IT-systemer: Kulturel kontekst, kommunikation, brugbarhed og brugbarhedsevaluering

Karakter: 12

ITSY exam january 2019

Student number: 20184159

Name: Thomas Kristian Margon Vinther

Field of study: Informatik

Usability and usability evaluation

1: What is usability? Provide a definition and describe in detail the components that comprise usability.

Usability is mainly about how easy a system is to use. If you make an incredible system that would be able to help millions of people, it would be useless if people didn’t know how to use it. This is why usability is important. The definition of usability is: “the 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.”[1] This means that it should be easy for the user to use and navigate in a system, it should be easy to achieve required goals, and it should be easy to use the system again after having used it once. Usability is also a part of UX (user experience), which is a broader term that describes the combined experience that the user has with a specific system or almost anything else for that matter.

Usability is comprised of five quality components. These are learnability, efficiency, memorability, error, and satisfaction.[2]

Learnability is about how easy the system is to use for a first-time user. When a user tries a new system it should be as easy as possible for him/her to accomplish basic tasks. The learnability of course depends on how advanced the system is. Photoshop is for example harder to use for a first-time user than Microsoft Word. So the context of a system is important to keep in mind.

Efficiency is about how fast the user can perform tasks in the system once the user has learned to use the design. This means that a system in which intermediate users can perform tasks with little to none waste of time and effort, is very efficient.

Memorability is about how easy it is for users to memorize how to use the system. So if a user returns to use a system after having already used it prior, how easy it is for him/her to start using it again. If a user has a too hard time accomplishing basic tasks in a system after having used it prior, the memorability is bad. Memorability is a bit harder to measure than the other components because it requires the users to be absent from the system for a specified amount of time.

The component errors focuses on the amount and degree of errors that a user gets using a system. If a user gets a lot of errors using a system there is probably space for improvements. The severity of the errors is also important to measure because errors of high severity represent a bigger problem. If it is hard to recover from an error that also means that it becomes a bigger problem.

Satisfaction is about how pleasant the system is to use. If the design is good, the satisfaction is affected positively. Satisfaction is very important if you want your users to return to your system/website after having used it.

2: What are the differences between heuristic inspection and usability testing?

A heuristic inspection is done by experts and a usability test is done by test users. In a heuristic inspection usability experts looks at the system and compare it to accepted usability principles. These principles are for example visibility of system status, user control and freedom, consistency and standards, and error prevention.[3] In a usability test the test user is given some tasks to complete in a system. If you want to perform a usability test you would need to find some test users. The amount of test users is dependent on what you want to test but Nielsen Norman Group states that 5 test users is almost always close to user testing´s maximum benefit-cost ratio.[4] Both the heuristic inspection and the usability testing produce a list of usability problems.

The primary advantage of heuristic inspections is that they are easy and quick to conduct. There is no need to find test users, all you need is a heuristic evaluation expert. This means that heuristic inspections more often than not are less expensive than usability tests.[5] The fact that they are quick and easy to conduct, also means that it is easier for designers to obtain feedback early in the system-building process.

The disadvantage of heuristic inspections is that they most often aren’t completely representative of the system target group. Therefore you might risk that the problems that the experts find are less genuine than the problems found from usability testing. The experts might also find more minor issues and fewer major issues.5

Performing a usability test has the advantage that the test users are representative of the target audience. This means that the problems that they find are more “real” then the problems a heuristic expert would find. This test also gives insight into how users use the website. One advantage of this is that you can use it to track time of how quickly test users complete certain tasks, which gives you an insight into the learnability and efficiency of your system. Another advantage is that this makes a “thinking aloud” test possible. In a thinking aloud test the users should say what they are thinking while using a system. This gives some great insight into which areas of the system could be improved.

The disadvantage of usability tests is that they can be unnatural because the test user is in a “sterile” situation, because they most likely are in a usability laboratory. The laboratory and the persons enabling the usability test might also make the test person a bit nervous, which might distort some results. A usability test is also quite difficult and takes a lot of time, which also means that it can be quite expensive.

3: What does it mean for a system to have high, or low usability?

Having a high usability means that the system does well within the 5 components that comprise usability. As mentioned previously these are learnability, efficiency, memorability, error, and satisfaction.

So if a system is easy to learn, easy to use efficiently, easy to memorize, easy to avoid errors or at least easy to recover from errors, and satisfying to use, that means that the system has a high usability – and vice versa.

4: Describe the process of defining tasks for a think aloud session. What are the necessary steps we need to follow? How can we make sure that the tasks are suitable?

When you want to define tasks for a think aloud session it is important to plan the context of use. This includes where the system is used, who it is used by, why it is used and how the design is used. This helps to ensure that the chosen tasks are suitable, and that the test results are valid. To ensure reliability you should also be sure that the tasks are based on realistic user situations.

