INTRODUCTION

The U.S. Federal requirements for accessibility and accessible design are expected to have a profound effect on the market. It is important that companies be able to offer accessible devices and software, and to be able to document the accessibility of those products. In the context of Section 508 and other related requirements, accessibility refers to both the physical accessibility of a device (i.e., the ability to access the physical features of the device), and the functional accessibility of the device (i.e., the ability to exercise the functions of the device). Thus, an accessibility evaluation must address both physical and functional aspects of a device.

The physical aspects of a device that must be accessible include the displays and controls by which the user operates the device, as well as other structures such as lids, cables, and access panels. The ability to position one's self in an appropriate operating position is also part of physical accessibility, and is particularly relevant for users who use a wheelchair. The functional aspects of a device that must be accessible include use of modes, menus, presets, programmable keys, and other features that are typically controlled by software.

Accessibility is not a simple matter of yes-it-is or no-it-isn't. First, a device may be accessible to users who have one impairment (e.g., a speech disorder), but be inaccessible to users who have a different impairment (e.g., a vision disorder.) Second, a device may be accessible to users performing some tasks (e.g., making a copy), but inaccessible to users performing other tasks (e.g., loading paper into a copier.) A complete description of accessibility, therefore, requires separate consideration of various impairments and user tasks.

A useful way to categorize tasks, for describing accessibility, is to divide them into primary, secondary, and tertiary tasks. In this scheme, primary tasks are the tasks associated with the core functionality of the device. The typical uses of a device involve primary tasks; these tasks may also be called Priority 1 tasks. Secondary tasks are tasks associated with advanced features or ancillary functions, and may be called Priority 2 tasks. Secondary tasks are usually performed far less frequently than primary tasks. Tertiary tasks are tasks that users rarely perform, and might be restricted to users with special training. Tertiary tasks may be called Priority 3 tasks.

Categorizing tasks in this way allows accessibility to be characterized at three levels. Level-1 accessibility refers to the accessibility of the device with respect to performing primary tasks (Priority 1 tasks). Level-2 accessibility refers to the accessibility of the device with respect to performing secondary (Priority 2) tasks, and Level-3 accessibility refers to the accessibility of the device with respect to performing tertiary (Priority 3) tasks.

The two-dimensional nature of accessibility is illustrated in the following table. Eight common categories of impairments are crossed with the three levels of accessibility, producing 24 cells - each requiring an assessment of accessibility.

It could be argued that tasks could be categorized into more levels, and that impairments could be categorized in greater detail. The level of detail reflected in the table above provides a useful aggregation of data that would be generated at greater levels of detail. Aggregation at this level allows all devices to be characterized in the same dimensions. Greater levels of detail in categorizing tasks would likely produce categories that do not apply to some devices, whereas the three categories shown above do apply to all devices. In the extreme, one could list each separate task that a user could perform with a device, and list every possible manifestation of impairment. If one did so, the characterization of accessibility would be unique for each device and would make it more difficult to compare similar devices. To use greater detail than shown in the table above is therefore of questionable utility in providing a summary of the accessibility of a device.

The accessibility evaluation methodology documented in this report was developed by the Georgia Tech Research Institute (GTRI) as a method of measuring the functional and physical accessibility of a product. This methodology is summative in nature and is, therefore, suitable for the evaluation of existing products or products nearing the end of their design cycle. The general purpose of a summative evaluation is to determine if the product meets some set of defined requirements. In this case, accessibility is the requirement that is being evaluated. In contrast, a formative evaluation is designed to provide feedback to the design team in order to further enhance or refine the design. Additional methods and considerations would need to be developed before this methodology could adequately be applied to formative evaluations. The methodology described in this report is oriented to producing definitive outcomes rather than good ideas for further consideration. Additionally, this methodology is designed to evaluate a single product and does not specifically address the comparison of two or more products to determine which is best, or whether a new product is as good as an established competitor.

The accessibility evaluation should not be confused with a general usability evaluation. While data may be collected on the general usability of a product, the focus of this evaluation is to determine if the product is accessible. A product may be accessible and still not very usable. A problem that affects all users equally is a usability problem and not necessarily an accessibility problem. For example, a well-designed user interface, from an accessibility standpoint, could be designed so that it is difficult to recover from mistakes. In this example, the accessibility of the system may be judged to be good while the usability of the system may be judged to be poor.

The accessibility evaluation methodology contained in this report is suitable for the evaluation of hardware, software, documentation, and customer support. In fact, in many cases, an evaluation will involve all four components. For example, the evaluation of a multi-function printer, scanner and fax machine, may involve hardware testing of the actual device, software testing of the device drivers and supporting utilities, documentation testing of the manuals, and customer support testing of provided services.

A primary reason for basing evaluations on tasks, rather than components or features of a device, is that Section 508 requires an outcome, namely, that the device be usable by individuals with certain impairments. It is therefore necessary to consider what users actually do when they use the device, rather than simply considering the various components or functions of the device.

If a given component is involved in a large number of user tasks, and that component cannot be used by individuals with a certain impairment, then all the tasks that require use of that component are rendered inaccessible even if other components also involved in the tasks are accessible. For example, imagine an office appliance that has a touch screen that is inaccessible for blind users. If all tasks associated with that appliance require use of the touch screen at some point in performing the task, then the device is completely inaccessible to a blind user, even if there are a number of other components of the device that are accessible.

This methodology calls for two separate evaluations. The first evaluation is a checklist evaluation that is to be performed by a qualified human factors expert or accessibility professional. GTRI has compiled a detailed checklist from guidelines found in Section 508, Section 255, the accessibility literature, sound human factors standards and guidelines, and our own design experience. The checklist consists of items that lend themselves to direct measurement or observation. These checklist items can be found in Appendix A. The second evaluation involves user-in-the-loop testing, which is required to evaluate the accessibility of the system under test for specific population groups.

Previous Page

|

Next Page

| Back to Table of Contents