Consensual study: Measurement of aesthetic response to different visual design

Author: Christian Rohrer

Introduced

I recently published an article in Jakob Nielsen's Alertbox aspires, which mentions a method of "measuring beauty", and many people let me talk more about "consensual research". Consensual research is not just about measurement, because we can also use them to inform or enlighten you about the direction of different visual designs that you might be considering. In this article I refer to a methodology graph of user studies, which is defined in this diagram as an attitude study that can be qualitative or quantitative (as shown in the figure below, which is a hybrid method below the middle of the diagram).

Subjective issues

Consensual research is not well known, although visual design is very important to the user interface. Paul Howe once wrote, "I've done a lot of tests on aesthetic preferences, but it often makes me think it's the least scientific part of my user research." This "less scientific" perception of visual design comes from the following reasons.

It is partly due to the way in which visual design is used, and it often comes from the belief that the designer can best evoke the desired response. But what I want to say is that this approach is not the only one, not even the biggest problem in creating the design direction-it's the performance of the design direction and it's often done in the most subjective way. For example, a 10243.html "> visual designer might say that when presenting a different design direction to decision makers:

"I think the C design is better than A and B." Because I think the C design triggers the right emotional feedback from our users, it's the most important brand features, trust and fun, closer to "

The problem with this type of presentation is that it defines the decision as a subjective interpretation of the design through internal consistency, and that the decision maker may feel that his or her explanation or instinctive response to the design is just as effective as the designer. And, at least in the sense, they are both single individuals. Even if designers have years of work experience and know which type of design can trigger which type of response, the right of decision-makers is to evaluate, make the two consistent, and, if not agreed, will often become a stalemate, not to make things in a positive direction.

Here, consensual research can help solve two problems:

1, inform the design team, why the different design direction can trigger the target user specific response (in order to improve the design direction)

2, accurate measurement of specific adjectives (such as brand characteristics) of the visual design direction, so as to help make the final decision

This makes the subjectivity exist where it should be: The target user's view of the design, not the designer or the decision maker. This allows designers and decision makers to make a choice.

Another question is, "How can we predict real-world behavior?" "The answer is that we can't do this with consensual research because it's a measure of attitude, not a measure of behavior (as we all know, what people say and do is often very different)." You never expect to rely on a fundamental attitude study to predict real-world behavior, but desirability is thought to occur in the initial phase of people interacting with the product, not in the next phase. The latter is more affected by the stable interaction design and finally satisfies the basic user requirements. The positive (and subjective) aesthetic response helps your target users start using your site, but assuming you don't make any particularly bad choices (such as poor comparisons or small goals), it may not affect overall ease of use too much.

Types of consensual studies

There are two types of consensual studies: qualitative and quantitative. In qualitative studies, the subjects were taken to a laboratory or conference room and presented with different visual design orientations (e.g., mood boards or high-level designs), or visual design interfaces (e.g., high-fidelity prototypes or home pages). Here are 3 examples of different visual design directions from Yahoo! The personal desirability of the study was Jeralyn Reese and Michelle Reamy a few years ago.

At this time, the subjects were given some index cards, each card has a description (usually an adjective), then, the participants were asked to indicate which card is the best match with which design. Here are some examples of some of the descriptions on the card:

Easily attainable agreeable hindrance of patronizing stressful

An attractive and easy-to-use personal time consuming

Noticeable effective high quality predictable save time

Busy authorized variable relevant too professional

The exciting threat of cooperation reliable trustworthy

Complicated, familiar, seductive, inflexible, uncontrollable.

A fast, seductive, simplistic, non-traditional

A bewildering, flexible, worthless, dull, unpredictable

There are links to fresh organized tips available

Customizable interesting unstoppable direct valuable

After the participants selected some cards for each design, the researchers asked them why they made the choice. The main benefit of this approach is to find out why specific designs cause specific reactions. This provides the design team with the material they need to improve the next version.

Because the sample size is too small, and the recruitment of the acting may introduce additional deviations, this is not a suitable quantitative research method. The quantitative study of desirability is a qualitative study presented several years later. The method is to implant various types of images with a design orientation in a questionnaire. This allows researchers to obtain larger, more representative target user samples, often with more predictable results.

Previously mentioned Yahoo! A quantitative approach is used here in the study of personal desirability. This is very important because changing visual designers is a very big decision: This is the first time this company is willing to change Yahoo! Channel Design (mail, personal, economic), and compete with the visual design style of its competitors, rather than continue to use the general Yahoo! that can be found online Design style. (The 3 designs were finally won).

Presenting the results of a consensual study

Just doing research is not enough, especially when there is disagreement. An easy way to render results is to use Venn graphs. Let's say we have three design orientations, "simple", "fashion", and "fun," and the result profile looks like this:

Even so, it is important to show more details of the method, such as a significant test of the findings. Mike Katz,rian Van der Merwe and Christina Hildebrand have pushed the consensual research approach deeper in ebay. They used paired comparisons in their surveys, which allowed them to present their results as follows:

In another study, they presented the results of a significant test:

Publication of consensual research papers

The consensual study was first presented by Microsoft's Joey Benedek and Trish Miner in the UPA2002 annual article, named "Measurement desirability: A new approach to assessing acceptability in a usability lab environment". Later, Microsoft presented a case study using the method on CHI2004 to explain how they modified the method to apply the revised focus group discussion in this qualitative research approach. As far as I know, the quantitative research version of consensual research has been published.

Summary

In general, consensual research is a great way to understand beauty and the visual design direction you may be considering, and ultimately to measure how specific a design triggers a specific response. This is useful when you try to make a good first impression on your target audience and want them to have a deeper interaction with your site or product so they can find out if it reaches their intrinsic needs. If everything goes well with other aspects of the user experience, they become a loyal user.

Original address: http://www.xdstrategy.com/2008/10/28/desirability_studies/

Source Address: http://www.psycb.com/%E3%......%8E%E6%84%9F/