User Tools
Site Tools
Sidebar
design:exploratory_methods
Table of Contents
Exploratory Methods
Here is a brief summary of some methods that can be used to help designers understand design problems. (Based on [Jon92])
Stating Objectives
- Identify the situation within which the design is to operate.
- Identify features of this situation with which the design must be compatible if it is to be accepted by sponsors. Compatibility with these features is the ultimate objective. They include:
- the sponsors' expectations and the reasons for them
- the resources available
- the essential objectives
- Ensure that statements identifying the objectives are compatible with each other and with information that becomes available while designing.
Literature Searching
- Identify the purposes for which published information is being sought.
- Identify the kinds of publication that are likely to contain information that can be relied upon for such purposes.
- Select the most relevant of the standard methods of beginning a literature search, such as
- encyclopedias and other reference texts
- library catalogues, journal abstracts, and indices
- librarians and industrial information experts
- individual experts by phone, email, post, interview, etc
- personal network of colleagues
- Minimize the search cost by allowing for retrieval delays and by continuously evaluating both the choice of sources and the applicability of collected data
- Keep accurate and complete references to documents that are found to be usable
- Keep local collections of publications sufficiently small and temporary to permit rapid retrieval (somewhat obsolete with respect to massive online sources currently available)
Searching for Visual Inconsistencies
- Examine samples and photographs of existing designs.
- Identify apparent inconsistencies and contradictions in the arrangement and purposes of design components.
- Infer reasons for these inconsistencies and look for evidence of causes of design change.
- Envisage ways of removing inconsistency and of adapting to the inferred external causes of change.
Interviewing Users
- Identify user situations that are relevant to the design situation that is being explored
- Seek the agreement of all the persons within the user situation who could be affected by the interviewer's presence or by a new design
- Encourage users to describe and to demonstrate any apsects of their activity that are important to them
- Direct the conversation toward aspects of the user's activity which seem relevant to the situation that is being explored
- Record both circumstantial and critical findings during the interview or within a short time of it
- When appropriate, obtain the user's comments on the conclusions drawn from an interview
Questionnaires
- Identify the design decisions that are to be influenced by replies to the questionnaire
- Identify the kinds of information that are critical to the taking of these decisions
- Identify the kinds of people who have rapid access to the kinds of information needed
- Carry out a pre-pilot investigation to gain insight into the knowledge of potential respondents
- Write a pilot questionnaire that fits both the known principles of questionnaire design and the particular situation, including
- Ask for the minimum information needed for the purpose at hand
- Ask questions that the intended respondents are able to answer
- Require simple yes/no, or numeric, or multiple-choice style, answers
- Ask questions most likely to be answered truthfully and without bias
- Do not be unnecessarily inquisitorial (think of ridiculous “government forms”)
- Circulate the pilot questionnaire to test the questions, the variability of the answers, and the method of analysis
- Select an appropriate sample of the kind of people having rapid access to the information that is sought
- Administer the questionnaire
- Extract from the replies the data most helpful to designers
Investigating User Behaviour
- Consult and observe experienced and inexperienced users of similar equipment before designing the new
- Carry out a man-machine systems analysis to define tasks, user abilities, and the design requirements for interfaces (the parts of the product that affect users)
- Observe or simulate critical aspects of the behaviour of both skilled and unskilled users of the proposed design
- Record limiting values which must not be exceeded if users are to carry out the actions required without error, injury, or discomfort
Systemic Testing
- Identify the features of the existing situation that are not as desired
- Identify sources of highly variable behaviour within the existing situation
- Apply or relax gross constraints to these sources of variability and record the effects upon the features that are not as desired – record also the effects upon other features in the situation
- Select the most promising and the least harmful of the tested constraints as an avenue for planning and achieving the desired changes
Selecting Scales of Measurement
- Pose the questions that are to be answered by measurement
- Determine the acceptable error and the acceptable cost of measurement
- Select an appropriate measurement scale – there are six
- Nominal – colours, nationalities, smells, professions, area codes
- Partial order – ordered by single of few of all possible criteria
- Ordinal – absolute ordering by rank
- Interval – degrees centigrade
- Rational – metres, dollars, ohms
- Multi-dimensional – mpg, mental age, force
- Plan the measuring procedure to be compatible with the above
Data Logging and Data Reduction
- Identify the uncertainties that are critical to the success or failure of the range of design options under consideration
- Identify the degree to which critical uncertainties have to be reduced
- Identify the time and resources available for reducing critical uncertainties
- Review the available methods of data logging and data reduction, noting, in each case, the accuracy, speed, cost and the kinds of question that can be answered
- Select methods of data logging and data reduction that are compatible with the above and with each other
- Continually check the relevance of partial results to the critical uncertainties and alter the procedure as necessary
References
[Jon92].
J.C. Jones. 1992. Design methods. John Wiley & Sons, New York.
design/exploratory_methods.txt · Last modified: 2020.03.12 13:30 (external edit)
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 4.0 International
