Methods for gathering information

While we may be able to see that there is a problem, knowing more about the situation can help us identify a potential solution or get validation that the solution we have in mind might actually work. Creating a research plan is a useful way to frame such research to make sure you are focused towards a single end goal.

Regardless of the scale of the problem, I always find taking at least a moment to write down the answers to the research plan prompts as a great way to get the core problem straight in my head. I often find that it helps me identify assumptions that would have hampered things moving forward and focusses my attention on what really matters. As with other elements of the process it’s highly scalable – you might just have an objective and a couple of questions – or it might fan out if the scale of the initiative is broader.

Marty van de Weyer

A ‘research plan’ sounds very formalised, but it is ultimately just posing yourself a few questions to draw your focus and to force you to outline what you are trying to do. Firstly you can create a research objective where you state the core thing that you are trying to find out or do in a simple sentence or two: 

The next stage is to consider what questions you have, or that you would need to answer, to respond to your research objective:

In many cases this may be enough, and help you to see that you might need to take a different approach than the one that you had been considering.

A simple research plan like this is also a useful tool to share with stakeholders to get their input. It lays out what you will be doing as part of this work and gives them a chance to reflect and raise anything that they had been thinking about, but not yet articulated to you.

Once you have figured out your research questions, you can then consider what might be appropriate methods to seek answers. While there are a range of human centred design methods for collecting information about a problem, you will likely have to select methods in relation to the scope you have available to investigate. For this reason, the list of methods below is broken down in that way. Ultimately, trying to use these methods in any way is better than not using them at all because of the aura around ‘doing it right’. There is of course value to knowing more about the methods and making the data more useful, but the important thing is to try them so that you can learn through doing.

Firstly it’s important to briefly consider the questions below to help you target the types of methods that might be the most effective.

• Who would be able to answer it/give you meaningful information (users themselves? A peer? An expert?).
• Is there any data that you could collect or access which would give an answer (or just a viewpoint) in relation to a particular question?
• Could you use some form of observation?

If the problem is small, or you need a quick solution, or you are just dabbling with the human centred design space, you may want to try some simple checks:

• Just ask students/teaching staff – can you send a simple email to a few individual students? Or your peers? Will that give you the information you need?
• Desktop research – if you are looking for an expert’s perspective, sometimes looking for other examples of research in the space will give you what you need to respond to a particular problem.
• Quick survey – ask some questions and get people to hold their hands up in class or in a meeting to respond. You can also do similar things in shared documents and digital whiteboards in a meeting, and then the data is already recorded for you.
• Basic observation – what do students do in class, or what do they do as part of a remote interaction? What is their behaviour based around the communication you give them or the actions you ask them to perform?

If the problem is a bit more complicated or you are looking for greater confidence in a potential solution you might like to consider some more in depth methods.

• In depth survey – Taking time to develop key questions (potentially piloting those questions with a small group of users to check they work) and then circulating a more formal digital survey more widely can provide a broader and more confidence inspiring dataset. 
• Interviews – Formalising the ‘just ask’ process into interviews by designing your questions, setting a flow and recording so that you can re-listen, pull out key ideas and place them aside for more structured interpretation.
• Formalised observation/Extracting behaviours from observational data – Take the time to figure out particular things that you are trying to track and put tools in place to do so. Structure and check the data against precedents to see what changes have occurred.

When using any of the methods mentioned above that involve asking people direct questions, please consider the next resource on creating user conscious questions. 

There are also methods that you can use to unpack and visualise parts of the problem space. These are discussed in greater detail in the final resource in this collection.

• Mapping – illustrating the problem to find trends and gaps.
• Personas – building characters to represent your key user types.

Repeating from above, the method chosen should be driven by the problem and the capacity that method has to provide meaningful and useful information. Thinking about how particular methods will get you what you need to have confidence and make decisions is as meaningful as using the methods themselves.

If there is scope you can use more than one method to feed in to another for greater efficiency and reliability. For example, a quick survey could be an effective way to bring to light the key areas that people are concerned about, which you could then target more specifically in direct questioning/interviews.