The critical goal of transfer. Arguably transfer is the aim of any education. Given that there is too much for anyone to learn; given that unpredictability is inevitable; given that being flexible and adaptive with one’s repertoire is key to any future success, it stands to reason that we should focus our ‘backward-design’ efforts on the goal of transfer, regardless of what and who we teach (and in spite of pressures to merely ‘cover content’ – which ironically inhibits transfer and worsens test scores, as I discuss below and in the next post). The point of school is not to get good at school but to effectively parlay what we learned in school in other learning and in life.
This notion is now front and center in the latest Understanding by Design (UbD) book, Creating High-Quality Units. The new Template highlights transfer goals since “understanding” surely implies, among other things “effective use of content.” And we have worked hard to help readers and users of UbD understand that the TMA troika is their complex obligation: transfer of learning, meaning-making, and content acquisition. Learning stuff is not the goal, it’s the means.
Furthermore, if you ask people to identify their long-term goals for the year or their career, they almost always identify transfer goals: read widely and deeply, independently; relate current affairs to history and become involved civically; solve all kinds of non-routine problems in and beyond math, etc. Great!
But… few teachers plan, teach, and assesses as if this were the case. Most teachers’ long-term goals are not reflected in the sum total of their assignments and assessments – and that’s why UbD remains needed. The overwhelming reality, in even the best schools, is that your task as a student is by and large to learn stuff and be tested on whether you learned it.
In this post, I want to go back to basics and remind readers of what transfer is and isn’t as a goal. In my next post I want to look at various released test items that plainly reveal that the most challenging test items demand transfer, not recall. And in my third post I will discuss a few key impediments to effectively teaching and assessing for transfer that we must remove, how we might begin to do so, and share some tools and tips for how to achieve better results.
Definition of transfer. Let’s begin with a simple overview of transfer from the first paragraph of the most helpful summary on the subject: Chapter 3 on ‘Learning and Transfer’ from the book How People Learn from the National Academy of Sciences (available for free here). Here is how transfer is defined and justified as a goal:

[Transfer is] the ability to extend what has been learned in one context to new contexts. Educators hope that students will transfer learning from one problem to another within a course, from one year in school to another, between school and home, and from school to workplace. Assumptions about transfer accompany the belief that it is better to broadly “educate” people than simply “train” them to perform particular tasks.

Note, then, a key term in the definition: context. And what this really means is contexts. You have not really learned something well unless you can extend or apply in a new context (framing of the task, audience, purpose, setting, etc.) what you learned in one context. You cannot just give me back what I taught you in a task that is framed just like the teaching tasks and the way I taught it and you practiced it. In the famous phrase in math, it can’t just be a ‘plug and chug’ prompt. There is a further implication in the definition that needs to be explicit: I can only be said to have transferred my learning if i did it autonomously, without much teacher reminders and guidance.
I often use the example of soccer in workshops to illustrate the point. As a coach, I often created drills for helping players learn to ‘create space’ on offense. But soccer is not the sum of the drills: can you now – on your own, in a sport with no scripts – apply those drills in the context of a fluid and novel game situation? Can you now ‘see’ when to use which of the skills we practiced – without my telling you what to do at every turn? That’s my aim as a coach and yours as a player.
John Wooden famously and paradoxically said that his aim as a coach was to be surprised by what his players did in a game. A player who has been so well educated and challenged can innovate, and often must, to win. The same thing is arguably true in all academic subjects.
The definition of transfer as the ability to handle novelty is consistent with what Bloom said about application in the Taxonomy:

Applying of appropriate abstraction without having to be prompted as to which abstraction is correct or without having to be shown how to use it in that situation.”[1]

“If the situations…are to involve application as we are defining it here, then they must either be situations new to the student or situations containing new elements as compared to the situation in which the abstraction was learned….. Ideally we are seeking a problem which will test the extent to which an individual has learned to apply the abstraction in a practical way….  Problems which themselves contain clues as to how they should be solved would not test application.”[2]

