This case study describes how training on keyboard access to Windows and hotkey macros enabled an employee with cerebral palsy to perform a complex task four times faster by keyboard than when using a combination of a keyboard, a mouse, and MouseKeys. She was able to complete a task that had previously taken over an hour in 16 minutes.
In her previous job, using a mouse resulted in painful back spasms that caused her to miss work. The spasms began to abate during this training program; six weeks later, she reported that the spasms were much less of a problem.
Although the cost to accommodate this employee was significant, her improved productivity should result in cost recovery within four months.
Joan's disability is cerebral palsy. She has good control over her left arm, but moves it slowly within a slightly restricted range of motion. She types with her left index finger and thumb. With her right arm, she has slightly less control and range of motion than with her left. Although she can type with her right hand, she prefers not to. Her speech is only slightly affected; she speaks clearly but somewhat slowly. Joan uses a powered wheelchair for mobility.
For many years Joan worked in a call centre. She later transferred to another department to work as an operational support person in a "paperless" office.
Joan verifies scanned documents against information stored in databases; issues certificates; and prepares correspondence.
To facilitate the comparing of documents, Joan and her co-workers have side-by-side monitors. The two monitors function as a single desktop; employees routinely drag documents from one monitor to the other, and resize documents to fit the screen. They use seven Windows-based applications, four of which were developed in-house.
The four proprietary applications were not designed with accessibility or usability in mind. Screens tend to be visually busy, keyboard support is spotty, and functional inconsistencies are common. For example, in one program, pressing Esc closes a certain window, has no effect on a second window, and exits the application without warning in a third window.
Joan accesses her PC with a Cherry mini-keyboard and numeric keypad (G84-4100 and -4700). She places the keypad to the left of the keyboard to give her easy access to Enter, backspace, arrow keys, and MouseKeys. When the workstation is set up properly and she uses both hands, Joan can reach all areas of the keyboard and keypad without undue strain.
Joan controls the pointer with MouseKeys and a regular mouse. She uses the mouse to move the pointer, keypad 5 to click, keypad + to double-click, and keypad 0 to drag-lock. In the past, Joan tried a trackball, but did not like it. She is not a good candidate for a touchpad.
Using a mouse and MouseKeys slow her down considerably. Although she can move the pointer and click on most objects in five to twenty seconds, there are many tasks — especially those that involve dragging — that can take several minutes. Sometimes she does not even complete these tasks. (Many of these operations can be done by pressing a single key on the keyboard, or by pressing a sequence of two or three keys.)
In addition to contributing to inefficiency, Joan experiences painful back spasms from using a regular mouse. In her previous job, she missed work due to mouse-induced pain. Because her range of motion is limited, she cannot maintain neutral working postures when accessing a mini-keyboard, keypad, and mouse. The three devices take up too much space.
Joan is very smart, but not technically-inclined. She works hard to learn new software, and tends to learn computer skills by rote. She internalizes sequences of steps by performing them repeatedly.
Joan's computer skills were rusty when I began working with her. For example, she had forgotten how to right-click; when to click and when to double-click; and that Windows has two different cursors (the insertion point and mouse pointer). Although she had used MouseKeys, she did not use them as effectively as possible.
Over a two-week period, I developed macros for Joan and provided her with training/coaching. (These measures were part of a complex accommodation that included building modifications, work area reorganizations, telephone modifications, ergonomic adjustments, and speech recognition training.)
Joan and I met nine times over a ten-day period for two- to five-hour training sessions. The training focused on keyboard techniques, but also included basic Windows and MouseKeys concepts and skills. My approach to keyboard access was to introduce general techniques rather than to have her memorize hotkeys. These techniques included working with menus, navigating to controls, task switching, discovering shortcuts, and using macros.
I developed 17 Macro Express macros to launch applications, workaround inconsistencies, and simplify multi-step procedures. I concentrated on tasks that Joan performed with difficulty. For example:
After about 30 contact hours, Joan was using 10 or 12 of the macros. She was much less reliant on the mouse; but when keyboard access was impractical or impossible, she was more likely to use MouseKeys than reach for the mouse. She also decided to try using her right hand for the directional keys on the right side of the keyboard.
The back spasms began to abate immediately. Six weeks later, she reported that the spasms were much less of a problem.
As she began to internalize new techniques and work habits, her productivity quadrupled: She completed a task that initially took her over an hour in 16 minutes. With practice, she may shave four to six minutes from this time. Experienced co-workers without disabilities perform the same task in about five minutes. Joan's supervisor recognizes that Joan cannot work as quickly as her co-workers, but is not overly concerned. He is obligated to meet production goals and to accommodate employees with disabilities to the point of undue hardship. Joan's accommodations do not cause undue hardship.
The cost to develop the macros and train Joan was approximately $4000.
Assuming that these accommodations increase the monetary value of her work by $50 per day, they will pay for themselves in four months.
Had this employee not been accommodated in the ways outlined in this paper, she may not have been successful in this job. Organizations may not be meeting their legal duty to accommodate if they fail to provide software customization (e.g., macros, user interface modifications, and scripting), alternative access training (e.g., keyboard-only techniques), and ergonomic adjustments.
I thank Gary Perlman and Jeffery Rose for their valuable suggestions on this paper.