Usability Testing Overview in Singapore
The concept of usability testing is to have a representative group of users typical of the product while observers and developers are watching, listening, and making notes. The product may be a website, software, or any other product that may not yet be formed. The test can be an early prototype test on paper, or a test of a finished product later.
Origin
Usability comes first from human factors. Ergonomics, also known as ergonomics, originated in World War II. Designers began to develop ergonomics by studying how to use machines and human capabilities when developing new weapons. This is a multidisciplinary subject, Including psychology, anthropometry, environmental medicine, engineering, statistics, industrial design, computer and so on.
ISO / IEC 9126-1 describes six aspects of software quality in product development, in order of functionality, reliability, usability, efficiency, maintainability, Portability. ISO / IEC 9126-1 defines “Quality in use” as a generalized goal: meeting the needs of target users and supporting users Quality, functionality, reliability, availability and availability determine the target user’s Quality of use, and support for users are concerned about the quality of maintenance and portability. Currently, ISO / IEC 9126-1 has two roles, first as part of a software design activity (definition of usability), and secondly as the ultimate goal of providing software to meet user needs.
Usability testing method
The so-called usability assessment, that is, the software “usability” to assess whether they meet the availability of standards. Currently, there are more than 20 usability testing methods, which can be divided into expert assessment and user assessment according to personnel involved in usability assessment. According to the software development stage in which assessment is conducted, the usability assessment can be divided into formative assessment and summative assessment. Formative assessment refers to the user in the software development or improvement process, users are required to test the product or prototype, after the test data collected to improve the product or design until the desired usability goals. The goal of formative assessment is to discover as many usability issues as possible and to improve software usability by fixing usability issues. The purpose of summative assessment is to cross-evaluate multiple versions or multiple products and compare the output assessment data. Website usability testing includes the steps of defining a clear goal and purpose, installing a test environment, selecting the right audience, testing and reporting the results.
Assessment method
Cognitive walkthrough
Cognitive Walkthroughs was proposed by Wharton et al. (1990), which first defines the target user, representative test tasks, the correct sequence of actions, the user interface for each task, and then rehearses the actions and raises questions, including Whether the user can establish the task to achieve the purpose, the user can obtain an effective plan of action, the user can use the appropriate steps, the user can assess whether the system based on the feedback to complete the task, the final comment, such as to achieve what effect, a Whether the actions are valid, whether a certain action is appropriate, and whether a situation is good. This method has the advantage of being able to use any fidelity prototype, including paper prototypes. The disadvantage of this method is that the evaluator is not a real user and can not represent the user well.
Heuristic evaluation
(Heuristic Evaluation) was proposed by Nielsen and Molich (1990). Multiple appraisers (usually 4 to 6 people) iterate through the various interfaces of the system based on the usability principle and independently evaluate the system, allowing each evaluator to discuss their own Find, find out the availability problem together. The advantage of this method lies in the fact that the expert’s decision is faster, the resource is less, the comprehensive assessment can be provided and the evaluation maneuverability is good, but there are also some shortcomings: First, it will be subjectively influenced by experts; secondly, there is no prescribed task, The third is to evaluate the latter part of the credibility of the evaluation as a result of reducing the credibility of the fourth is the assessment of experts and the user’s expectations there is a gap between the problems found only represent the meaning of experts.
User test method
User Test is to allow users to really use the software system, by the experimenter to observe the process of experiment, recording and measurement. This method can accurately feedback the user’s performance, reflecting the needs of users, is a very effective method. User testing can be divided into laboratory testing and field testing. Laboratory tests are conducted in the usability testing lab, where the field tests are viewed and tested by usability testers to the user’s actual use site.
After user testing, the evaluator needs to compile and summarize the data obtained during the test, such as the mean, median, range and standard deviation of the completed time, the percentage of successful tasks completed by the user, and the user’s various preferences for a single interaction Histogram representation and so on. Then analyze the data and write a final test report based on the severity and urgency of the problem.
Precautions
You test the product, not the user
For some users, “test” has a negative meaning. We need to work hard to make sure they do not think the test is for them. We want to make them understand that they are helping us to test prototypes or websites. In fact, we can not use the term “test”. Instead, we invite participants to help us, “Brave to prototype.” When users find it hard to get the job done, we should change the site, not the user. At the same time, we should also consider how well the site fits the typical user’s goals, rather than focusing on how well the user is doing on this task.
