Understanding User Needs

There are a number of methods that can be utilised to understand the needs of users; however, depending on the method we use we may only understand the surface layer of what users think they want, rather than pushing into the deeper understanding of what users really need, as illustrated below. This section will delve into a number of methods, listing them from surface gathering methods to gaining deeper understanding, that can be used in healthcare. Each method highlights what the method is, and why, when, and how you could use each during the development process.

Surveys

Gain people’s thoughts, feelings, and opinions based on their responses to a list of questions.

What and Why

Surveys are a means to gather data from a chosen population (which might be more or less precisely defined) of people based on a list of questions. There are two types of surveys questions, which produce different types of answers. The first are closed questions, which offer a predefined list of responses, typically using a 5-point Likert scale. These questions produce quantitative results, that is, results that are measurable and require statistical analysis. The second are open questions where the participant can provide whatever information they think is relevant to the question. These questions produce qualitative data. Most surveys will incorporate a mixture of both open and closed questions, depending on what response the researchers think will provide the information necessary to help them improve the system.

When

Surveys can be used at various times throughout the course of product development. They can be used at the beginning of a project to gain an understanding of users’ perceptions on a topic or after running an evaluation to assess the users’ experience when using the system.

How

When developing a survey it is important that the objective of the survey is well understood to help focus the questions. Decide which questions are best given as closed questions and those that should be open questions. With closed questions ensure that the question is not biased towards a particular answer, and that responses are consistent between similar questions. Surveys should also be as short as possible; this reduces the risk of participants not finishing the survey, or getting frustrated with it and answering questions without reading them. Once the survey is created it should be rigorously tested, to ensure that participants provide the correct type of answers and that the questions are easy to understand. In some situations, there are existing validated questionnaires such as the System Usability Scale (SUS), or the NASA Task Load Index (TLX); these can be very useful if they address the questions that matter in the study. The method by which the survey results are analysed should also be well understood and tested.

Pros

Can be used in a variety of ways and are not onerous on the participants.

Cons

Surveys are deceptively easy to create, but it is difficult to create a good survey that is able to gather insightful data. They only gather information on the participant’s opinions and not on what they actually do. If not properly designed the survey will not generate any informative results. Difficult to get participants to complete the survey completely, especially when there are lots of questions (particularly open questions).

Points to Ponder

Are the questions appropriate?
Is the survey too long?
Is the survey too repetitious?For closed questions, are the questions (or possible responses) biased?

Resources

For further information on creating great qualitative surveys, click here (25 min read including videos). For useful questions to improve feedback surveys, click here (4 min read).

Focus Groups

Bring together a group of people to understand general perceptions, beliefs, and attitudes.

What and Why

Focus groups are used to establish a consensus from a large group of users through discussing together an issue or task to understand how they would like to see it improved, and identify the barriers they currently face or broad opinions on a subject or product. In healthcare, focus groups could include various groups, such as clinicians, patients, nurses, etc. when exploring issues that relate to multiple users. However, in some cases a single user group may be all that is required (e.g. understanding nursing issues at the bedside may only necessitate talking with nurses). Therefore, the participants should be recruited based on their interaction with the problem space.

When

Focus groups can be run either during the ‘understanding user needs’ section or during the ‘design stage’, depending on what information you are trying to elicit from the participants. When observations may be difficult to conduct in a timely manner and you are seeking to identify general concerns and barriers for users, then conducting a focus group during ‘understanding the user needs’ can provide information to move forward on the project. If the purpose of the focus group is to gauge the sentiments from users about initial prototypes, then the focus group would be undertaken during the ‘design stage’ once a prototype for evaluation has been developed.

How

These sessions should be pre-planned and moderated by the researcher to guide discussion whose job it is to manage and guide the discussion without manipulating the course of a discussion towards a particular opinion. To prepare for running a focus group the following points should be covered:

•    Define the objective you wish to achieve
•    Select which users are required to attend
•    Determine the number of participants per session, how many sessions, and the location of the focus groups
•    Recruit participants based on defined criteria
•    Develop questions used to guide the discussion using open-ended questions
•    Run the session

The session itself is run in a similar manner to Interviews, with the main difference being that you are interviewing a group of individuals simultaneously, and encouraging conversation between them as well as with you.

Pros

Gather a lot of information quickly, provide good insight into an issue as it is being explored from multiple perspectives at the same time.

Cons

Difficult to organise due to busy schedules. Sometimes difficult to get participants involved in the conversation. Like interviews, only give a perception on what people think happens rather than what actually happens. The opinions of some may be drowned out by more vocal participants, or participants may not feel comfortable sharing opposing views to others in a group discussion. Requires a good moderator to achieve ensure everyone’s voice is heard.

Points to Ponder

What is the purpose of the focus group?
Who should be included in the focus group?
How many participants will be in each focus group?
Do the participants best encapsulate the demographic of potential users?
Have you gathered enough information or should you run another focus group?

