A Tail of Two Mice

Experiment

Introduction and Hypothesis

In a Microsoft Windows computer environment, a mouse is a tool that users
have become accustomed to and dependent upon to complete daily tasks.
Its invention has allowed the concept of "Direct Manipulation" to be possible
as well as powerful. The mouse can give the user a sense of control over their
computer (provided the interface is designed thoughtfully). The question naturally
arises as to what variety of mouse is "ideal" and for whom this is so. We will
attempt to show that a regular two-button mouse with additional scrolling bar is
an efficient choice for finding specified information in a textual list. This variety
of mouse gives the user all of the same functionality of a conventional mouse, with
one extra feature, the "scroll wheel". This small, rotating wheel embedded directly
between the standard right and left buttons allows web pages or other applications
to be scrolled without moving the mouse in order to grab the bar along the side of
the page to pull it down.

We have a developed a piece of software with MS Visual Basic 6.0 that requires
users to complete a series of tasks consisting of finding a specified name in
a list of two-hundred names. This software will be described in more
detail in the 'Materials' section below.

Based upon our own personal experience with such devices as well as that of
colleagues and friends, we hypothesize that the completion times will improve
with each successive task completed with the "scroll wheel" mouse while the
times will remain fairly constant when the tasks are completed with a
conventional mouse. We also expect the times to be generally lower
when the scroll wheel is utilized. The reason for this is that the user will
not have to actually waste time to move the mouse so that he/she may grab
the scroll bar on the side of the form.

The following is a breakdown of what our experiment measured:

Dependent Variables:

Times to complete successive tasks
Subjective Satisfaction survey

Independent Variables:

Pointing device

Treatments of Independent Variable:

IntelliMouse with "scroll wheel"
PS/2 Standard mouse

Hypothesis:
The times for the successive tasks completed with the "scroll wheel" mouse
will improve, while the times for the successive tasks completed
with the conventional mouse will remain constant. Upon becoming familiar
with the "scroll wheel" mouse, the amount of mouse movements will be
reduced, and the time to complete the task will become smaller.

Pilot Study Results

When the software was first developed, it was run on a total of six subjects.
All of the subjects were college students familiar with a Microsoft Windows
environment and conventional web browsers. None of the subjects had used
a mouse with a "scroll wheel" extensively and had not formed a solid
opinion of the device prior to taking this test. The exact same workstation was used for all
tests, the only variable being the pointing device. The first three subjects were given a
conventional mouse, while the remaining three were given the "scroll wheel" mouse.

The results showed a fairly significant time difference between the pointing
devices. The learning curve we had expected was not clearly demonstrated, however the
times for the subjects using the "scroll wheel" mouse were distinctly lower than the others.
With regards to the experiment itself, we realized that eye fatigue was going to be a factor
we had not anticipated. The subjects had made comments about the strain on their eyes,
so we decided to give them the option to take a break in between each of the three sets of five names.

Subjects

The subjects we selected for our experiment were all college students. Each subject, as with the pilot subjects, were all familiar with Win32 computing environments. Each subject was familiar with conventional web browsers and familiar with standard mice. None of the subjects had used "scroll wheel" mice extensively nor formed an opinion about the device prior to participating in our experiment. Each subject completed the test procedures in identical environments with the only independent variable being the pointing device. Fifteen of the subjects used the Microsoft IntelliMouse with the scroll wheel and another fifteen used a standard Microsoft PS/2 Mouse.

Materials

The materials for this experiment include:

300 MHz Pentium II Computer running Windows98
Microsoft Visual Basic 6.0
Microsoft IntelliMouse
Microsoft PS/2 Mouse
Subjective Satisfaction Survey
Experiment Software (see below)

The test given to the subjects was developed in Visual Basic 6.0. The first screen is an introduction, describing the forthcoming task in detail. Once the subject clicks 'OK', he/she begins the test. The task is divided into three sections. A text box is provided at the top of the window indicating the name that the subject is searching for. (See Figure 1) The subject then scrolls through the list below that consists of two hundred names. When the name is found, the subject will click on it and if the correct name was found, the next name is provided. If the subject indicated an incorrect name, he/she is reminded of the name they are searching for and they must continue searching for it. Each task consists of finding five names in the list. After the first task, there is a short break when a new list of names is loaded and the subject has a chance to rest their eyes. The time for all three tasks is displayed after completing the entire experiment. There are three tasks in all as we intend to measure the improvement of times with each successive task.

Full experiment available for download Here:

Figure 1

Procedures and Problems

For each subject using the IntelliMouse, the experiment proctor gave a brief demonstration of the functionality of the "scroll wheel" and a brief overview of the task. Each subject was asked to sign the standard experimental consent form prior to beginning the set of tasks. The subject was then asked to begin the task when they felt comfortable and had no further questions.