English 680V: Visual Rhetoric in a Technological Age (Purdue)

The following is a list of the keywords I used when answering different parts of these exploratory questions:

Accessibility
Design
Digital library
Digital Library
Information Design
Online Writing Lab
Participatory Design
Usability (testing)
User-centered theory
Visual Design

On August 26, 2005, the Writing Lab launched its redesigned Online Writing Lab (OWL). During the spring semester of that academic year, Dr. Salvo administered a pilot test to his English 515 course that would test the usability of Purdue’s OWL and the usability of the testing materials themselves. My involvement with the OWL began as a student in the course, recording during the usability tests and refining the testing materials that would be used to conduct the subsequent usability testing. Beginning on February 27 and concluding on March 3, 2006, the OWL Usability Research Team conducted its first round of usability testing. After analyzing our first generation (G1) data, we immediately began planning for our next round of testing, which we conducted between July 10 and July 17, 2006. The testing materials we used for G2 were largely those we used during G1; however, based on initial data findings, we added into the test a user-centered prototype, which we tested in addition to our initial testing materials. Since completing G2, we have presented our findings at CCCC, CCCC Computer Connection, and Purdue’s TLT Conference, and we are in the planning stages of G3 and G4.

These three generations of usability testing has forwarded a very compelling argument for iterative usability testing as the best way to ensure a user-centered digital library, one composed of appropriate, interconnected, and successful visual and textual elements. The goal, then, for OWL designers is to redesign the OWL based on participant data and feedback and to continue testing to ensure continuous improvement toward an intuitive, searchable, accessible site.

First-hand experience in conducting usability testing has offered me very unique insights to the ways that users interact with web spaces. During the task-based test, where users were asked to locate pieces of information within the site, we discovered that some participants required extraordinary amounts of time to complete tasks, up to ten minutes, at which point participants were asked to proceed to the next task. Participants’ willingness to continue looking for information proved that our participants wanted to be genuinely helpful and were willing to work really hard to complete the tasks set before them. Additionally, they offered constructive suggestions toward improvement during an after-test questionnaire. Given the context of the usability testing situation, we can presume that participants were more committed to completing their assigned tasks than if they were trying to access information on their own; in other words, we cannot expect actual users to search for ten minutes in one site to locate a piece of information. Another benefit of usability testing enabled us to observe how users interacted with the site—their pathways, numbers of clicks, and amounts of time to complete tasks. Because the testing recorders were able to observe participants through every step of the testing, noting places where participants experienced particular difficulty or frustration completing tasks, we were able to speculate about user expectations. This enabled us to witness participant reactions while using the site. Furthermore, usability testing affords us the opportunity to offer actual OWL users a participatory role in OWL’s redesign process.

There exists a wealth of research in the areas of usability studies, web design, visual communication, and a host of others relating to this project. But as Richard Buchanan claims in “Design Research and the New Learning,” there exists “a profound, irreducible gap between scientific understanding in this area [user behaviors and interaction] and the task of the designer” (16). My attempt with this project is to rectify this information shortage by exploring some of the ways web designers and usability experts can construct more visually effective digital libraries. Currently, my driving question will be how we can usability test the visual components of a digital library. What follows is a quick encapsulation of some of the recurring points in the research.
• Baker forwards HATS, an acronym for headings, access, typography, and spacing that gives a straightforward system for thinking about elements of design. While I think his acronym provides good base guidelines for incorporating design, it does not seem to consider the ways that design intersects with usability.
• Buchanan claims that usability investigations require consideration of “human and cultural factors” (16). That a product works is not enough; “it must also fit the hand and mind of the person who uses it” (16). This passage captures some very central messages of our usability testing; we recognize that the redesigned OWL of August 2005 represented a designer’s vision. But as we have noted, and the designer has concurred, usability testing can refine elements of the site that are inaccessible to the user.
• AICP, standing for alignment, invitation, credibility, persuasion, forms the basis of Hassett’s piece. Similar to Baker’s piece, I think these guidelines hold value for students in professional or technical writing courses when beginning to learn about design principles, but it does not appear to hold relevance for usability studies.
A similar trend in much of this research is for authors to outline practices that they use to teach document design in their courses. However, there is very little work done in the area of how design can be measured in terms of accessibility and usability.

