What Is a Usability Evaluation?

The concept of usability testing is to let a representative group of users perform typical operations on the product, while observers and developers observe, listen, and record.

Usability test

Right!

The concept of usability testing is to let a representative group of users perform typical operations on the product, while observers and developers observe, listen, and record.

The product may be a website, software, or any other product, and it may not have taken shape yet. The test can be an early paper prototype test or a post-production test.

A clear explanation of availability is provided in IS0-9241-11: Extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use.

Chinese name

Usability test

Earliest source

Human Factors Engineering

begins at

WWII

First record

1981

Usability originated from human factors. Human factors engineering

The first documented usability test appeared in 1981. then

ISO / IEC 9126-1 defines usability as "the ability of a software product to be understood, learned, used, and attracted by users in specific use scenarios" [ISO / IEC 9126-1. Software engineering-Product quality-Part 1 : Quality model [S]. International Standards Organization, 2001.]. ISO / IEC 9126-1 states that

The so-called usability evaluation is to evaluate the software "availability" and check whether it meets the usability standards. There are currently more than 20 usability assessment methods. According to the division of personnel involved in usability assessment, it can be divided into expert assessment and user assessment. According to the software development stage of the assessment, usability assessment can be divided into formative assessment and summary assessment. Formative evaluation refers to asking the user to test the product or prototype during the software development or improvement process, and use the data collected after the test to improve the product or design until the required usability goal is reached. The goal of formative evaluation is to find as many usability problems as possible, and to improve the usability of the software by fixing usability problems. The purpose of summary evaluation is to horizontally evaluate multiple versions or multiple products, and output evaluation data for comparison. Website usability testing includes the steps of defining clear goals and objectives, installing a test environment, selecting the right audience, testing and reporting results.

You test the product, not the user

For some users, "testing" has negative implications. We need to work hard to make sure they don't think the tests are directed at them. We want them to understand that they are helping us test a prototype or website. In fact, we can not use the term "test". Instead, we invite participants to help us, "the courage to try prototypes." When it is difficult for users to complete tasks, we should change the website, not the users. At the same time, we should also think about the extent to which the site can meet the goals of typical users, rather than focusing on how well the users are doing this task.

Rely more on user performance than on their preferences

Through testing we can measure the performance of users and their preferences. The user's performance includes whether it completed successfully, how long it took, errors that occurred, etc. Preferences include user self-reported satisfaction and comfort. Some designers believe that if their design caters to users' preferences, users will perform well on the site. But the evidence does not support this. In fact, user performance and their preferences for products are not one-to-one. One study found that about 70 percent of users agree that performance and preferences are linked. In other words, they perform well on the sites they like and poorly on the sites they don't like. However, a relatively large percentage of people (30%) believe that user performance and their preferences for products do not correspond one-to-one. They may perform well on sites they do not like, and may perform poorly on sites they like. There are multiple explanations for why people give higher ratings to websites that perform poorly. They may attribute the poor performance to themselves, not the website. In other words, they may worry that giving a lower rating will hurt the website designer, that is, our feelings. In other words, they may not complete the task, but think they have completed it successfully, and they are not aware of the problem. For all these reasons, we suggest you: rely more on the performance of users than their preferences.

Apply the test results you have

Usability testing is not just a milestone for checking project progress. You need to know that usability testing is not over when the last participant completes the task. The entire team must study the results carefully, set priorities, and modify or prototype the site based on the results.

Find the best solution based on user experience

Making any product, including most websites and software, requires thinking about how many different users work, experiences, issues, and needs. Most projects, including designing or modifying websites, deal with time, budget, and resource constraints. Balancing all aspects is a major challenge for most projects. When you weigh the pros and cons, it's best to prioritize those websites or software that will get the most users to complete the task. Studies have shown that the cost of supporting failed customers after product launch is much higher than the cost of product modification during development. You need to seriously consider the assumptions of users, use cases, and the results of usability tests, trying to find the ideal solution for different customer needs. Without the best solution, users will not be able to complete the task smoothly. Evidence shows that even if users use extended time to complete tasks in a less perfect product interface, it is far less successful than in a better product interface.

You can do usability tests on both formal and informal devices. With any type of device, you can use a variety of formal or informal methods.

Usability testing location

With any of the following settings, you can perform effective usability tests:

* Two or three-room stationary laboratory with audiovisual equipment

* Meeting room, user's home or studio, equipped with portable recording equipment

* Meeting room, user's home or studio, can be replaced by human eyes and notes without recording equipment

* Can be remotely controlled when the user is in different locations

Therefore, even if you do not have or cannot find a fixed laboratory, you should perform usability testing. Don't say, "Because we don't have an usability lab, we can't do usability testing." Just do it! You can do it in any space.

How many people are involved in usability testing

It depends. A typical test requires 8 to 16 people (per user group). If each user will spend an hour, it means that the test of each user group takes one to two working days. When your project is: * Prototype on paper or early development stage * Plan to test the entire development through several rounds * Have a fairly consistent user base If you only need people to help identify serious problems, you may only need 4 to 6 people. * If you have different groups of potential users (eg doctors, patients, researchers), you need user representatives for all of these groups. If you have requirements for users' computer operation or network experience, you also need to include less experienced and more experienced users. * If you are going to conduct a formal quantitative test of your product or system, you will need more people to get statistically meaningful results. For diagnostic usability tests, 6 to 8 users are usually not enough to reveal most of the problems of the product. * If you have been doing iterative (repetitive) usability testing during the website development process, many users will participate in one or the other version of the website test. Therefore, although there are fewer than 10 participants in each usability test, you may need 15 to 30 people to participate in the test before the website launch.

How much does it cost to do usability testing? The cost depends on the size of the site, the amount of your test, the number of user types expected, and how well you expect this test to be regular. If you already have a standard test procedure and available materials equipment, usability testing will be performed quickly and cheaply. If you or your user-recruitment company have a user database, the time spent on recruiting can be greatly saved, so it will cost less.

Usability testing budget

These factors should be considered:

* Planned time: Determine the main questions for the test, the types of users who need to be tested, the screening questionnaire for the recruited users, and the test scenarios.

* Recruitment cost: the time of the company personnel, the cost to the recruiting company (usually a good choice), usability experts need to spend time to familiarize themselves with the website and its production team, design the corresponding test scenarios, if you need to record the test process It also costs rent for a laboratory or a portable camera.

* Time spent by team watching users (for testing)

* Compensation or gift to test participants

* Time spent analyzing audiovisual materials, finding problems and recommending solutions

* Time spent discussing changes and modifications with developers, writing survey results and suggesting reports.

Remember, budget analysis involves multiple usability tests. Building the usability of a website (or product) is an iterative process. You will find that the budget for several small tests during development is more valuable than just one large test at the end of the project. Website usability testing is to achieve cross-form visual consistency, including testing the display when the screen resolution changes, the margins and column layouts, the color and size of the form, the font used for the label, the size of the button, the heat used or Shortcuts, labels used for animation / graphics, buttons and other controls, length of text box in the same field, format of date and time fields. ^[1]