Task analysis is usually considered the most appropriate technique for situations where the tasks are procedural, and the process can be observed. Other techniques may be more appropriate for tasks that are cognitive, and not easy to see or measure.
However, task analysis can fail even for the simple, procedural things if it is not done right. Conversely, it can be surprisingly successful even for the more complex, non observable areas if the analyst is very skilled at winnowing out the necessary substance. So, success or failure of a task analysis depends a lot on how well it is done.
Here are some characteristics of a good task analysis:
•The task statements themselves are clear and measurable. Collectively, they represent the important job outcomes at an appropriate level of detail.
•The skill and knowledge statements represent more than just a further breakdown of the task. They represent underlying concepts, principles, rules and factual information that trainees must possess in order to perform the task with the necessary level of understanding.
•A good task analysis avoids statements that merely represent "mini steps", and then "sub mini steps."
•All skill and knowledge statements are clear enough to be transformed by someone other than the task analyst into instructionally sound objectives without further breakdown. Generally, a good task analysis results in a number of objectives that is equal to or less than the number of task analysis components; never the reverse.
•On the other hand, unnecessary detail is avoided; most (at least 70%) skill or knowledge statements can be transformed directly, one to one, into a worthwhile learning objective.
•The analysis addresses task conditions and standards. In other words, if a condition "using the XYZ meter" applies, a skill statement may be defined to address the conditions: "can take voltage readings with the XYZ meter."
•The analysis reflects direct input from expert job incumbents. It is usually a good idea to utilize procedures and direct observations in the analysis process. But an analysis done without direct input from existing job incumbents is more likely to fail, no matter how well it was done.
•the expert job incumbent assigned to assist with the task analysis should have the respect of line organization involved. The learners and their supervisors know that the quality of their training is dependent on the quality of the technical input it is based upon. If they do not respect the person who represented them during task analysis, they may not even give the training a fair shot at being accepted. (This is not always a relevant issue, such as when the learners won"t even know who the experts were, but in cases where they do, it can be a major factor in their willingness to accept the training as valid).
•There are several approaches to task analysis. In some cases, you can use procedures or technical references to derive the analysis data. In others, direct observation is best. Interviews with job experts at some point is very important.
