User Involvement

User abilities, preferences and needs

The aim of the report on "User Ability, Acceptability, Preference and Need Assessment" is an in-depth analysis of the abilities, needs and preferences of potential users. In particular to investigate the potential users knowledge and needs with respect to word prediction and related tasks are important information when defining the functional specification of the FASTY system.

In the beginning of the project a potential user pool consisting of 58 primary and secondary users was set up by the partners FTB, ELI, IKuT, FUNDP, IGEL, MULT and FORTEC.

Two questionnaires were developed. A User Description Sheet (UDS) to be able to choose appropriate participants for this investigation. An extensive User Questionnaire with regards to the preferences and user needs in view of word prediction but also sections about the handicap, used software and hardware, PC usage and typing, and the readiness to be involved and to support FASTY.

The results of the user questionnaires are summarised below

• Nearly 75% of the primary users have mobility problems, some of them have extreme difficulties in PC operations. Half of the users have difficulties in communication.
• More than a third of the users do not use a standard mouse or a standard keyboard. Nearly half of the users do not operate the keyboard with both hands.
• Nearly all users will use the word prediction with MS Word or another word processing software. A third will use it also for Internet applications.
• Approximately a third of the users are still working with Windows 95. The remaining group is working with Windows 98 or higher. The primary users are working with older and slower hardware than the secondary users.
• 70-80% of the users are ready to be involved in or to support FASTY. Only half of the users are willing to supply private or office text for text collection and text collecting tools.
• About half of the primary users know word prediction programs even if they do not use them. Only a few users have some concrete ideas about word prediction.
• The majority of the users thought that a program with word prediction should be useful for them. They expect to write or type faster, to be able to write and communicate and to write correct sentences.
• Users want to have prediction with an insertion as automatically as possible. It is important that the predictor inserts more than a word, e.g. a space after a predicted word and after punctuations, capital letters after punctuations and a u after q. Abbreviation Expansion is also desirable.
• The user wants to choose the predicted selection with function keys or with the input device. The selection list should contain at least 1 or 5 predictions but no more than 10. The number of words in a prediction string should be adjustable.
• Grammar checking is desirable for a great majority of the users.
• A third of the users wanted opportunities of audible feedback.
• It is important to the users that new words are added automatically to the dictionary which should be editable.

Ethical Aspects

The Report on Ethical Aspects describes the actions and efforts that have been and will be taken to assure an appropriate ethical standard in the FASTY project and will serve as a guideline. The Community Directive on the protection of individuals with regard to the processing of personal data and on the free movement of such data (Directive 95/46/EC) together with the opinions of the European Committee of Ethic (ECE) have been the basis for this report.

The variety of subjects includes the behaviour towards the other partners involved in the project as well as the handling of collected data such as user questionnaires, user ability assessment data, and text samples. Since sensitive and confidential data is involved special care has to be taken in protection of the privacy of the participating disabled persons. Particularly persons suffering from any kind of communication disorder, which is one of the target groups in the FASTY project, deserve in general a high degree of protection.

All data should be transformed and kept in a form, which avoids a trace back to a specific user or permits identification of a specific user for no longer than it is necessary. This was a guideline in the development of the Text Collection Tool that was mentioned above.

Text Collection

Text was collected from users with the Text Collection Tool. As described before this text material is needed for the prediction part in the developing phase. Special care was taken to fulfil the ethical guidelines.

User Panel

The User Panel is in close contact with the developers and is asked for approval of user specific documents. Thus, a user centred design is achieved.