Quick, write a multiple choice question for this standard:
Engage effectively in a range of collaborative discussions (one-on-one, in groups, and teacher-led) with diverse partners on grade 8 topics, texts, and issues, building on others’ ideas and expressing their own clearly. (CCSS.ELA-Literacy.SL.8.1)
Okay okay that really wasn’t fair. It is after all a speaking and listening standard. Try this one:
Analyze data obtained from testing different materials to determine which materials have the properties that are best suited for an intended purpose. (NGSS. 2-PS1-2)
Possible…but difficult. It’s pretty obvious that it makes more sense to assess these standards with something besides a multiple choice question, something even beyond a short answer. In fact, we argue that the range of assessments needs to include something a little different: authentic assessments. There’s a few reasons why:
1) See how students do in the game; not just in drills
Authentic assessments mirror the real world as much as possible. This means using real situations and roles to design tasks for students to demonstrate their learning.
In their foundational book Understanding by Design Wiggins and McTighe explain the difference between an authentic task and how we normally think about assessments by using one of my favorite metaphors. They compare traditional assessment exercises to lay-up drills and authentic performance tasks to a full basketball game. The difference is that “an exercise involves a straightforward execution of a ‘move’ out of context” whereas a performance task requires “thought of the many choices and challenges that confront a performer in practice.”
Look back at that science standard. Writing a multiple choice question to assess it is feasible, but you would probably lose parts of the goal. For instance if you gave students the properties of different materials and asked them to choose which was best for the intended purpose, you would probably lose the analysis of the raw data from testing.
An authentic assessment, on the other hand, might ask students to take on the role of scientists looking for a metal that is malleable enough to use in high speed internet wiring. In this task students would analyze the data from scientific tests to determine which metal (if any) is most appropriate for creating internet service wires. Students would have to make a plan for how to analyze the data and explain their reasoning about their final decision.
Unlike a multiple choice or short answer question, in the performance tasks students would go through the full performance of the standard. Just as important, they might have to try out different approaches to figure out which metal is best, deal with ambiguity, and maybe even work collaboratively with their peers. In an authentic performance task, students would be learning not just the discrete moves, but how to apply those moves in context.
If we want students to be prepared for the challenges of the world, we need to give them opportunities to practice the full game, not just the drills.
Even if you managed to fit all the elements of a complex standard in multiple choice questions, at the end of the assessment you would really only know if students were right or if they were wrong.
This means you would miss out on understanding the most important part of the goal: the student’s thinking. How did the student reason that this material was best for this intended purpose? What was his or her logic?
Missing that means you miss out on the opportunity to see what’s happening in students’ heads and provide feedback to improve it (pretty important right?).
3) Provide opportunities for creative thinking and problem solving
What happens on a multiple choice test if the student analyzes the data and comes to a logically reasoned conclusion that isn’t one of the answer choices? What if a student has a question that adds complexity to the problem (say he or she challenges the validity of the experiment in question). What then?
Authentic performance tasks allow for this creative thinking that we want to encourage.
4) Measure transfer
More about this very important element tomorrow!
Need another reason to include authentic performance tasks in your assessment repertoire? Performance tasks will be a big part of the next generation of assessments that will be coming to a classroom near you (and your classroom) in 2014-2015. Check out a few examples of performance tasks from both PARCC and Smarter Balance (the two consortia designing these assessments).
So how do you design authentic performance tasks?
Wiggins & McTighe suggests designing assessments by thinking about what evidence you would like the task to yield and then creating a goal, role, audience, situation, product, and standards (GRASPs).
Think about how you might design a performance task around the ELA standard we started with. Share your idea below.