Qualitative research

People's behavior is something very complex, that involves too many variables, the amount of data alone is not enough to understand the actions.

Unlike quantitative research, which seeks to know in quantity (how many / how much) and obtains this information indirectly through other tools (analytic tools), qualitative research seeks to understand why, how and what, in a more detailed way that can help to better understand the complexity of the problem to get a solution.

The vast majority is obtained in a direct way through observation and a set of questions that meet what the person is doing in real time.

Qualitative research allows to understand:
- Behaviors, attitudes and aptitudes of potential users of the product
- Technical, business and environmental contexts involving the product
- Vocabulary and other aspects of the domain in question
- How existing products are used

It is possible to collect qualitative data through interviews with the involved parties (stakeholders). In these interviews with stakeholders, it is possible to better understand the technical context in which the product is located and to make known to those who are developing it how the user research will be conducted.

It is important to distinguish the concept of consumer from user. Consumer is a user who has decided to purchase or purchase the product. When the product satisfies the user's goals, the probability of being purchased is greater.

The main focus of researching should be users of the product in question. A large part of the source of qualitative data comes from interviews and observations made from the user. Most people are unable to properly justify their actions or behaviors, so it is not enough just to prepare a questionnaire, hence the need to collect data in other formats, such as video, so that they can be analyzed later. Some of these qualitative data are mostly directly linked to the product and are collected by the marketing team, from plans on the brand (brand strategy) to market research.

However, one of the most important methods for collecting qualitative information are ethnographic interviews.

“Ultimately, we believe that properly conducted open-ended interviews are quite capable of exploring these aspects of the user’s mental model. By asking the right questions and paying close attention to how a subject explains his activities and the domain, you can decipher how he mentally associates different bits of functionality and information.”
Cooper & Reimann & Cronin

Contextual inquiry

Contextual inquiry is an interview method created by Beyer and Holtzblatt that aims to obtain information related to the context in which the interaction takes place.

Based on the master apprentice model, where the elaborated questions are asked so that the person is the master and the interviewer is the apprentice.

It consists of formulating a set of general questions, then observing and asking any necessary questions while the user works in their regular environment (context).

As this type of interview is carried out in the regular environment in which the user performs his functions, the type of qualitative data obtained is more realistic than when analyzed in the laboratory.

Usually this type of interviews are done in the initial phase of the design process, as they easily obtain precious information about the user's workflow, social, technical and physical environment.

Its results help to improve the interaction process, to better understand what is really important to users. It also helps to discover information that can enable new projects (not previously thought out), as discovering various problems that the user encounters during the day to-day, who still don't have a solution or aren't efficient enough. However, this method has some limitations and may not be enough.

Ethnographic interviews

Ethnographic interviews are a combination of immersive observations and direct interviews.

Initially, candidates have to be identified. This will be the starting point in identifying personas (a concept that will be clarified later). However, the personas chosen for this type of interviews are mostly approximate hypotheses of those that will be created in the future. This is because it is a step that takes place before the personas are definitively established.

These personas hypotheses must be worked out by answering these three questions:
- What kind of different people will use this product?
- How might their needs and behaviors vary?
- What types of behaviors and environments need to be explored?

After the temporary personas have been created, it is necessary to plan the interviews that will be done. According to the authors, these interviews should last approximately one hour (depending on the complexity of the subject) and should be carried out by a team of two designers (one conducts the interview and points out small notes, while the other points out more detailed notes).

There are three different types of phases in which interviews are conducted:

Early interviews, which focus more on the user's point of view. The questions are less technical and detailed, with more openness.

Middle interviews, occur when patterns of how the person uses the product start to be visible. Some questions are starting to be more focused on these behavior patterns recognized by designers.

Later interviews, use more specific questions that help to better understand the person's role and behavior, thus obtaining answers that justify the patterns of use (behavior) previously obtained.

