results

The results may be posted here. Ideally a comparison of the statistics over time will give some indication to the impact that changes to English Wikipedia have caused.

Example formats for results

The following are suggested examples of how collected data might be displayed.

Number of users who performed edits each month grouped by number of edits

Month Number of edits in one month
1 - 9 10 - 99 100 - 999 1000+
January 2009 # (%) # (%) # (%) # (%)
February 2009 # (%) # (%) # (%) # (%)