Simulation Levels in Software Training
By Anthony Karrer, Alan Laser, and Laura Sund Martin
Using technology to teach technology.
Web-based training is a common method for instructing end-users on how to work with software applications effectively. A key aspect to WBT programs is the use of simulations. However, even relatively simple software applications can be extremely complex and require a large range of user interactions. But building a simulation of every application feature makes the training module as complicated as the application. For this reason, instructional designers employ several techniques to simplify simulations for training, including screen capture, point-and-click, data input, multiple paths, and full simulation.
Level 1: screen capture
Screen capture is the simplest type of simulation, displaying actual screens used within the application while voice-over audio explains the action. Screen captures are essentially moving pictures that follow a set script. They present highly accurate depictions of an application but with a low level of interactivity.
Screen captures are ideal for simple demonstrations of an application, such as training managers or senior-level employees who need a basic understanding of the product rather than in-depth knowledge of how to use the application.
For example, consider an example from Digital Insight Corporation that trains end-users on an Internet banking application. A screen capture presents the log-in screen. A moving cursor points to the "User Number" and "Password" text boxes and demonstrates how to fill them in. Then the cursor clicks on the "Enter" button. The screen capture doesn’t let learners click on the button, alter text boxes, or control the cursor. The text in the side bar of Figure 2 describes the screen-capture simulation.
Fig. 2: Screen Capture Simulation
© Digital Insight Corporation. Used with permission. Digital Insight and Beyond Internet Banking are trademarks of Digital Insight Corporation.
Level 2: point-and-click
Point-and-click provides the most basic interactive simulation. A point-and-click simulation presents an application screen and lets learners click on hotspots, which are pre-defined interactive regions. Point-and-click simulations target specific areas and avoid actual data input. When learners click on hotspots, the simulation reacts by presenting feedback, moving the simulation forward, or asking learners to click on another hotspot.
Unfortunately, a point-and-click simulation doesn't behave like the actual application. Similar to screen captures, point-and-click simulations are useful for training end-users who don't need an in-depth understanding of the application. They're also practical for illustrating applications that are relatively straightforward.
A point-and-click version of the Digital Insight Internet banking example has learners click on the "User Number" and "Password" text boxes and the "Enter" button. However, learners still are unable to enter a user number and password or click on hotspots out of order. The text in the sidebar of Figure 3 provides instruction for working through a point-and-click simulation of Digital Insight’s training application.
Fig. 3: Point-and-Click Simulation
© Digital Insight Corporation. Used with permission. Digital Insight and Beyond Internet Banking are trademarks of Digital Insight Corporation.
Level 3: data input
Data input simulations provide robust application-style interactivity. In addition to simple clicks, they simulate input elements, such as menus, drop-down areas, check boxes, radio buttons, and text fields. In short, they have the same data input capabilities of the actual application. Although data input simulations don’t mimic all the behaviors of the application, they allow the learner to enter data in a realistic format. But beware: Not all input types are equal.
The most simple type of data input simulation asks learners to click on a button, radio button, or drop-down menu. These are clearly defined actions that result in correct or incorrect answers, making it easy for the training engine to cycle forward to the next step.
Complicated data input simulations ask learners to enter text and click a button that signals the next task. The difficulty with this type of simulation is determining whether learners must enter text exactly as specified. A good rule of thumb is to error-trap questions that may affect the functioning of the actual application. For example, if the application in the Internet banking example requires a dollar sign ($) when entering the amount of transferred funds, then the training application should require it as well.
Another issue is telling learners what to enter into text fields. It's important to clarify for learners any required data, or to not validate input. Consider the login example: The name that learners must enter needs to be explicitly stated.
In general, simulations at the data input level can replicate many application behaviors with a high degree of accuracy. But data input simulations are limited by increased cost and complexity. Basically, the number and complexity of behaviors simulated depends on design resources. However, from an instructional design standpoint, data input simulations can increase knowledge retention.
A data input simulation of the Internet banking application has learners click in the "User Number" text box and enter a valid user number, click in the "Password" box and enter a valid password, and click "Enter" to complete the lesson. The program evaluates each step, and the learner either moves forward or receives an error message. The text in the sidebar of Figure 4 provides step-by-step instruction for working through a data input simulation of Digital Insight’s training application.
Fig. 4 Data Input Simulation
© Digital Insight Corporation. Used with permission. Digital Insight and Beyond Internet Banking are trademarks of Digital Insight Corporation.
Level 4: multiple input paths
Simulations with multiple input paths offer several options to complete an instructional task. This illustrates to learners the notion that software application screens and features are accessible from a variety of paths. For example, there are multiple ways to italicize text in Microsoft Word. One option is to select the text with the mouse and click the "Italics" button on the menu bar. Another option is to highlight the text and simultaneously press the Ctrl key and the I key. And a third option is to highlight text and click on the "Format" button in the menu bar, followed by selecting "Font" and "Italics" buttons in the appropriate drop-down menus.
Multiple path simulations are complicated because they must contain and track several possible interactions. In other words, when only one data input is sought, the program is only concerned with validating a single piece of data. When several data inputs are sought, the program must validate all appropriate data and offer feedback to the learner before moving to the next lesson.
In the Digital Insight log-in example, both text boxes must be validated when learners click the "Enter" button. But suppose the actual application allows the user to enter a password first before the user number. Should the simulated version of the application allow that functionality as well? Perhaps. But the Digital Insight training program doesn't allow for multiple path inputs because it would unnecessarily increase the complexity level.
Level 5: full simulation
A full simulation includes all possible interaction points of the actual software application. The easiest way to create a full simulation is to add training functionality to the application in the form of wizards and so forth, rather than create training with application functions.
The only real distinction between full simulation and multiple paths is that the number of choices is limited in multiple paths. Consider the Word example: The learner has more than one equally correct answer. The difference between a full simulation and a multiple path simulation is what happens when learners make a mistake, such as clicking on "File" rather than "Format." Does the menu appear as a drop-down box, or do they immediately get an error? Under multiple paths, learners are still constrained by the three paths that the instructional designer established as correct. No attempt is made to simulate other aspects of the program. However, in a full simulation, learners should be able to see what is under the "File" menu to learn that it's the incorrect answer.
Indeed, full simulations have few limitations with regard to accuracy and realism, but they require more effort and design skills to build. For instance, in the Microsoft Word example, the designer must specify when, how, and why learners receive feedback. After three incorrect answers? After 30 seconds of inactivity? Likewise, a full simulation of the Digital Insight Internet banking example could explore every screen and feature, but only a few actions would produce successful feedback. Although that can be a good way to review application functions, it can also frustrate and overwhelm learners.
The interaction levels of software simulations run the gamut from screen capture to full simulation. If learners need a broad understanding of the application, screen capture or point-and-click simulations are sufficient. If learners need to become proficient users of the application, data input simulations are more appropriate. Finally, although multiple path and full simulations appear to be excellent training methods on the surface, instructional design challenges often make them difficult to build. Remember: Determining the right simulation depends on the complexity of the application, the learning goal, and requirements for development time and costs.
Published: September 2001