More rely on the user’s performance, rather than their preferences
Through testing , e can measure the user’s performance, as well as their preferences. The user’s performance includes whether the successful completion, the time spent, the error generated and so on. Preferences include user self-reported satisfaction and comfort. Some designers think that if their design caters to the user’s preferences, the user will perform well on the site. But the evidence does not support this. In fact, the user’s performance and their preferences do not correspond to the product. One study found that about 70% of users agree that performance and preferences are linked. In other words, they perform well on their favorite websites and perform poorly on websites they do not like. However, a relatively large proportion of people (30%) believe that users’ performance and their preference for products do not correspond to each other. They may perform well on sites they do not like, and may not perform well on their favorite sites. There are many explanations as to why people give higher ratings to websites that do not perform well. They may attribute poor performance to themselves, not websites. Or, they may worry that giving a lower rating hurts website designers, which is our feelings. Or, they may not have completed the task, but consider themselves successfully completed, they do not realize the problem. For all these reasons, we recommend that you: rely more on user performance than on their preferences.
Apply your test results
Usability testing is more than just a milestone for checking the progress of a project. You know, usability testing is not over when the last participant has done the job. The entire team must carefully study the results, set priorities, and modify the site or prototype based on the results.
Based on the user experience, identify the best solution to the problem
Making any product, including most websites and software, requires thinking about how many different users work, user experience, issue, and need. Most projects, including designing or revising websites, deal with time, budget and resource constraints. Balancing all aspects represents a major challenge for most projects. When you weigh the pros and cons, it’s best to prioritize websites or software that enable the most users to do their job. Studies have shown that the cost of supporting failed customers is significantly higher than the cost of product modifications at development time. You need to carefully consider the consequences of assuming users, usage scenarios, and usability tests in an attempt to find the ideal solution for different customer needs. Can not find the best solution, the user will not be able to complete the task smoothly. Evidence shows that even users who spend more time doing their job in a less-than-perfect product interface are far less likely to be successful in a better product interface.
Practicality
You can do usability testing whether you use formal or informal devices. With any type of device, you can use a variety of formal or informal methods.
The location of the usability test
You can have a valid usability test using any of these settings:
- Two or three-room fixed lab with audiovisual equipment
- Meeting room, user’s home or studio, with portable recording device
- Conference room, user’s home or studio, no recording device can also be replaced by eye observation and notes
- When the user can remote control in different locations
So, even if you do not or can not find a permanent lab, you should do a usability test. Do not say, “Because we do not have a usability lab, we can not do usability testing.” Just do it! You can do it in any space.
Usability tests need how many people involved
Look at the situation. A typical test requires 8 to 16 people (per user group). If each user will spend an hour, it means one to two business days per user group test. When your project is: * Prototype on paper or early development stage * Plan to test the entire development in a few rounds * Have a fairly consistent user base You may only need 4 to 6 if you help someone find a serious problem. * If you have different potential user groups (eg physicians, patients, researchers), you need user representatives for all these groups. If you have a requirement on the user’s computer operation or network experience, you also need to include less experienced and more experienced users. * If you want to conduct a formal quantitative test of your product or system, you will need more people for statistically meaningful results. For diagnostic usability testing, 6 to 8 users are usually not enough to expose most of the product’s problems. If you have been doing iterative (repetitive) usability testing during website development, there are many users participating in one or another version of the web test. So, although there are fewer than 10 test participants per availability test, you may need to take 15 to 30 people to test before the site is launched.
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 users expected, and the extent to which you expect the test to be formal. If you already have a standard test program and available materials and equipment, usability testing will be conducted quickly and very cheaply. If you or your user hiring company has a user database, the time it takes to recruit can be substantial and therefore less expensive.
Usability testing budget
These factors should be considered:
- The time it takes to plan: Identify the main questions of the test, the types of users that need to be tested, the screening questionnaire for the hiring user, and the test scenario.
- Recruitment costs: The time spent on the company, the cost to the hiring firm (usually a good choice), and the usability experts spend time familiarizing themselves with the site and their production team to design the appropriate test scenarios. If you need to document the testing process, It also costs you to rent a lab or a video camcorder.
- The amount of time the team spent watching the user (testing)
- Remuneration or gift paid to test participants
- Analyze audiovisual materials, find out the problems that exist, and how long it takes to recommend solutions
- Discuss with developers how long it will take to change and modify the program, write down the results of the survey, and report on the recommendations.
Remember, the budget analysis includes more than one usability test. The availability of a website (or product) is an iterative process. You’ll find that the budget used for several small tests during development is more valuable than just a large test at the end of the project. Website usability testing is to achieve cross-form visual consistency, including display, border and column layout when testing screen resolutions, color and size of the form, font used by the label, size of the button, heat used or Shortcuts, labels for controls such as animations / graphics used, buttons, the length of the text box in the same field, the format of the date and time fields.
read more:
