The concept of usability testing is to let a group ofRepresentativenessUsers perform typical operations on the product, andobserverWatch, listen and take notes with developers.
The product may be a website, software, or any other product, and it may not have formed yet.The test can be an early paper prototype test or a later product test.
In IS0-9241 - 11 forusabilityExtent 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.
Availability originated from human factors.Human Factors EngineeringErgonomics（ergonomics）, Originated fromWorld War IIWhen designing new weapons, designers studied how to use machines, human capacity limits and characteristics, which gave birth to ergonomics, a discipline involving many fields, including psychologyAnthropometry、Environmental medicine、engineering, statisticsindustrial design, computers, etc.
Historical development
Announce
edit
The first documented usability test occurred in 1981.at that timeXeroxSubordinatePalo Alto Research Center An employee of the company recorded the process of introducing usability testing into the development of Xerox Star Workstation (Xerox 8010 Information System).However, since there are only about 25000 sets in total, Xerox Star system is considered as a typical business failure case.
In 1984, Intuit, an American financial software companyInc.In itsPersonal financial managementDuring the development of software Quicken, usability testing is introduced.Suzanne E. Taylor mentioned in his 2003 industry bestseller Inside Intuit that "in the first usability test example, this practice has later becomeIndustry practiceLeFevre gathered some people from the street to test Quicken at the same time. After each testProgrammerCan improve the software. "One of the founders of Intuit IncScott CookHe also said, "We did usability testing in 1984, five years earlier than others. It's different from the usability testing in the sold population, and it's routine and coreDesign processOne of the links is also very different.
After more than 20 years of development and application, usability testing has become an essential and important link in all stages of product (service) design, development, improvement and maintenance.Its value lies in finding possible problems in products (services) early in the initial stage, and providing improvement plans before development or production, thus saving designdevelopment cost 。When the sales of products (services) are weak or problems occur in the use process, but the key problem cannot be found in a timely and accurate manner, usability testing can greatly improve the efficiency of problem solving.Through the usability test, we can not only know the degree of users' recognition of products (services), but also know some implicitUser behaviorlaw.
international standard
Announce
edit
ISO/IEC 9126-1 defines availability as ",software product The ability to be understood, learned, used and attracted by users "[ISO/IEC 9126-1Software engineering – Product quality – Part 1: Quality model[S]. International Standards Organization, 2001Product development processinsoftware qualityThe six aspects of, in turn, functionality, reliability, and availability（usability）、Effectiveness(efficiency), maintainability（maintainability）, portability.ISO/IEC 9126-1 will“Service quality(Quality in use) "is a broad goal: to meet the use quality of target users and support users. Functionality, reliability, effectiveness and availability determine the use quality of target users in specific scenarios, while support users care about the quality of maintainability and portability.At present, ISO/IEC 9126-1 has two functions, first as a concretesoftware design Part of the activity (usability definition), followed by providing software to meetuser demand The ultimate goal of.
international standardISO 9241-11 Usability is defined as "a specific user can effectively, efficiently and satisfactorily use the product to achieve a specific goal under a specific use scenario" [ISO9241-11Ergonomic requirements for office work with visual display terminals (VDT's) – Part 11: Guidance on usability[S]. International Standards Organization, 1998.] ISO 9241-11 summarizes availability into three aspects:Effectiveness(effectiveness), the accuracy and completeness that users achieve by using the system to complete various tasks;Efficiency: resources consumed by users to complete tasks according to accuracy and integrity, including intelligence, physical strength, time, materials oreconomic resources ；Satisfaction(satisfaction), the user's subjective reaction to the use of the system, describing the comfort and acceptance of using the product.
Nielsen (1994) believed that the utility andusability(usability) constitutes the factor whether the system can be used to achieve specific goals, which is called usability [Nielsen J. Availability Engineering [M]Liu ZhengjieBeijing:China Machine Press，2004：16-24．】。Availability is defined as "whether users can use it wellFunctions of the system”, divided into five factors: learnability, users can use the system to complete relevant tasks in a short time;Efficiency: After users learn to use the system, they can use the system efficiently;canMemory（memorability）, users can still use the system after not using it for a period of time;Errors. Users can make fewer errors when using the system. The system must preventCatastrophic errorhappen;Satisfaction(satisfaction), users are subjectively satisfied with the system.
Shackel (1991) defined usability asFunctional characteristicsThe system can be easily and effectively used by specific user groups. After specific training and user support, the system can complete tasks in a specific range in a specific environment. "The availability is divided into four factors: effectiveness, learnability, and flexibility（flexibility）, attitude.
Test method
Announce
edit
The so-called usability evaluation is to evaluate the "usability" of software and check whether it meets the usability standards.At present, there are more than 20 usability evaluation methods, which can be divided into expert evaluation and user evaluation according to the personnel involved in usability evaluation;According to thesoftware developmentAt the stage, usability evaluation can be divided into formative evaluation and summative evaluation.Formative assessment refers to inviting users to test the product or prototype during the process of software development or improvementImprove productOr design until the required availability target is reached.The goal of formative assessment is to find as many usability problems as possible, and improve software usability by fixing usability problems. The purpose of summative assessment is to horizontally evaluate multiple versions or products, and output evaluation data for comparison.The website usability test includes the following steps: defining clear goals and objectives, installingtesting environment, select the appropriate audience, test and report the results.
Evaluation method
Announce
edit
cognitive rehearsal
(Cognitive Walkthroughs) was proposed by Wharton et alTarget user、RepresentativenessTest task, correct action sequence of each taskuser interfaceAnd then rehearse the action and constantlyAsk questionsIncluding whether the user can establish and achieve the task purpose, whether the user can obtain an effective action plan, whether the user can adopt appropriate operation steps, and whether the user canfeedback informationEvaluate whether the task is completed, and finally make comments, such as what effect to achieve, whether an action is effective, whether an action is appropriate, and whether a situation is good.The advantage of this method is that anyLow FidelityPrototypes, including paper prototypes.The disadvantage of this method is that the evaluator is not a real user and cannot represent the user well.
Heuristic evaluation
（Heuristic Evaluation）Nielsen and Molich (1990) put forward that multiple evaluators (usually 4 to 6 people) should repeatedly browse the various parts of the system according to the usability principleInterface, independent evaluation system, which allows all evaluators to discuss their findings after completing the evaluation independently and jointly find out the usability problems.The advantage of this method is that experts make decisions quickly, use less resources, and can provideComprehensive evaluation, EvaluationmaneuverabilityOkay, but there are alsodeficiencies: First, it will be subject to the subjective influence of experts; second, it will notMandate, which will lead to the inconsistency of expert evaluation. Third, the credibility of the evaluation is reduced due to the reason of the evaluator in the later stage of the evaluation. Fourth, there is a gap between the expert evaluation and the user's expectations, and the problems found can only represent the meaning of the experts.
User test method
(User Test) is to let users really usesoftware systemThe experimental process is observed, recorded and measured by the experimenter.This method can accurately feedback the user's performance and reflect the user's needs. It is a very effective method.User testing can be divided into laboratory testing andField test。The laboratory test is conducted inAvailability testing laboratoryThe field test is conducted by usability testers on the user's actual use site.
For some users, "testing" has a negative connotation.We should try to make sure that they do not think the test is aimed at them.We want them to understand that they are helping us test prototypes or websites.In fact, we can not use the term "test".On the contrary, we invite participants to help us, "dare to try prototype".When the user is difficult to complete the task, we should change the website, not the user.At the same time, we should also consider the extent to which the website can meet the goals of those typical users, rather than focusing on how well users do this task.
Rely more on users' performance than their preferences
Through testing, we can measure the performance of users and their preferences.The user's performance includes whether it was successfully completed, the time taken, the errors generated, and so on.Preferences include user self-reportedSatisfactionandComfort。 Some designers believe that if their design can cater to users' preferences, users will have a good performance on the website.But the evidence does not support this.In fact, users' performance and their preference for products are notOne-to-one correspondence。A study found that about 70% of users agree that performance is related to preferences.In other words, they perform well on favorite websites and poorly on disliked websites.However, a relatively large proportion of people (30%) believe that the performance of users and their preference for products are not one-to-one.They may perform well on websites they don't like and poorly on websites they like.There are many explanations for why people give high comments on websites with poor performance.They may blame themselves, not the website, for their poor performance.In other words, they may worry that giving a lower evaluation will hurt the website designers, that is, our feelings.In other words, they may not have completed the task, but they think they have successfully completed it, and they do not realize the problem.For all these reasons, we recommend that you rely more on users' performance than their preferences.
Apply the test results you have mastered
Usability testing is not just a milestone used to check the project progress. You should know that when the last participant completes the task, the usability testing is not finished.The whole team must carefully study the results, set priorities, and modify the website prototype based on the results.
To manufacture any product, including most websites and software, you need to consider the working methods, experiences, problems and needs of many different users.Most projects, including designing or modifying websitesprocessing time , budget and resources.Balancing all aspects is a major challenge for most projects.When you weigh the pros and cons, it's best to give priority to developing websites or software that can enable the most users to complete tasks.Research shows that after the product is launched, the cost of supporting failed customers is far higher than the cost of product modification during development.You need to carefully consider the hypothetical users, usage scenarios, and usability test results, and try to find outcustomer demandThe ideal solution for.If the best solution cannot be found, users will not be able to complete the task smoothly.Evidence shows that even if users extend the use time to complete tasks in the product interface, it is far less than in a better product interfaceSense of success。
practicability
Announce
edit
Whether you use formal or informal equipment, you can do usability testing.With any type of equipment, you can use a variety of formal or informal methods.
Location of usability test
Using any of the following settings, you can conduct an effective usability test:
*Fixed laboratory with two or three rooms, equipped with audio-visual equipment
*Meeting room, user's home or studio, equipped with portable recording equipment
*Meeting rooms, users' homes or studios, without recording equipment, can also be replaced by human eye observation and notes
*When users can control remotely in different locations
Therefore, even if you don't have or can't find a fixed laboratory, you should alsousabilityTest.Don't say, "Because we don't have a usability lab, we can't do usability testing." Just do it!You can do it in any space.
How many people need to participate in the usability test
It just depends.A typical test requires 8 to 16 people (perUser group）。If each user will spend one hour, it means that the test of each user group will take one to two working days.When your project is in: * paper prototype or early stageDevelopment phase*It is planned to pass several rounds of testing and the whole development * is quite consistentUser groupIf you only need people to help you find serious problems, you may only need four or six people*If you have differentPotential usersGroups (e.g. doctors, patients, researchers), you need user representatives from all these groups.If youComputer operationOr network experience is required, and users with less experience and more experience should also be included*If you want to formallyQuantitative test, you will need more people to get statistically meaningful results.For diagnostic usability testing, 6 to 8 users are usually not enough to reveal most problems of the product*If onWebsite developmentDuring the process, you have been doing iterative (repeated) usability testing, and many users will participate in one or another version of theWebsite test。Therefore, although there are less than 10 test participants in each usability test, you may need 15 to 30 people to participate in the test before the website is launched.
How much does usability testing cost?The cost depends on the size of the website, the amount of your tests, the number of expected user types, and how formal you expect this test to be.If you have a standardTest procedureWith available materials and equipment, usability testing will be carried out quickly and cheaply.If you or your user recruitment company has a user database, you can save a lot of time for recruitment, so the cost will be less.
Budget for usability testing
These factors should be considered:
*Time spent in planning: determine the main problems of the test, the types of users to be tested, the screening questionnaires of the recruited users, and the test scenarios.
*Recruitment cost: the time spent by the company's staff, the cost to the recruitment company (usually a good choice), the usability experts need to spend time getting familiar with the website and its production team, designing corresponding test scenarios, and if you need to recordTest processIt also costs the rent of laboratory or portable video equipment.
*Time spent by the team observing users (testing)
*Rewards or gifts paid to test participants
*AnalysisAudiovisual materials, the time taken to find the existing problems and recommend solutions
*Time spent discussing changes and modification plans with developers, and writing survey results and recommendation reports.
Remember that budget analysis involves multiple availability tests.Building the usability of a website (or product) is an iterative process.You will find that the budget for a few small tests during the development process is more valuable than that for only one large test at the end of the project.Website usability testing is to achieve cross form visual consistency, including testingScreen resolutionDisplay when changing,MarginsAnd column layout, form color and size, font used for labels, button size, hot spring orShortcut key, labels of animation/graphics, buttons and other controls used, the same asOne fieldOfText BoxLength, date and time field format.
