Layout Testing with Greeked Pages
Background
Page layout, content and navigation are not always designed at the same time. It is often necessary to work through at least part of these processes separately. As a result, it may not be possible to test layouts with realistic content until a relatively late stage in the design process, meaning that usability problems relating to the layout may not be found at the appropriate time.
Various solutions exist for this problem. One is the possibility of testing early prototype layouts containing 'greeked' text - that is, the 'lorem imsum' placeholder text commonly used for layout design [1]. A method for testing the recognisability of page elements was discussed in Neilsen's Alertbox back in 1998 [2], though the concept originated with Thomas S. Tullis [3].
Technique
Testing will require several users - around six is helpful without being excessively time-consuming. Ensure that they have not seen or discussed the layouts before the test! First, create a list of elements that should be visible upon the layout. Nielsen provides a list of nine standard elements that are likely to present on all intranet pages - but in your particular case you may wish to alter this list a little to encompass all of the types of element present on this template.
Give each test user a copy of each page - in random sequence, to eliminate any systematic error that might result from carrying the experience with the first page through to the second. Ask the test user to draw labelled blocks around the parts of the page that correspond to the elements you have identified. Depending on circumstances, you may find that encouraging the user to 'think aloud' may provide useful information, but be careful not to 'lead' the user to a preferred solution.
Finally, ask the user to give a simple mark out of ten for 'appeal'. This is not a very scientific measure, but is nonetheless of interest since this allows you to contrast the user's subjective measure of preference against the data that you have gathered (the number of elements correctly identified). Nielsen points out that the less usable page is often given a higher average mark by the user.
Scoring The Test
With the information provided, draw a simple table:
| Layout | Correctly Identified Page Elements | Subjective Appeal |
|---|---|---|
| 1 | N% (eg. 65%) | # (e.g. 5/10) |
| 2 | M% (eg. 75%) | # (e.g. 6/10) |
This provides you with a basic score. You will probably also find your notes from think-aloud sessions to be very useful in identifying the causes of common misunderstandings and recommending potential solutions.
When Should Page Template Evaluation Be Carried Out?
This technique can be applied on example designs, so there is no need to create a prototype Web site; interface ideas can be mocked up using graphics software. These mockups can be tested before any actual development takes place. For this reason, the template testing approach can be helpful when commissioning layout template or graphical design work. Most projects will benefit from a user-centred design process, an approach that focuses on supporting every stage of the development process with user-centred activities, so consider building approaches like this one into your development plans where possible.
Conclusions
If a developing design is tested frequently, most usability problems can be found and solved at an early stage. The testing of prototype page layouts is a simple and cheap technique that can help to tease out problems with page layout and visual elements. Testing early and often can save money by finding these problems when they are still cheap and simple to solve.
It is useful to make use of various methods of usability testing during an iterative design and development cycle, since the various techniques often reveal different sets of usability problems - testing a greeked page template allows us to separate the usability of the layout itself and the usability of the content that will be placed within this content [2]. It is also important to evaluate issues such as content, navigation mechanisms and page functionality, by means such as heuristic evaluation and the cognitive walkthrough - see QA Focus documents on these subjects [4] [5]. Note that greeked template testing does look at several usability heuristics: Aesthetic & minimalist design and Consistency and standards are important factors in creating a layout that scores highly on this test.
Finally, running tests like this one can help you gain a detailed understanding of user reactions to the interface that you are designing or developing.
References
- Lorem Ipsum Generator,
<http://lorem-ipsum.perbang.dk/> - Testing Greeked Page Templates, Jakob Nielsen,
<http://www.useit.com/alertbox/980517.html> - A method for evaluating Web page design concepts, T.S. Tullis. In ACM Conference on Computer-Human Interaction CHI 98 Summary (Los Angeles, CA, 18-23 April 1998), pp. 323-324.
- Introduction To Cognitive Walkthroughs, QA Focus briefing document no. 87,
<http://www.ukoln.ac.uk/qa-focus/documents/briefings/briefing-87/> - Heuristic Evaluation, QA Focus briefing document no. 89,
<http://www.ukoln.ac.uk/qa-focus/documents/briefings/briefing-89/>
Further Information
- Getting Started With User Centred Design, UsabilityNet,
<http://www.usabilitynet.org/tools/gettingstarted.htm>