When you want to define the tasks there are a couple of necessary steps you have to follow. First you have to know which basic tasks that the user uses the system for. For example: If you want to test a website that sells plane tickets, you need to define the basic tasks that the user has to do in order to buy the wanted tickets. This might look something like: open the index page -> search for flight to New York from Billund -> buy ticket.

You also have to decide on which parts/functions of the system that you want to test. If we follow the previous example: Maybe you don’t want to test how easy it is for users to buy tickets. Maybe you want to test how easy it is for users to contact the support to get help. That means that the chosen tasks should be different dependent on the context of use.

It is also important that the task descriptions are very easy to understand for the user. If you make some tasks that the user can’t understand then they are basically worthless. But it is also important that the tasks don’t give away part of the solution. Following the previous example, a wrong task may be: “click on help in the main navigation bar and then submit a support ticket.” A better solution would be to write: “find the support page and submit a support ticket.” This way the user is better tested for their ability to navigate the design themselves.

It is also important to keep in mind how long each task would take to solve. A complete usability test with introduction, test and debriefing is normally about 60 – 90 minutes.[6] If the test is longer you risk that the test user gets tired. So you should generally aim for tasks that take about 45 minutes, because then there is space for the introduction and debriefing too.

Cultuaral context and communication

1: Define and explain communication according to both James Carey (Carey, 1992) and Ole
Thyssen (Thyssen, 1994). How are these approaches to communication different and/or similar?

James Carey thinks that there are 2 views of communication. The transmission view and the ritual view. Both of these views he says is from religious origins. The transmission view states that communication is imparting, sending, transmitting or giving information to others.[7] This might also be the most common understanding. The transmission view is a lot like transportation and the idea is therefore that communication is transmission of messages for the purpose of control.[8] The ritual view states that communication is more focused on maintaining a society and not just sending messages from one person to another. The purpose of communication is therefore cohesion that creates a fellowship and bonds people together.

Ole Thyssen’s viewpoint is more based on the theories from Niklas Luhmann[9]. He views the society as a social system that derives its meaning from a system of communication. One of the things that Ole Thyssen views different than James Carey is that he thinks that communication is not a transmission, but rather a selection/selections. He thinks that communication is comprised of four different types of selections: information, message, understanding and association between sender and receiver.[10] Each selection also have a risk of failing the communication. The association selection is quite similar to the view of James Carey, because this is about the receivers willingness to accept the communication offer – which corresponds well with James Careys thoughts that communications is about cohesion.

Ole Thyssen also states that because communication is comprised of selections, a communication is still present even if none of the persons are saying or doing anything. This could for example be in an elevator where people don’t talk to each other, but even the decision to not engage in a communication flow with each other is communication in itself. The signal to not wanting to communicate is therefore also communication. This is both similar and different to the views of James Carey. Different because he sees communication as a transmission and similar because he sees communication as method to create fellowship – or in this case: choose not to create fellowship.


2: Define and give examples of interactivity by applying the concept of interactivity (Andersen,
2001: Signs of Learning in Computer Cultures) to the IT-system you work on your P1 semester
project. How is interactivity represented in this IT-system?

The concept of interactivity can be presented by two dimensions. The first is about the infrastructure and the second is about the creation. These dimensions can be used to give insight into understanding how computers, learning, and culture impacts each other. The infrastructure part is about who is producing the information and who is controlling the information. This can be seen on the figure below.

The production and distribution of information can be administered by a user or a center. Therefore, there is a total of 4 different structures. These are transmission, registration, consultation, and conversation.

In my P1 project we wrote about AKU-Aalborg, which is a public student housing service in Aalborg. If we take a look at the infrastructure part in relation to AKU-Aalborg, we can for example see that the user have to make an account in order to get on the waiting list for an apartment. This account-creating-process is a registration because the information is produced by the user but the distribution is controlled by the center. Transmission and conversation is almost not on AKU-Aalborg´s website at all. But consultation is a huge part of the website, because there is a lot of information in regards to the task of finding a student housing. This information is produced by the center and controlled by the user as he/she navigates the website and therefore it is a consultation.

The creation dimension also consists of four aspects. These are perspective, play, interactivity-as-emergence, and pragmatism. In terms of interactivity Luis O. Arata uses these aspects to say that ”… an interactive approach favors the use of multiple points of view which can coexist even if they appear mutually exclusive; it celebrates the creative value of play; it is a catalyst for emergence; and it tends to be ultimately pragmatic.” (Arata 1999, p. 2). If we look at AKU-Aalborg from a creation perspective, we can for example see that it to some extend takes use of the play aspect as users have to play around on the site to find apartments of their liking. The search functions in the website is very limited so users have to look around and see what they can get with the limited search functions that is given.

