Using research to identify needs
Designing products which can be used by everyone is a simple aim. This has led many designers to think it is just a matter of common sense and can be done from general knowledge without any consultation or research. Many products have been let down because designers have made assumptions about their users or have been ignorant of some of the practical difficulties bad design has caused. Standards committees have not always had adequate representation from older or disabled consumers, and information about what these groups find easy or difficult has been very limited. RICA's ongoing work on the usability of domestic appliances has thrown up many examples. For others see www.baddesigns.com
Design for all
It is worth restating the fact that design for all means just that. The basic idea is to create products which are usable by the widest range of people. To do this, there are strong arguments for carrying out research among groups which have the most difficulty. As the Netherlands Design Institute put it 'if designers succeed in understanding the problems and limitations of older users and apply this knowledge, a much larger group of non-model consumers will stand to gain - getting things right for them, might hold the key to getting things right for all of us'. An added gain is that consultation with older consumers can lead to more innovative products, including answers to 'urgent, sometimes completely new questions'.
"Today [older people] want products that do not draw attention to their owners' special needs and can be used - and desired - by all" Bridget Bodoano, The Independent newspaper
The basic toolkit
All method of research are really ways of consulting with potential customers. They have different advantages and disadvantages and some are more difficult and more expensive than others. In this section we give you some idea of what these differences are, and where to go for more information.
Whatever method of consulting users is selected, it is important to schedule it early in the product design or standard making process.
Research can be formal or informal. The difference is in structure, size and approach and in the authority and reliability of its conclusions.
"Our research shows that the design of many mainstream products is disabling - it excludes potential customers. It is time to increase marketability with design for a broader spectrum of the population" RICA
Survey researchers tend to divide research into quantitative and qualitative studies. Within these very broad categories there are very many individual techniques. Designers have their own research typologies. For example, IDEO www.ideo.com , an international design consultancy, have developed a system of some nine broad categories which further divide into over 50 sub categories.
The broad headings are
- Co-design
- Co-research
- Current customer information
- Direct design experience
- Expert Observation
- Future Creator
- Imagine and Act out
- Professional trackers
- Stimulus and interview
This classification has been used by the Presence Forum, a European project that ran from 1998 to 2000, in its review of user research for design, published as The Methods Lab (see Contacts).
The European Institute for Design and Disability campaigns for inclusive design, and can also provide guidance.
However the research is described, it is essential to be clear about what you want to find out, and then choose the appropriate method.
Below we describe some of the main ways of focusing on ease of use/most qualitative techniques. These are often more appropriate for development projects and are cheaper than large scale work. Much of the research described by the Presence Forum are applied versions of these techniques.
Consultation
The views of those who have relevant professional experience is essential. These could include designers, but also people experienced in the needs of elderly and disabled people, such as occupational therapists.
Desk research
This is basically finding out what has been done, who is in the field, what is being researched and what has been published. It saves reinventing the wheel and should tell you who the experts are. Literature searches can be organised through research libraries or carried out directly using bibliographic databases or through the Internet. However searching is a skilled job if you want to be sure that your search is comprehensive. Literature sources may include, for example, specialist journals, reports on evaluations, specialist bibliographies, published information, indices of current activities in universities as well as standards.
It can be worth thinking laterally about the areas searched - findings from research carried out in one product area may be relevant to other areas.
Most research reports give details of methodology and this can be invaluable.
Quantitative research
These are larger scale studies which include surveys and censuses of different kinds. They may be used by manufacturers to test the acceptability of new designs or in their general market research.
Sample surveys
A large number of users is selected to represent a given population. Surveys are a way of collecting people's views and experiences and can quantify and test hypothesis and generally attach number to things. However they can only explore subjects to a limited extent. Surveys can also be used to collect information about long term experiences, reliability, mishaps or accidents. They can be carried out by telephone, personal interview or by post or direct distribution. It is important to ensure that potential users are not excluded, for example street surveys do not reach people who cannot get out and about much.
Information about market research is available from the European Society for Opinion and Marketing Research, www.esomar.nl. Discussions on survey methods are published regularly in the International Journal of the Market Research Society, quarterly, MRS, London, www.mrs.org.uk.
Qualitative research
This takes many forms. They usually collect information from smaller groups of people, but in greater depth. Traditionally qualitative research has been used as a means of exploring concepts and revealing issues to be quantified by larger scale surveys, although the second quantitative stages are often dropped.
Qualitative research is flexible, good at revealing underlying principles, can be fast and is usually cheaper than larger scale projects. Its findings - supported by a wealth of direct quotes, photographs and other materials collected in the course of the study - can be extremely persuasive. On the other hand there is the danger that small samples may represent only themselves, limiting their usefulness.
In the context of design for all the aim of exploratory research is to find out more about how the product is used by consumers - this includes not only how easy it is to use, but how it fits in with the environment, and people's lifestyles and aspirations.
Observation
Observing and interviewing users of products where the product is normally used can be revealing. Consumers may use - and misuse - products in unsuspected ways. Sequences of operation, problems and interaction with different environments can become immediately apparent.
