A key strategy for measuring learning is to build great assessment tasks …
An Assessment Task (AT) is defined as something that a person who has learned Subject XYZ can do well.
- A person who has learned arithmetic well can (add + subtract + multiply + divide) numbers correctly and quickly.
- A person who has learned scientific research well can define a problem, design a research study that addresses the problem, gather data, draw conclusions, write a journal paper and get the paper published.
- A person who has learned chemistry well can answer questions or explain ideas about the periodic table in accurate, fast, and concise ways.
Below, I’ll list some ideas about what a great assessment task looks like. In the comments section of this blog I’d welcome your ideas about (a) what you like about the list, (b) ideas for making this list better.
Relevant. The Assessment Task (AT) should be super important to the topic being learned. Examples:
- Writing Class: Assess how well a learner can write a paragraph.
- Plumbing Class: Assess how well a learner can join pipes together.
- Life Skills Class. Assess how well a learner can set their own goals and then reach these goals.
- Statistics Class: Assess how well a learner can present and interpret data with a histogram.
Appropriate (for the learner). When I am learning something new, I start as a beginner. As I figure out more and more ideas and how to put the ideas together, this is like walking along a trail that has waypoints (see image). If someone is at waypoint #3, the ATs should be appropriate for this point in the learning journey. If someone is at waypoint #7, the ATs should be appropriate for this level.
Simple (for the learner). The Assessment Task (AT) and the criteria for success should be super clear to the learner; i.e., nearly zero ambiguity on what is being asked and what constitutes doing the task well. In a later post, I’ll explain how to build tasks like this. Note that simple for the learner is not the same thing as easy or trivial.
Measurable. The AT should provide data that allows the (coach/manager/teacher) to gauge how well the person has learned. The scale should span from zero (no learning yet) to infinity (because there is a always a better way to do something …).
Regarding the learner, measurement can sometimes (but not always) be done by the learner himself. For example, I’m a novice swimmer so I lack the qualifications to gauge how well I’m doing my swimming stroke. To get data, I”ll ask a friend who is an expert swimming instructor. In skiing, I am a certified instructor, I have been to racing clinics, etc. and so I am somewhat good at gauging how well I’m doing. Big Idea: Novices are not usually able to gage (i.e. measure) learning, whereas experts are.
Effective (for other parties)
- The AT should take the (coach/manager/teacher) very little time to use. Avoid complexity for managers and coaches. Avoid “grading hell’ for classroom teachers.
- The AT should protect the (coach/manager/teacher) from anger directed at them because the learner did not do well. In later posts, I’ll explain how to remove the (teacher/manager/mentor) out of the line of fire.
- The AT should be easy (low time) for the developer to build and improve.