Blog

Online Card Sorting Pros and Cons

Posted by Brian

Online card sorting applications increase the accessibility of running a card sort. They make it fast, cheap, and easy (depending on the scope and scale of the sort) to set up the sort, and to recruit participants. Also, the participants can complete the sort at their convenience. Another major time-saver of using an online tool is that the data is recorded as the participant performs the sort, rather than having to be compiled and manually entered afterwards.

Some drawbacks of using an online tool for card sorting (as with using any online tool to replicate any manual process) are that the tool can impose its own constraints on the process, and there is also always a chance that technical issues will frustrate the participant or cause the participant to abort the process altogether. Also, an online sort doesn’t provide any verbalized cues about the participant’s organizational rationale. Although this could be overcome by using a card sort tool such as  OptimalSort with a screensharing or conferencing tool to run a remote moderated sort.

Garrett’s UX Planes

Posted by Brian

In his book The Elements of User Experience, Jesse James Garrett provides a conceptual framework for user experience design that encompasses five planes:

  • Surface (images and text the user sees)
  • Skeleton (placement of elements)
  • Structure (interaction design and information architecture)
  • Scope (range of content and features)
  • Strategy (organizational goals and user needs)

There are dependencies between these five planes, and these dependencies impact design choices.

I’ve always liked Garrett’s framework. I think he captures the essential components of UX, and organizes them into a visual structure that can be used to drive the design process. Easily understandable frameworks are always helpful.

There is a wide range of heuristics, methods, and techniques that can be applied to the design process within any given plane of Garrett’s model. But his framework allows practitioners who may have different specialties (graphic design, content production, marketing, IA, etc.) to develop a shared understanding of the overall process and an appreciation for the interdependencies of their work.

I also enjoyed Garrett’s discussion of different architectural approaches to structuring content. I thought he made some important points about the distinctions between top-down and bottom-up architectures. He explained that top-down approaches are driven by strategic concerns, and can sometimes gloss over important content features. Bottom-up approaches, on the other hand, are tailored to site functionality or content, but can make it difficult to accommodate future changes.

Surveys for UI Evaluation

Posted by Brian

What are some advantages and disadvantages of using survey documents for evaluating user interfaces?

People are used to taking surveys, so they are generally widely accepted. They are easy for managers and designers to understand, an inexpensive way to gather feedback, and can result in a large base of statistical data. Of course, if the survey isn’t well-crafted and properly targeted from the beginning, the resulting data won’t be worth much.

Surveys can reveal much more about a user’s real feelings than simple analytics or quantitative measurements. However, some responses can be biased. But overall, I think surveys are very effective as long as clear goals are established before designing the survey, and the questions are well thought out and clear. I also think it’s a good idea to test a survey on a small group before deploying it to thousands of users.

Personally, I generally don’t mind taking surveys. But I must say I always bail out of a website survey if I was told it would be “brief,” and then I’m ten minutes into it with no end in sight. For those types of random, pop-up type surveys, I think it’s important to let the user know exactly what to expect.

On Norman

Posted by Brian

Don Norman’s The Design of Everyday Things is a classic in the design world. He uses simple case studies and proposes simple principles that we all already know through the mundane experience of existing and interacting with objects and artifacts.

We are people. We use things. Sometimes these things don’t work the way they should, or the way we think they should, so we need people like Don Norman to explain why.

Sometimes good design is a product of research and a deep understanding of form and function, and sometimes it’s just a happy accident. We can study it forever, but sometimes good design is nothing more than a minor modification of something that almost worked in the first place. That’s called innovation.

I love Norman’s book because he simply and beautifully illustrates some great points, and he coins a few good words that have become part of the design and user experience lexicon.

I don’t know if Norman’s book ever made me think differently about how things are designed, but it has certainly served to justify my outrage about poor design. Whenever I open a “re-sealable” package that doesn’t actually re-seal, I think of Don Norman. If nothing else, Norman got a lot of people thinking that even the simplest things that frustrate us because of their inherently poor design can be improved, if we bothered to care about improving such things.

Elements of Informed Consent

Posted by Brian

Informed consent is based on the idea that research study subjects should be given sufficient information needed to make an informed and rational decision regarding their participation in the study. The principles that form the basis of informed consent are respect for persons, beneficence, and justice. These principles were established with the publication of the Belmont Report, published in 1979 by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research [Lazar et al. 2010].

Participants should fully understand the reason for the study, the procedures involved in the study, the risks of participation, and how they can obtain more information about the study. Additionally, participation in the study should be voluntarily and participants should not be coerced in any way [Lazar et al. 2010]. There are three fundamental components of informed consent: voluntariness, comprehension, and disclosure. Voluntariness implies that the participant has not been unduly influenced by researchers. Comprehension assumes that participants have the capacity to make an informed choice about their participation. Disclosure requires researchers to inform participants about the purpose of the study, any inherent risks, potential benefits, research alternatives, confidentiality, compensation in the event of injury, how participants can get more information about the research, and the conditions of participation [NIH 2008].

Lazar et al. suggest a general template for informed consent documents, to include the following sections: title and purpose, description of procedures, duration of the study, risks, benefits, alternatives to participation, confidentiality, costs and expenses, participants’ rights, and contact information if participants have any additional questions or concerns [2010]. Many institutions follow this general format for their own informed consent documents, although they may address additional concerns. For example, some additional elements that Iowa State’s Institutional Review Board suggests addressing when appropriate include: currently unforeseeable risks; circumstances whereby the subject’s participation could be terminated without the subject’s consent; the consequences of the subject’s decision to withdraw from the research, and the procedures for subject withdrawal; and procedures for providing significant new findings to the subject [2010].

Also, it is important that informed consent information is communicated using simple language that is easily understood by the subject, and using whatever means required to convey the information clearly.

REFERENCES

Iowa State University Office for Responsible Research. 2010. Elements of Informed Consent. Iowa State University, Ames, IA.

Jonathan Lazar, Jinjuan Heidi Feng, and Harry Hocheiser. 2010. Research Methods in Human-Computer Interaction. Wiley, Chichester, West Sussex, UK.

NIH Office of Extramural Research. 2008. Protecting Human Research Participants. National Institutes of Health, Bethesda, MD.

The Future of IA

Posted by Brian

In a recent interview, Peter Morville made some salient points that addressed the importance of medium-independence for information architecture, incorporating information flows and feedback loops into information ecosystems, facilitating understanding, and developing information architectures that are based on a deeper awareness of culture and cognition.

Use context, organizational goals, user needs, and incorporating appropriate mental models and metaphors have always been an important part of information design. But when considering the future of wearables, ubicomp, and the Internet of Things, information architecture will become increasingly complex.

Information architectures will have to be much more scalable and flexible than in the past. Also, there will likely be both visible and invisible information architectures. Practitioners will still develop traditional information architectures that support users interacting with a visible interface. But they will also have to manage architectures that support users performing information-seeking tasks that don’t require a visible interface, and that also support devices interacting with other devices.

More and more, physical spaces are becoming information spaces. This will make it increasingly difficult to manage the movement of users and devices through these blended environments using traditional information architecture implementations.