The main advantage of observation is that users themselves may be unable to answer certain questions fully. They are unlikely to notice every potential hazard or be aware of adopting an awkward posture, for example. Users may not be able to give accurate performance data - such as how long it took to complete a task. And users perceptions will be related to experience - those who have already tried a better product may be less satisfied that those who have not.
Crucial to the success of observation techniques is the training of observers - it is a more difficult job than it appears. For the results of the user trial to be reliable, the observers must be consistent in their own judgements and comparable with each other.
Unstructured and semi-structured interviews
These are interviews which do not depend on a list of predetermined questions. In unstructured interviews the researcher formulates questions around a list of broad subject headings and is free to follow any leads or develop points. Semi-structured interviews are similar except that the list of subjects is more detailed and specific. In both cases respondents are encouraged to talk freely around the set topics. Respondents are all asked for certain information yet can develop their views at length. Interviewers can choose how to put their questions and how much to explore and probe.
Group discussions and focus groups
Here a group (usually between 6 and 10 people) discuss a subject, guided by a moderator. Their strength lies in the interaction between participants. Group members remind each other of experiences, opinions are tested by being bounced from one member to another. They provide a good forum for discussing products because examples or prototypes can be on hand to stimulate responses.
Assessment based on ergonomic data
Measurements of the size and shape of body or parts of the body and other anatomical data such as ranges of movement and strength or dexterity are known as anthropometric measurements. These are usually collected for a sample of people who represent typical users of the product being tested. This information can guide product designers.
However although some relevant measurements exist, not enough information has existed about the characteristics of elderly people or about the actions they find easiest (shapes gripped, force available to turn something, for example). An excellent source has now become available with the publication of Older Adultdata (see Contacts).
Other contacts for anthropometric information
Robert Feeney and Associates (RFA) have carried out large scale trials with disabled people on reach characteristics and space requirements when carrying out everyday activities.
International Ergonomics Association (IEA) is the association of ergonomics and human factors societies around the world www.iea.cc
Human Factors and Ergonomics Society www.hfes.org
ErgoWeb is a commercial ergonomics company providing consultancy, training, innovative software solutions and information internationally www.ergoweb.com
Expert assessment
Opinions are collected from experts, based on their knowledge of the type of product and how and where it might be used. Experts in design for all are disabled people, occupational therapists or technicians.
User trials
These are the most common form of ergonomic assessment. They involve those who use the product, or could do so. They can take place in users' homes, a laboratory, test house or in any other appropriate environment. Users' reactions are recorded. Some user trials are observed.
In a laboratory The major advantage is the degree of control you have in managing the trial. It can be structured so that each person uses the product in the same way - following the same order of operation and spending the same time on each task. More elaborate measuring techniques can be set up in a laboratory, and this may be important for some products. Closer observation and supervision of the user test is also possible. Users can be helped if they have problems in completing a questionnaire or get into other difficulties.
The major drawback is that the setting is unnatural. However well the domestic environment is simulated, a laboratory is still a laboratory. Another problem with laboratory user tests (though not confined to them) is the effect of the trial supervisor on the behaviour of users. People may try to do what you expect of them - even if you are not aware of showing any kind of bias. There are also the practical problems of scheduling and getting people to the site.
In the field These will generally take place in users' homes, though other natural settings, such as in the street or in shops, may be appropriate. The main advantage is that the user is on home ground. Problems involving the interaction of the product with aspects of the home setting are more likely to be demonstrated. Users will almost certainly feel more relaxed and therefore perhaps less anxious to 'please' the test supervisor. A major problem is the lack of control over test conditions. Each user's home will be different and for products which have a significant interaction with the environment, this may be very relevant.
Unsupervised trials It is also possible to conduct unsupervised trials in users' homes. Instructions are provided with a questionnaire for the user to complete. One of the advantages of unsupervised user trials is their reduced cost. Less time on the part of the project staff is required and this may mean that a larger number of users can be included.
A final advantage is that the environment in which the product is used is totally natural. Not only is the user in his or her own home, but there is also no stranger watching. The major drawback of unsupervised trials is the lack of control over the experimental situation. There is no way of knowing how the products really are used, and no observation of areas of misuse or hazardous use.
Selecting users and respondents
Whatever method of research is chosen, the key decision revolves around who is to participate in the trials. A rule generally adopted is that users should be 'ordinary' people in that they have not made any special study of products or equipment.
Within that, it is usually necessary to devise some kind of classification system in order to ensure that people with different characteristics are included - both sexes, a cross section of age ranges for example. Large studies can structure a sample which is representative of the whole population of the people under study and may even be able to choose people randomly from this population.
In practice this is rarely possible. With small samples it is especially important that great care is taken in the selection of people, because great weight is given to a small number of observations. For ease of use assessments there are persuasive arguments for choosing people who represent extremes in the population. This should cover anthropometric differences and other characteristics which are relevant - psychological, physiological or environmental, for example. The outcomes of this research are much more likely to embody design for all principles and, as RICA has consistently found, designs which accommodate the people who have the most difficulty are easier for everyone else too.
Next Page: Product group checklists