Resources

For further information on how to run a focus group, click here (12 min read).

Interviews

Provide insight into user needs and help establish context of the user’s environment. They also help to identify or clarify key points of frustration.

What and Why

Interviews are perceived as an easy method to perform, but in actuality can be difficult to conduct well; however, they can produce good insights into the needs of users.  They are used to explore issues and hear individuals’ stories. Interviews are versatile, as they can be conducted in person, over the telephone, or by video link. Interviews can also range from unstructured, which involves going in with no clear focus and letting the questions come naturally, to semi-structured, where the interview is guided by a script but allows the interviewer to explore issues in more detail by adding follow-on questions, to structured, where the interviewer cannot deviate from the questions they have created. In relation to human factors, semi-structured interviews are the most common as they provide guidance to the interviewer as to what they are going to ask, but provide the flexibility to delve into topics that warrant further exploration.

When

Interviews can be conducted at various times during the ‘understanding user needs’ phase. If the researcher is unfamiliar with the environment they are examining, then an interview prior to conducting observations could provide them with the necessary context. Interviews post-observations help to gain further insight from the users with identified issues when contextual inquiries are not possible. When used in healthcare, interviews should have a clear objective and the duration should be taken into account (keep it short).
To enrich data collected from interviews, probes can be used to help guide discussion or to uncover information specific to what issues you are exploring. Probes can take on various shapes and sizes, from graphs of data, screenshots or video of observations, or cards that promote new thoughts.

How

Introduction:	Introduce yourself, explain the intention of the interview, reassure them about ethical issues and data management, gain signed consent, ask to record.
Warm-up:	Ask general non-threatening questions.
Main body:	Present questions in a logical order and start with broad questions and then narrow down. Avoid closed questions.
A cool-off period:	Include a few easy questions to wind down. Give the interviewee the chance to add anything else they feel is important, or to ask any questions in return.
Closure:	Thank the interviewee, make it obvious that the interview has ended, and switch off the recording device.

Pros

Delivers insights into users’ perceptions of a task. Ability to gain insights from a range of different perspectives, such as healthcare professionals, patients, technicians, etc.

Cons

Only provide a user’s perspective on what they think they accomplish, and in some cases are idealised perspectives on the situation. Interviews can also be very time consuming and, within healthcare, may be difficult to arrange as healthcare professionals are already time poor.

Points to Ponder

What is the purpose of the interview?
What relationship does the interviewee have to what you are trying to understand?
Do you have a good understanding of the users’ needs?
Have you properly understood situations the participant has described?
Have you uncovered the information you are looking for?

Resources

For further information on the pros and cons of conducting user interviews, click here (48 min read including videos). Additional tips for conducting qualitative interviews, click here (10 min read).

Observations

Used to observe and understand user’s behaviours and workflows to identify current issues and uncover opportunities to improve user experiences.

What and Why

When observing people interacting with technology, we are able to see how that technology works within the context of their workflow. This allows us to see what people actually do, as opposed to what they say they do. Observers in healthcare settings typically take on the role of outside observation; that is, the observer is not participating in the environment they are observing. It is also important that when observing healthcare settings multiple perspectives may be required as professionals move around patients. Through observations we want to understand what people are doing, how they are doing it, and with whom they are doing it. We also should be observing the moments that bring satisfaction to those whose work we are observing, and their moments of frustration. Identifying these key points help us to understand when technology can be helpful or a hindrance to the users.

When

Observations are an important method used to identify user needs. Depending on the researcher’s familiarity with the environment and the users, observations may be carried out prior to interviews. They provide a good first step to familiarise oneself with the context the user works in, and understanding the users’ needs in context. Observations may also highlight other areas, previously unknown to the researcher, that require further exploration. For example, if you are observing pain medication delivery at the bedside, you may find that the issue of late delivery is not at the bedside, but at the drug station where bedside nurses get access to the medication.

How

The AEIOU framework can help identify what you need to focus on when conducting observations.

Activities	What are the actions that the people are performing?
Environments	Where do the activities take place? What resources are available to them? What is the environment like (noise, temperature, lighting)? Where are people standing and other objects in the room positioned?
Interactions	How are people interacting (verbally, mannerisms, through technology)? Who are they interacting with? What are the interactions about?
Objects and Tech	What objects and technology are present? How are they being used? How are people using particular objects to perform their tasks (gripping, manipulating, typing, touching)?
Users	Who is present? What is their role? How are they involved in the activity?

Pros

Provide an unbiased glimpse into the practices of users, able to uncover issues that users are unaware of.

Cons

Time consuming, small sample size, huge data set, Hawthorne effect (people alter behaviour in response to being observed). Often hard to interpret data when the work is unfamiliar to the observer.