Many teachers just expect transfer to happen if content is well taught. No research supports this view. Students who have not been taught for transfer overwhelmingly respond as follows to a ‘novel’ but do-able challenge: We didn’t cover this; I don’t know what to do. In David Perkins’ famous example, it is like the Physics student in college who complained that, while all the problems studied in class involved shooting cannons into the air, the exam question that involved dropping cannon balls down shafts was unfair because “we never studied any hole problems.”
That achieving transfer is far more difficult than we grasp or care to acknowledge is also clear from soccer.  True story about a former player of mine, in a game. When I yelled out to her to apply what we had been learning all week she yelled back in the game: “But the other team isn’t lining up the way we did the drills!!” Indeed. Yet this humorous anecdote has a serious consequence: even well-taught students don’t transfer their learning very well. Many students do poorly on high-stakes tests because they don’t see that an unfamiliar-looking test question is related to something they learned.
In effect, whether in soccer, mathematics or US history, the learners have to be able to see on their own in this ‘new’ task how past learning applies – without the past learning being explicitly prompted.  And, in more challenging transfer tasks, they are thus going to need some creative insight as well as flexibility in adapting prior learning to a very unfamiliar-looking unscaffolded task.
Confronting students with ‘novel’ tasks. Note, then, that the key idea in aiming for and (especially) assessing for transfer is that the student has to successfully confront a “novel” challenge before we should conclude that they really got it. What “novel” means here is: an unfamiliar-looking task (as framed) that none the less should be doable by the student – if they really learned the related content with understanding.
Here’s a simple example: if I teach the 5-paragraph essay, I should be sure to ‘test’ student understanding of the genre by asking them to read and write a 4 or 7 paragraph essay. But as the now-famous item from the MCAS English test in Massachusetts a few years ago revealed, when students were asked to classify a 17-paragraph piece of writing, only 31% correctly said ‘essay’ from the choices – and reported to newspaper reporters that it “couldn’t be an essay because it didn’t have 5 paragraphs.
A vital lesson flows from this issue of novelty. Just because a teacher-designed challenge is hands-on and educationally worthy doesn’t mean that it requires much independent application of prior learning. If the task is familiar and the work is scaffolded, little transfer of learning is required. So, the typical hands-on project – done for all the right reasons – does not assess for transfer if the student 1) gets help all along the way in completing the project, 2) the work is highly contextualized, and 3) little demand is typically made whereby the student must draw general and transferable lessons from the doing of this and other projects. In fact, since such projects are usually so teacher-scaffolded and highly specific they may well inhibit later transfer of the same abilities and ideas in question! I grew flowers, but we didn’t ‘cover’ herbs, so…
Here’s the other irony, to be addressed in my next post: transfer is precisely what a challenging multiple-choice test question demands of the learner. Learners have to handle questions that look different from the ones they studied – with no hints or ability to question the teacher. The most difficult tests questions involve transferable ideas and processes, not obscure facts. Most ‘test prep’ is thus an utter failure because it conflates the format with the rigor: teachers wrongly focus on practicing the test format (using low-level and familiar items) instead of practicing the test goal where the harder questions require transfer of learning.
In the next installment, I want to analyze released test items that make very concrete and clear how educators often misunderstand tests and thus proper preparation for them; and unintentionally undercut transfer, with unfortunate outcomes.

[1]  Bloom et al. (1954), p. 120.
[2]  Bloom et al. (1954), pp. 125-126. Italics added.



16 Responses

  1. I think the challenge with middle school is at what age, grade, or stage is a student able to contextualize the learning. If I teach the swing rhythm or the minor pentatonic scale to grade 7 students, I still have to tell them to use those skills when they switch instruments. When they don’t get it, having their peers jump in and give them the ten second translation helps a great deal. It takes a lot of practice to transfer musical ideas, but to be able for them to self-assess what to do next is the goal for any subject. In music class, the surprise comes when the students say, “we wrote a song”.
    I’ve been criticized as being too vague when teaching (using the Socratic method), but it is this abstraction that allows you to utilize the knowledge in a variety of contexts. Too much too soon and the students are lost. Too little too late and you end up training them, not educating them.

