Extension:InlineEditor/Usability study 1

From MediaWiki.org
Jump to navigation Jump to search

Usability testing[edit]

The test of November 16th 2010 in Groningen has been done with two groups: first year and master students of Information Science. These groups are heavily biased towards young, bright, and technological skilled people. However, it should indicate some of the more obvious problems of all the tested editors. Each participant has been asked to perform a few tasks in a 15 minute think-aloud experiment. The screen and voice has been recorded using an open-source screencapture program, and published here on Wikimedia Commons. For this a contract has been signed by all participants. The name of the participants will not be published.

The test has been conducted on a private wiki filled with the page of Groningen taken from the Dutch Wikipedia. This article has been shortened (as it is quite long) and optimized for the current algorithms of the editors.

There are two new editing environments that are tested, along the current, original editing interface. The first is one where the user can choose to edit based on different functionality: "Text" (sentences), "Lists", "References", "Media" (images, etc.), "Templates", "Full page". The advantage of this is to be able to give help or an optimized editing interface for each of these different functionalities. The disadvantage is that the user will not learn to edit larger parts of wikitext.

The other approach is to differentiate based on size: "Sentences", "Paragraphs", "Sections", "Full page". The advantage of this is that users encounter more wikitext whenever they move up a level, but they have to do more research themselves as on how wikitext works.

As it is a small-scale test, quantitative measurements are be unreliable. Even more so because of the think-aloud setting. Because the user will talk during the process, tasks are done at a different speed than usual. Therefore, only qualitative measurements have been used, such as the way a user completes a task, what kind of problems the user runs into, how the user feels while doing a task, etc.

# Gender Education Editor Wikipedia viewing Wikipedia editing Personal suggestions
#1 Female Graduate Original almost every day never edit button also besides first section
#2 Female Graduate Blocks never full editor; WYSIWYG
#3(a)(b) Male Graduate Functional once a week never making the adding/managing of references easier
#4 Male Graduate Blocks once a day never
#5 Male Graduate Functional every week never better terminology; less content; less guidance / text to read
#6 Male Undergraduate Functional few times a week never
#7 Male Undergraduate Blocks once a day never
#8 Male Undergraduate Original 2 times a month 1 time syntax highlighting
#9 Male Undergraduate Blocks often never edit mode bar larger; per sentence editing is great; full editor isn't that bad when you use it often
#10 Male Undergraduate Blocks 5 times a week 1 time making the adding of references easier