These are some of the methods mentioned by the authors, so that an ethnographic interview can be properly carried out:
- Interview where interactions take place
- Avoid a set of fixed questions
- Focus on goals first, then on tasks
- Avoid making the user the designer
- Avoid discussions about technology
- Promote the use of a narrative (storytelling)
- Asking to demonstrate and talk about an action
- Avoid questions that suggest a solution or opinion

Focus groups

Group interviews (focus groups) are normally prepared by the marketing department, where after analyzing the target audience, they gather a number of people (between 3 to 12) in a room and make a set of questions and exercises that lead to a discussion of the group in general. They are usually concerned with the visual aspect of the product or its appearance (largely due to the fact that they are mainly made by marketing people), however for design methodologies their results can be misleading and not very credible, as usually the group's opinion, it easily becomes a personal opinion due to the force that the group ends up exerting psychologically on each person.
They do not determine usability quality.

After all, they are particularly good at determining what an audience wants and their results are most accurate in the early stages of development.

Usability testing and User testing

Assuming that all users are equal and act in the same way is wrong.

Each user perceives things differently, often depending on their environment and the type of actions they are used to performing. Discarding a tool because it assumes that all users do not like it is also a wrong procedure, unless it is supported by some research or investigation.

However, the goal is not to discard tools or options for assuming something, but rather to adapt as needed to improve the user's usability. How do you determine that something contributes to good usability? You carry out a usability test.

Currently, the concept of usability testing is confused with user testing.

The difference is that user testing is concerned with understanding what kind of solutions users need, while usability testing is in charge of understanding how the user reacts with the product or interface.

A usability test is actually a cluster of techniques related to the collection of user characteristics when interacting with the product, in order to improve its usability.

It is usually measured by how successful the user is in solving specific tasks and problems he encounters. Their results reveal where users encounter the most problems, as well as where they are most successful.

User testing is a good way to gain quantitative value on problems that were found in usability testing so that solutions can be provided to a majority.

Usability tests are carried out after the conceptual phase of the product, where there is already enough material and information to build a prototype that can be tested.

It turns out that in the early days of product development, the prototype will have to be low-fidelity, or even something different that only simulates its behavior, but allows you to have a sense of how the user reacts to that feature, even if it is not yet implemented.

The important thing here is to test people's reactions as soon as possible, as the information obtained will influence the product's development.

“Testing reminds you that not everyone thinks the way you do, knows what you know, uses the Web the way you do.”
Steve Krug

It is necessary to go testing according to the different production parts of a project.
Usability testing is not dependent on the number of participants and is carried out one-by-one.

It is enough to design the test for at least one person to draw conclusions on how to improve the product. A test in the early phase has more value than the final phase, even if the final phase has more participants. The participants should not be limited only by the target audience (although they are the most important), the important thing is to test independently of the audience, as problems will always arise.

It is not a static process, on the contrary, it is a dynamic process, where any change has to be tested, even if it is supposed to be an improvement.

There may not be a perfect solution immediately after the fix, but as it is a cyclical process, any and all changes are improved until they are established as efficient or usable. The user's live reaction is very important to confirm these changes.

A usability test is not a test on illustrative elements, colors, shadows and so on, which are often interpreted by the user as personal taste.
The interest is not whether it is visually appealing, but functional, easy to understand and communicate.

It is important to focus on specific aspects that the user found problematic or difficult to use. Comprehensive or general answers do not help detect the error or problem that is creating difficulties for them.

In the end you need to write a report, as well as in all kinds of different tests and interviews.

The more detailed and accurate the report is, the easier it is to improve the system.

It should be noted that for each session performed, a short, concise and easy-to-interpret report is required, as this will be used by several different departments that are involved in product development. Its conclusions cannot be subjective and of different interpretations, they have to be conclusive.

Eyetracking

It is a technique that allows you to study and follow the behavior of the user's eye.

With this it is possible to determine several things, such as which elements are receiving more attention, the fluidity of the content and its visual hierarchy, etc…
The two most common ways to present this information are: heat maps and gaze plots.

Both exhibit the behavior of the eye, just in a different way.

Heat maps are better for understanding which zones the user focuses on the most.

Gaze plots allow you to analyze the order in which the user looked at the various elements.