3: By applying the concept of the “model user” (da. Thorlacius’ “den implicitte modtager”) try to
establish a profile of the user of the IT-system you work on your P1 semester project (exercising logical deduction as demonstrated by Sherlock and Mycroft Holmes). How may we deduce the “model user” by pointing out and creating relations between signs?

AKU-Aalborg is a student housing company paid for by the public with the intention to rent out apartments cheaply to students.[11] Therefore the primary users must be students or soon-to-be students. The overall service is also cheaper than most other services because the service itself is free and the rent for the apartments is cheaper. The students that visits this site is therefore low likely to be rich people, because rich people might not want to wait in queue to get a good apartment when they could just pay more for one and get it now. The model user must also be someone that is decent at planning for value in the long term because they have to wait in queues for quite a while in order to get an apartment. He/she must also have a couple years of studying left because you can’t have an apartment through AKU-Aalborg if you aren’t studying and most people won’t go through the hassle of moving for just living somewhere for a short period of time for then just to move again shortly after. The people that are studying are mostly in their youths but not young enough to go in high school. So it would be fair to assume that the model user of AKU-Aalborg’s service is in his/her twenties. Since the service is only available in Denmark it is also fair to assume that most users would be Danish.

Many of AKU-Aalborg’s available apartments are colleges.[12] From that you might assume that the people living there would be more outgoing than the average, but I don’t think this is the case, because the fact that they are cheaper simply attracts all kinds of people.


4: Discuss how it is possible to establish links between a paradigm of communication and a
paradigm of usability based on your work with the course materials, lectures and lecture
exercises and the abovementioned questions. How can the study of communication inform an
evaluation of IT-systems?

The study of communication can inform an evaluation of IT-systems, because they can tell something about how the system is perceived. If you make a usability test you can find out which problems a given system has. But if you study the system from a communication point of view, you might find out why the problems exist and how they can be solved. The communication part helps in that you can analyze the things that the user does and why he does it. For example if a user can’t navigate through a website, you could analyze it from a communication point of view to find out why he can’t navigate the system. It might be because of semiotics in that he perceives the signs and styles of the website in a different way than intended. In a usability test you can also make a think-aloud test so that the test user says everything that he thinks aloud. If you make such a test you get better insight in the user’s perception of the website, and from there it might be easier to analyze and better the communicative experience.

A mistake you could find in a usability test might be that the system simple changes the layout too much across the website. This would worsen the “fatisk function” (Thorlacius)[13] in that the communicative connection between user and the system would be worsened. And if the “referentielle function” (Thorlacius) isn’t there to tell the user that he is in fact still on the same website, it might bother him and worsen the overall experience. An annoyed user is also bad for the website because the frustration might make it harder for him/her to complete his/her goal with the system. This goal could be to purchase something and therefore the system-owners would lose profit. If you didn’t analyze this from a communicative point of view it might be harder to find out exactly what would fix this problem.

If there is too much text on a website it would make it harder to get an overview of the site. This also might lead to users not reading all the text which then might lead to usability problems because the users can’t figure out how to complete his or her goal. If you take a look at this from Ole Thyssen’s point of view, it might have something to do with the selection of communication. If there is too much text the user might make the selection to not receive more of the communication that the text was supposed to give. From a James Carey point of view the error might exist because there is too much ritual communication in the text. Therefore you should maybe consider “sacrificing” some of the text that is meant to bond the user and system/company together, thus keeping him/her reading the parts of the text that is required for him/her to reach the desired goal.

  1. https://www.interaction-design.org/literature/topics/usability ↑

  2. https://www.nngroup.com/articles/usability-101-introduction-to-usability/ ↑

  3. https://www.usability.gov/how-to-and-tools/methods/heuristic-evaluation.html ↑

  4. https://www.nngroup.com/articles/how-many-test-users/ ↑

  5. https://www.usability.gov/how-to-and-tools/methods/heuristic-evaluation.html ↑

  6. https://www.nngroup.com/articles/time-budgets-for-usability-sessions/ ↑

  7. Carey (1992) A cultural approach to communication ↑

  8. http://www.postcolonialweb.org/poldiscourse/2views.html ↑

  9. Thyssen(1994).Teorien om kommunikation hos Niklas Luhmann File ↑

  10. Thyssen(1994).Teorien om kommunikation hos Niklas Luhmann File, Page 78 ↑

  11. http://www.aku-aalborg.dk/media/2077/vedtaegt-for-aku-aalborg-godkendt-13-nov-2017.pdf ↑

  12. https://vl.aku-aalborg.dk/BoligOversigt.aspx?ID=0 ↑

  13. http://lundberg.at/teori/Lisbeth%20Thorlacius%20webmodel-kopi.pdf ↑