Points to Ponder

Who are the main users you want to focus on?
How do the key users perform the given task(s)?
What aspects of the task are frustrating to the user?
Does anyone else interact with the task / activities you are investigating?
Are you able to capture all of the data you require?
What area of the observed environment is of most interest to you?
What routines do the participants undertake?
What are they doing that you didn’t expect them to do?

Resources

For additional information on observations, specifically how to prepare and conduct, click here (10 min read).

Contextual Inquiry

Are a combination of interviews and observations, providing opportunities for the researcher to inquire of the user additional information to improve understanding.

What and Why

Contextual Inquiry is a method that combines observations and interviews. It allows the researcher to be observing the practices of the participants, and provides the freedom for inquiry into why they are performing the actions they are performing at the time and how they understand their work. This allows for a great deal of information and context to be understood quickly based on the user’s experience performing the task, and acknowledges the user as an expert at performing their tasks. Contextual Inquiry can also provide important details to models relating to communication, roles and relationships of those being observed, sequence of events, impact of the environment, and reasons behind the structure of the system.

When

When possible, contextual inquiries should be conducted as one of the first methods of understanding the user’s needs as it allows the researcher to probe the user for details they do not understand. In some scenarios they can be performed instead of observations, as the interactions of the users may be minimal and easily captured during the inquiry. However, in some healthcare settings contextual inquiry is impractical due to patients being present, the cognitive effort already required for healthcare professionals to perform their tasks, and the ability to record both sound and vision when people are working within sterile conditions and patient privacy needs to be maintained.

How

Contextual inquiry begins with observation and then from what is being observed querying users to gain further understanding. For example: Why did you ask the patient to roll over? What information is helping you to make that decision? How do you know how much medication to give the patient? Therefore, contextual inquiries should use a combination of observation and interview techniques. Optionally, the insights from contextual inquiry can be represented in a set of work models that describe the flow of communication between people, the physical environment that shapes work, the design of key artefacts that support the work, the task structure (called a “sequence model”) and an overview of the culture and policies that shape work.

Pros

Quick to perform and provide a lot of additional context that cannot be gathered from observations alone. Promotes that the user is an expert and that the researcher is trying to understand how they perform their tasks.

Cons

Not always achievable due to the constraints of healthcare. Large data set, and prone to the Hawthorne effect (observation alters behaviour). Often hard to interpret data when the work is unfamiliar to the observer.

Points to Ponder

Where is the user receiving their information?
Why do they perform the actions they do?
What is the result of the action?
How do they know when to take an action?

Resources

For additional information on contextual inquiries, click here (3 min read). For additional information on using contextual design work models, click here (20 min read).

Task Analysis

Understand the tasks users perform to reach their goal/s.

What and Why

Task Analysis is about understanding the tasks required by participants to achieve their goal(s). This is important as it allows the researcher to understand the entire workflow for individuals or a team working together to perform a task (e.g. surgery), and highlight the tasks that cause the most frustration for the user(s). Task analysis allows the researcher to understand how the participants progress towards their goal and what is required for them to achieve along the path towards their goal.

There are several types of task analysis that can be completed, depending on the situation; however, in the case of healthcare, there are two main task analysis that can be performed: physical and cognitive. Physical task analysis is the process of identifying the actions that users perform to achieve their goal within their environment and includes capturing any subtasks. A cognitive task analysis is focused on the “thinking” tasks where decision-making, problem-solving, and judgements relating to the goal are taking place.

Performing a task analysis allows for the users’ process flow to be captured and identifies the means by which they interact with the environment around them and can highlight when and how they make decisions to progress towards their goal. It provides insight into opportunities for improvement that may not otherwise be obvious. Completing a task analysis provides essential information to understand where new technology will be situated within the process flow, identifies what other aspects of the process will be affected, and how the cognitive load of the user can be reduced.

When

Task analysis should be conducted as early as possible and, in most instances, can be completed alongside an observational or contextual inquiry study. Identification of the tasks that cause the most physical and cognitive effort for users at the beginning of the design process allows for more appropriate prototypes to be generated in later development phases, which will have the most impact on improving performance of the user. Additionally, a second task analysis can be carried out when conducting ‘in the wild’ testing, to ensure that the technology is being used when expected and to see how users integrate the technology into their workflow and identify how it improves their decision making.

Pros

Identify key tasks that require attention, can be produced in line with other methods, provides the researcher with a clear understanding of what is involved to achieve a set task.

Cons

Having reliable observational, contextual inquiry, and interview data that can be structured into a valid task analysis.

Points to Ponder

What are the goal(s)?
Does the task contribute to achieving the goal?
How do the tasks relate to each other? 
Are there any tasks that are unnecessarily repeated?
Who does the task affect?
What decisions are being made during each task?

Resources

For a more in-depth review of task analysis, click here (15 min read).