    • You are spot on in terms of the research. Transfer can only happen when there is reflection, analysis, and generalization from the lessons learned. But, obviously, if the intllectual work is abstract and premature it won’t help. Here’s a good example from music: Have kids play chords to a song on the piano in 1st, 2nd and 3rd position. Ask them to generalize which positions in sequence lead to the best flow of music and why. Then have them listen to some simple piano or organ songs in which the positioning clearly makes a difference (Toto, Hold the Line, Gimme Some Lovin) and ask them to describe why it works. Another idea is to transpose songs into lower keys that singers have to sing. This happens in our band all the time – we move to a key that is easy for the musicians to play (say, G) but the key now makes the singer lose the emotional pitch needed for the song.

  2. Hi Grant!
    I have a question for you.
    We are currently teaching 9th grade physics with learning goal quizzes within the model based unit and then giving students a video that illustrates the unit.
    Students are given just enough data to find out the relevant possible data the unit focuses on. (Such as finding an intersection between a constant velocity vehicle and a constant acceleration vehicle).
    Accomplishing all this will earn them a ‘B’. In order to get an ‘A’, they must make a prediction or “what would happen if”, question and answer it with evidence.
    My question is, is this a transfer task?
    I think it is, as it is always a fairly novel situation, and we offer little or no guidance. In fact our directions are very minimal, along the lines of, “tell me everything you can about the forces and motion of the objects”.
    I do not think we can give students a task that requires knowledge that is beyond (or past !) what we have taught, wherein they would need to choose the appropriate model to make sense of the situation. I mean, for heavens sake, we can’t expect students to use acceleration equations on an acceleration test!
    We do have practice videos that go above and beyond what we teach, because a scrimmage should be more difficult than the game!
    Any feedback you can give? Any way you can see how to improve what we do to make it authentic and transfer based? Thanks!

    • Great question! It’s a little hard for me to answer the question with just the info you have provided because so much depends on precisely what the prompt is and what the prior knowledge needs to be. But from the sound of it, it seems like a good transfer task. In order to answer a what if question, you need to go beyond mere plugging in into force/vector formulas.
      That said, a lot depends upon how you assess the answers. The student may propose a perfectly plausible answer to a hypothetical that is not quite what you intended; a student may get the correct answer without fully understanding why it must be right. In other words, it may be that the quality of the reasoning is more salient than the accuracy of the answer – or some combo of the two. (This is a common issue in English where kids may not answer the question accurately but the argument is detailed, evidence-based and plausible.)
      As for authenticity: perhaps the ‘what if’ scenarios can depend upon whether or not all possible salient forces have been identified, as in studying movement in the skies and realizing that perhaps an unseen star/planet is causing anomalous data. Which suggests that a key part of the setup has to be that there are 3-4 completely different plausible models for the data presented and one has to not only argue for one interpretation but argue against the others. That’s real.
      I saw a fabulous 8th grade task in which some of what you envision was done – predict the movement of objects held underwater in which the mass of the objects varied. If you’ll send me an email I’ll send you the task and the other resources the teacher gave me.

  3. […] “If the situations…are to involve application as we are defining it here, then they must either be situations new to the student or situations containing new elements as compared to the situation in which the abstraction was learned….. Ideally we are seeking a problem which will test the extent to which an individual has learned to apply the abstraction in a practical way….  Problems which themselves contain clues as to how they should be solved would not test application.”[2] […]

  4. A right teacher should be able to understand his or her Learners, their abilities and suitable learning approach and techniques to use for best learning outcomes.
    A good teacher makes a considerable value-added difference in learning. He should be able to change what is in his control to enable the Learners achieve the desired outcomes.

Leave a Reply

Your email address will not be published. Required fields are marked *