Evaluating Learning Through Transfer

Evaluating Learning Through Transfer

By Rachel Phillips

Assessment is vital for learning. It’s important to know where students are, how they’ve grown, and what they need to do next to continue to improve. Assessment also helps drive instructional decisions. Assessment is another form of feedback, one that serves both the learner and the teacher. We know it should happen frequently, but how it should happen is often in question.

Research has shown that multiple-choice tests are not a great measure of deep learning and understanding; rather, they often just measure what students remember in that moment. They also measure test-taking skills and logic—whether, for example, a student is able to eliminate obvious answers. Certainly, logic is incredibly important (a test like the LSAT is very focused on logic), but we’re trying to develop young historians, not young lawyers. Research has also shown that other types of assessments—such as projects and essays—give students opportunities to demonstrate more-thorough learning. However, teachers don’t typically have the capacity to consistently analyze student work on a regular basis. Another method for measuring learning, one that is often discussed but infrequently used, is via the idea of transfer.

What is transfer? Transfer is the concept of applying something you know or have learned in one context to another context. The ability to do this shows deep understanding, so instead of stopping what you’re doing to assess what a student memorizes, you can give them a transfer task to assess their level of competence and understanding. Ultimately, doing this also allows for more authentic and fewer test-like activities, and the teacher is then able to focus on mastery rather than procedural or simple content knowledge. For example, in history, a student who has a deep understanding of causation would be able to conduct a causal analysis of most historical events, rather than having only an understanding of the causes or consequences of one event.

Before digging into how you might use transfer to assess students, it’s important to recognize that there are different kinds of transfer, each of which indicates how much a student has learned with understanding. Near transfer, negative transfer, far transfer, flexible transfer...this list goes on, but for our purposes, we’re only going to discuss near and far transfer.

First up: near transfer. Near transfer is taking knowledge and applying it to something very similar. Mathematical word problems are a common example of this. Often the numbers, names, or objects mentioned in a problem change, but the underlying ideas are the same. Typically, students just need procedural knowledge to solve word problems rather than deep understanding. In our curriculum, an example of near transfer might be having a student fill out the Causation Tool based on one historical event, and then filling it out again for another, very similar historical event such as the causes of revolutions.

Now onto what some might think of as the gold standard—far or flexible transfer. Far or flexible transfer is taking an underlying concept and applying it to an entirely different context. One of my favorite examples of this, which comes from a famous 1908 study by Scholckow and Judd, is about throwing darts. Two groups of students were asked to throw darts at an underwater target that was about a foot below the water. One group threw the darts, the other threw the darts and learned about light refraction. Both groups performed similarly at the dart throwing exercise. Then, the target was moved—it was now only four inches underwater. The group that had received instruction about light refraction did much better hitting the new target. Because they had learned about refraction, which relates to how a target’s position can be deceptive, they were able to apply this knowledge and adjust their dart throwing to hit the target more successfully.

Throwing darts at a target underwater. By OER, CC BY 4.0. Throwing darts at a target underwater. By OER, CC BY 4.0.

So, beyond having students throw darts at underwater targets, how can you design transfer tasks for your own classroom? Well, one way is to start by using the tools from the practice progressions found in both BHP and WHP to help students develop conceptual understanding of topics, then have them do historical analysis of an unrelated historical topic without the use of those tools, and then assess how they do on those tasks. Another fun far- and flexible-transfer task is part of the Little Big History Project from BHP. This asks students to take their deep knowledge of thresholds and apply it to something from today that they’re interested in—and those interests don’t, at face value, seem related to the thresholds at all. For example, a student might choose to study sunglasses, or sports, or sandwiches, and then have to decide how that thing relates to the creation of the stars or plants. This is a great way to gauge deep understanding via far transfer.

Admittedly, transfer tasks are difficult to come up with because they require some creativity and, of course, deep conceptual knowledge. But that’s why collective learning (and our teacher community) is so important!  What are some of your ideas for transfer tasks in your classrooms? Post your ideas in the community!

About the author: Rachel Phillips, PhD, is a learning scientist who leads research and evaluation efforts for OER Project, and also develops curriculum for their courses. She is elementary certified, has taught in K–12 schools, and currently serves as an adjunct professor for graduate courses in American University's School of Education. Rachel was formerly Director of Research and Evaluation at Code.org, faculty at the University of Washington, and program director for National Science Foundation-funded research. Her work focuses on the intersections of learning sciences and equity in formal educational spaces.

Cover image: Venn diagram, by OER, CC BY 4.0.

  • I LOVE to use transfer tasks when introducing skills practices. I always start with a non-historical example that’s semi relevant to students (ie uses some kind of pop culture reference)

    So for CCOT I display things that have played music over time - from the phonograph to Alexa devices. Then I use this to get them to identify things that stayed the same and things that changed. What big trends do they see/what do they have in common and what’s different. Then we talk about how what they did was analysis for CCOT and practice with a historical example (like the CCOT practices for overview articles) 

    Or for contextulalization I display an image from a scene from a popular movie or TV show - Edna from The Incredibles when she’s saying “No Capes” or a screenshot of A Stranger Things demogorgen battle. I then ask the students to explain in as much detail as possible everything I would need to know if I had never seen the movie/show to understand what is about to happen in the scene. 

    This year we did a causation tool for each revolution. I modeled the first one, I guided them through the third one, they worked in partner as I nudged them for more details “and then what?”, and the fourth they were to complete independently. Overall this group was more successful in completing well detailed and accurate independent causal maps than previous groups where we did less.  

    Does the skill always transfer and get applied perfectly? No. But it’s a solid way to assess if they actually understand the skill or not before having them apply it with historical information.