The methods in the research of usability do not in principle differ from the studies of other goals for products. We can differentiate three approaches, each of which have a different relation to the application of the results:
Because of the great variety of products and their uses, almost all studies of usability focus at one given type of products only. We can only take some of these studies as examples - there are very little studies of usability of products in general (as a contrast to the studies of, for example, the aesthetics of products).
In the descriptive type of studying a product the target is to register its usability. On a rudimentary level, such registering is made continuously in every large industry in order to confirm that each departing product item is usable. If it isn't, then the item will be rectified or rejected. Indeed, usability is the principal concern in most systems of quality control of manufacturers. The same is true for the systems that many countries have set up for verifying the healthiness of foodstuffs and the safety of vehicles, and the public critique that exists for some products. However, these audits of usability are usually not regarded as research because they consist mostly of simple measurements only, they do not analyze the material later, and no general patterns or generalizable theories result from them.
To be sure, such studies have also been made which do seek general theory. They focus almost always on a specific type of products. For some products, such as architecture and furniture, there are historical studies which illustrate the use and usability of their objects, beside their other properties. Besides, there are sociological, psychological and ethnographic studies where the use of a given product is pivotal. An example is Bostadssociologi [Sociology of dwelling] by Holm and Boalt. Then there are studies which analyze the usability of various products when used by the elderly and other people with reduced abilities.
The general principles of descriptive study are explained on the page Descriptive Theory.
The empirical gathering of facts about product use is seldom possible without users, especially when studying products of the interactive type. There are several alternatives for gathering facts from the users of a product:
When planning the structure of concepts and questionnaires in your study you can make use of recent reports from studies of interactive usability, as explained in the paragraphs Usability as a measurement and Subjective usability measurements on the www-page Usability of Interactive Products. If you are in doubt about which concepts belong together and which not, you can test it by submitting the data to analysis of variance or factor analysis.
The analysis of observations and users' opinions can be done with qualitative methods, if the study consists of a small number of cases only (see Case study). When analyzing larger amounts of material a quantitative format of data is often preferred, because it allows using powerful methods of Quantitative Analysis. A temporal view of the development of objects or their use can be constructed with variables (Time Series) or with qualitative factors, see Explaining a Development; the latter method can be useful when studying changes in the behavior of consumers and in the use of products.
Generally can be said that usability is so heavily value-loaded a concept that it is hard to differentiate between the descriptive and normative approaches in its study. Accordingly, it is perhaps unnecessary to recommend an objective (or "disinterested") attitude for an descriptive study of usability and allow subjective attitudes only in the normative work. Besides, it is clear that sometimes the normative aspect lies hidden behind the descriptive study as well. The reason for staying (perhaps just initially) on the level of information and not proceeding directly to action can be that the relationships between important variables related to usability are not yet exactly known and more studies on a descriptive level are needed before it will be possible to formulate normative recommendations. There are numerous examples of excellent studies that include an descriptive phase and a normative final phase, for example the study of the usability of home kitchens for the elderly, by Keiski (1998).
Theories of design and of fabrication are those bases of knowledge that are generally used in planning these two activities. "Generally" means that these theories are applied in more than one enterprise and in the manufacture of many if not all the products of a certain type. Modifying these theories is thus an effective method for improving current products or their fabrication processes.
To be sure, theory is not always the sole basis, not even the most important one, for design today. Another basis is often found in the exemplars given by earlier products. This concerns especially questions of use and usability, because both normally change only a little if at all when a new generation of products comes to the market. Therefore the designers of products can often base their work on the assumption that the novelty will be used very much like the preceding one. No new research will in this case be necessary, if all goes well.
However, exemplars cannot eternally supplant research and theory, because the use of even the most traditional product will change some day. Besides, sometimes a completely novel product must be created from scratch - for example in the field of informatics - and some products such as large buildings and ships are so complicated that no exemplar for them can be found. The designers of these demanding projects would then need to do much research, often in great hurry, if other researchers had not already done at least part of the work and provided a generally applicable basis for design: a design theory.
It was said earlier that when studying usability - a definitely desirable characteristic of products - the normative aspect is nearly always present, at least indirectly. This actualizes taking up the question of viewpoint: whose subjective assessments shall be recorded in the descriptive study and whose wishes shall steer the normative proposals?
The answer may seem simple: the user's viewpoint is normally the right one. But who is the user? Many products are marketed in various countries where the opinions of people can be very different. Besides, the product sometimes generates not only benefits to the user, but also disadvantages to other people. Therefore the researcher should try to keep a clear account of the distribution of benefits, disadvantages and opinions between people and in time, see Point of View.
When selecting the members of the research team, two types of people would be especially welcome: experienced users of comparable products, and also persons from the intended target group of customers. People of the latter type are often difficult to find and a group of volunteers is used instead. It will bring about some bias into the results obtained, but the damage can be corrected in the later phases of the project, when there are better chances to locate appropriate potential customers for testing the product proposal.
The philosophy behind normative theory is explained in Preparing Design Theory, as well as its normal procedures. There is a summary of recent studies, especially concerning interactive products, in the paragraph Usability as a design approach on the page Usability of Interactive Products.
The target of economy sometimes gets mixed up in the study of usability. Often there are alternative ways of acquiring the desired service from the product, and the costs are different, which gives occasion to economic optimization. Some usual models for optimization are explained under the title Normative Study of Economy.
Presentation of the findings can be difficult if these consist of novel modes of activity around new products which do not yet exist physically. One possible mode of presentation is the format of storyboard or "comic strip" like the one on the right, from Keinonen, 2000, p. 217.
