GOMS is the acronym of the term Goals, Operators, Methods and Selection rules. It is an evaluation model developed by Stuart Card, Thomas P. Moran and Allen Newell in 1983. The model is a specialized human information processor model designed to observe human-computer interaction. Based on this GOMS model, many other models for analysis have been developed. In the following post, we would look at what constitutes the GOMS, the advantages & disadvantages of the GOMS and how GOMS has evolved over the years.
Goals are defined as what the user wants to accomplish. Operators are the actions performed by the user to accomplish the goal. Methods are a combination or a sequence of operators used in order to accomplish the goal. In many cases, there is more than one method that can be used to achieve the same goal. This is where the Selection rules come into play, where the rules are used to describe when a user would choose a method over another method.
By breaking down a user’s interaction with the computer into these 4 simple elementary actions, the interface can be studied. The evaluator can also pin-point on specific aspects to improve on.
Advantages & Disadvantages
The GOMS may not be the most accurate of all the evaluation methods available, but like all methods, it has its advantages and disadvantages.
One advantage is that it is fairly simple, cost-efficient and less time-consuming to calculate the GOMS estimate of an interaction. In order to do so, the average methods-time measurement data for each specific task has to be previously measured experimentally, and needs to have a high degree of accuracy. By using the GOMS, each step that the user uses to interact with the interface is broken down into detailed steps. For each of these detailed steps, the time used can be measured and the total time a user needs to complete a desired task is just a simple addition of the time used for each of these detailed steps.
However, the major disadvantage of the GOMS is pretty similar to most disadvantages of HCI evaluation methods. That is, the GOMS is not able to handle user unpredictability. The GOMS method relies heavily on the prediction of user methods. It does not take into account user behaviour such as fatigue, social factors, etc. Furthermore, GOMS assumes that the user knows what to do throughout the whole process of completing the task, as such, GOMS applies to veteran users and not those who are new to the interface.
Also, as we can see from the overview of the GOMS, only the usability of the system is considered, not the functionality. What this means is that the evaluation does not improve on the functionality of the system.
Furthermore, out of all the GOMS models (which we will look at in a short while), only the KLM method does not require a deep understanding of GOMS in order to evaluate. As such, if a company decides to use GOMS for evaluation, they would need to hire someone who has the expertise of GOMS evaluation in order to effectively make use of the GOMS.
Evolution of GOMS
The very first version was the plain GOMS created by the original three founders, which is now commonly referred to as the CMN-GOMS, taking after the names of creators, Stuart Card, Thomas P. Moran and Allen Newell. What this method does is that it follows a rigid goal-method-operation-selection rules structure. This structure allows the evaluator to represent all the tasks in a pseudo-code format, and at the same time, a guide is provided to assist in formulating selection rules. The method is also able to estimate how much time it takes for a user to complete a task.
Based on the CMN-GOMS, the keystroke-level model (KLM) was developed. This model is an 11-step model that can be used to estimate time taken to complete simple data input with just the mouse and the keyboard. With KLM-GOMS, it is easier to use than the other GOMS, and evaluators often find more efficient ways to complete a task by analysing individual steps and sift out unneeded steps. The KLM-GOMS is best suited to evaluate tasks that take, on average, less than 5 minutes to complete, due to its constraints. The 11 steps are as follows.
Step 1–Obtain a working prototype of computer interface or a step by step operational description of a task.
Step 2–Identify the goals or the desired outcome of work .
Step 3–For each of these goals, find subgoals or tasks that achieve the main goals.
Step 4–Identify methods to main goals and all subgoals.
Step 5–Convert description of methods to pseudo-code (the terminology that is described above).
Step 6–State any and all assumptions used in the making of pseudo-code and goals.
Step 7–Determine appropriate mental or keystroke operators for each step.
Step 8–Assign time values to mental or keystroke operators.
Step 9–Add up execution times for operators.
Step 10-Adjust total time of task to be sensitive by age of expected.
Step 11-Verify validity of results
CPM-GOMS is the Cognitive Perceptual Motor GOMS model developed in 1988 by Bonnie John, a former student of Allen Newell. It is basically similar to the CMN-GOMS method in most of its model, except that it determines which actions can be done parallel at the same time and create blocks whereby the many actions are performed at the same time. What this means that CPM-GOMS allows for multi-tasking when evaluating the time needed to complete a task.
A more in-depth look at the different GOMS, together with examples can be found on http://members.tripod.com/elena_chmil/thegomsmodel/id10.html.
For most parts, the GOMS is not a very advanced method of evaluation, but it is able to provide us with a fast and efficient evaluation method to calculate time required to complete a task. The way GOMS has evolved over the years have also provided us with different methods that we can use to evaluate human computer interaction.