I think a very common misconception regarding usability testing is an underestimation of the value of iterative testing. The work we have conducted so far has uncovered tremendous insight into the behaviors of men and women. Additionally, our second, third and fourth generations of testing will test hypotheses developed in early stages of testing. This data allows us to join into the conversations about user-centered design; web design; usability method and methodology; findings based on gender, usage, native language; etc. Some common misconceptions regarding design are that design should be dealt with upon completion of text, that it is a one-time task, that it is static, and that it does not contribute significantly to success of a text. During our tests, participants responded particularly well to their chance to build their own sites using pre-set components of the site in varying sizes and shapes. One participant, in a radical move, discarded all the shapes presented and drew a prototype on a blank sheet of paper. This move offers a new avenue for further testing. As Kathryn Henderson said in “The Role of Material Objects in the Design Process: A Comparison of Two Design Cultures and How They Contend with Automation,” “Actor network theory lets us treat design documents—sketches, drawings, diagrams—as participants in the interactions that construct new technologies” (140).

The push toward information digitality has open new spaces for conducting research into the accessibility and usability of online spaces. The new generation of students is more comfortable searching for information online than perhaps any generation before it. But these students bring with them certain expectations about the ways that web spaces should be constructed and information should be accessed. More frequently, digital libraries are becoming prime sources of information ranging from recipes to dating. That said, we are faced with the challenge of presenting the most usable, more accessible OWL possible

Symptoms of poor design can stem from a number of issues. First, design can fail if it interrupts, contradicts, or detracts from the message of the content. As Michael J. Hassett claims in his article, clip art can destroy a text’s credibility, giving it the impression of being cheap or unsophisticated (66). Poor design can also result from a lack of user participation in the creation, testing, and revision of a text. In addition, the fact that idiosyncrasies in data can result from small or unrepresentative pools of participants presents another good argument for iterative rounds of testing.

One of the driving forces of Purdue’s OWL usability testing is to maintain its commitment to the mission of a land-grant state university. That is, Purdue should strive to make its information as accessible as possible. Failure to conduct testing on the site contradicts some of Purdue’s most central educational and ethical obligations. OWL designers have obligations to ensure the highest accessibility rates to the largest number of people. The only way to determine majority preferences is through iterative rounds of testing that will test hypotheses regarding user preferences of, behaviors around, and reactions to the site.

Highlights of which usability experts and designers should be aware follow:
• Users typically like, and expect, what they have seen. In Blokzjil’s “The Instructor as Stagehand,” students showed preferences toward certain aspects of powerpoint presentations and toward the use of laser pointers during presentations, when those certain elements or the use of the laser pointer was incorporated into the presentations they watched. This knowledge suggests that consideration of conventionalized forms should be considered during preliminary stages of design. If users are presented with design elements that contradict their prior experience, they are likely to express dissatisfaction.
• An iterative usability testing methodology is the only way to ensure accurate, representative data. Without it, inconsistencies could skew the data, and results could become outdated.
• The question facing me in this presentation, and the question I pose to designers and usability experts is to consider ways that we can begin to interpret data against the particular needs of users to create a usable, accessible web space supported by iterative usability testing and superior design.

Looking toward the future, I envision a shift in design research toward building and maintaining more effective, user-centered digital libraries. Web spaces are beginning to offer users customizable options to align their websites more closely with their unique preferences. However, perhaps more unexpected, I see this shift encompassing more design-based customizable options—color schemes, hierarchies, layout, visuals, etc. I see potential for huge advances to be made in the areas of visual design and usability studies.