The forum of publication of the proposals may need some consideration so that they reach the right audience and maximize the practical outcome, see General Normative Report, on the page Reporting. The most usual format for design theory is a handbook for the designers of a certain type of products. These usually aim at enumerating all the customary requirements that products are expected to meet, and on this list usability is often the first.
Other usual channels of publication for design theory are standards, either national or international ones. They can be either compulsory or voluntary, and confirmed on various levels from international organizations to private companies. Voluntary standards are often developed as joint projects of the major manufacturers of a certain product in a country.
Some requirements of usability have been judged so important by the officials in various countries that these have authorized the specifications as governmental regulations. These usually stake out the allowable minimal or maximal limits for important variables of products, but do not otherwise restrain the design.
Usability is often the most important requirement for a product, and to achieve it in new products, product design is the most direct and most effective method. Indeed, the aspect of usability is normally present in all phases of a product development project.
Traditionally, the starting point in product design is an existing product or a new product idea which the designers then refine to be more usable, but lately a few products have originated the other way round, starting from a novel mode of activity. The designers have first invented a novel activity and first thereafter or simultaneously they have created a product which matches the activity. Examples of these are some playconsoles and camera-telephones. On the right can be seen a product idea which aims at preparing to people who live in normal apartments a possibility to a novel activity: to train and simulate mountain climbing. It was created by Decathlon, a French producer of sports equipment (Keinonen et al., 2004 b, 78).
Developing a product idea. The greatest chances to radically new thinking in product design is in its initial stages: in the tentative product concept phase, often called "product idea" or "design driver", when it is relatively easy to modify any property of the product, including its usability.
The choice of method for studying the use of a product depends on how far you intend to proceed from the present state of things.
If you just want to remove the problems encountered in the use of existing products, it is often easiest to use extrinsic methods, like Methods Engineering, or the customary methods of descriptive observation, questionnaire or interview. The researcher can also try to clarify the context of the future activity, by using the the methods of Forecasting.
The extrinsic approach is usually restricted to one or two dimensions of product usability. The researcher can select these, for example, from Shackel's categorization (1991, 24, figure on the right).
If you wish to go deeper and develop new modes of use, perhaps even new products to cater for these modes, you will often find that when using the extrinsic approach it is difficult to keep in mind and even think about all the possible dimensions and variations of using a product, despite the helpful models of Shackel and others. To handle these variations, one possible approach is to abandon the study of distinct dimensions and instead study holistically the use of the product, taking in account simultaneously all the aspects that it can entail. Such a study can be difficult for a researcher who works in the extrinsic way, perhaps alone, and it can often better be done in a small group where experienced users (or expected future users) of the product try and prove various models of the product and discuss them. Suitable techniques for this type of study can be adapted from:
In the beginning the development team perhaps only comes to think about existing products but usually it will soon expand its deliberations to new, improved products that have not yet been produced.
However, just discussing hypothetical activities does not stimulate innovation as much as the presence of a tangible instrument of action. Now exactly the real instrument of action is usually not yet available, because it can be created first when the development project is finished. Therefore a substitute must be used, such as:
Presence of a mock-up or an exemplar allows using some special modes of innovative creation, like:
In the above modes of study one or more members of the team serve as actors and the rest of the team as audience. Subsequently the entire team shifts to normal innovative conception of potentially feasible designs. Such a shift is, indeed, typical to many methods of innovation: a phase of free fantasy is followed by more realistic outlining. In the beginning of the innovative work the development team is thus allowed to neglect seemingly imperative practical restrictions, postponing the solution of technical problems until the time of eventually applying the new ideas into practice.
The specification phase. Usually the most research-intensive stage in product development is the phase of detailed product concept or specification, where the largest amounts of data are processed and exact requirements for the new product are enumerated. This is the right moment to specify the necessities concerning usability as far as they can be clearly expressed or measured.
For example, the requirements concerning usability of a new building are usually presented as a list of all necessary rooms together with their areas, intended functions and for this purpose needed equipment. Beside the list, there often is a diagram like the one on the right, showing desired connections between the rooms.
It has to be noted, however, that it is often quite difficult to specify in advance all detailed usability requirements, and far from all new design projects in practice include any detailed definition of them. As was already remarked above, it is possible to replace specifications with exemplars given by earlier products, especially in the questions of usability. This is because the use of products normally changes only very slowly, and therefore the designers of products can often quite successfully base their work on the assumption that the novelty will be used very much like the preceding ones.
The form-finding phase. In the synthetic phase of design more and more details of the product become fixed and radical changes thus become more difficult. Fruitful moments for generating improvements to a draft of design could still in this phase be the sessions arranged in order to activate innovation such as brainstorming. In the final stages of design the feasibility of modifying the product diminishes, but of course changes must be made if the final evaluating of prototypes or market testing indicates such a necessity.
Although the normal design process described above includes doing first a study of functional requirements and then designing on this basis, it is worthwhile to keep in mind that a design with very scant tolerances can have its risks, too. For example, a kitchen that has exact the right dimensions for the present activity, can become too small when a new type of kitchen furniture should be added. Instead, it can often be advisable to make the product modifiable or flexible so that it can be used in different ways. For example, a person car could be modifiable for transporting goods, or an office building could be designed with thin internal walls which can easily be moved. Sometimes flexibility can be achieved simply with slight over-dimensioning.
August 3, 2007.
Comments to the author:
Original location: http://www2.uiah.fi/projects/metodi