There are a variety of quantitative methods to evaluate an interface. One particular system is the Goals, objects, methods, and selection rules (GOMS) which first became popular in the 1980s. This system works particularly well for a worker given a specific repetitive task and a computer. As a task becomes more reptitive, efficiency becomes increasingly important. Even a few seconds saved for each iteration of the task makes a large difference. Interface designers consider quantitative mesasurements such as GOMS to be a tool in an aresenal of techniques needed to analyze an interface.
GOMS predicts the time it will take a user to execute a task using a given interface. It breaks down tasks into four different serial elemental gestures. During the development of GOMS, laboratory experiments determined the average time it took a user to perform each of the four gestures listed in the table below.
|K = 0.2 sec; Keying, time for pressing a key on the keyboard.||P = 1.1 sec; Pointing, time required for pointing to a position on the display.|
|H = 0.4 sec; Homing, the time for a user to move his hand from keyboard to mouse.||M = 1.35 sec; Mentally preparing, time to prepare for the next step.|
Along with these symbols (K,P,H, an M) there is a complex set of rules established to determined how to place the symbols when analyzing an interface. To provide an overview of how the system works, the rules necessary to make the GOMS analysis are not presented, but simply the final GOMS formulation.
Let's explore GOMS analysis for software which needs to convert temperature between Celsius and Farenheit (example adopted from Raskin, Humane Interface). Imagine an interface like this:
Here's another completely different approach to solving the problem.
The most efficient interface would require only the keystrokes of numbers representing the numerical value needing to be converted.