May 2014
Mon Tue Wed Thu Fri Sat Sun
« Apr   Jun »

Day 12/05/2014

Can students learn something about failure from the design world?

Can students learn something about failure from the design world? This is not a novel idea for this blog, but it is something that I am going to begin to explore empirically (with research studies) over the next year.



Lately I’ve been thinking a lot about the design world’s mantra of “fail early, fail often,” or the more recent idea of “fail forward.” In the tech industry, there’s talk of a “failure fetish.” As outlined in their book Fail Fast, Fail Often: How Losing Can Help You Win, Babineaux and Krumboltz explain that the idea behind the mantra is that successes arise out of hundreds of mistakes and failures. The earlier you fail, the earlier you begin to learn.  Therefore, you should push ahead with an idea early, knowing full well that you won’t get it right on the first try, in order to gather feedback. The President of Pixar calls it a process of going from “suck to non-suck,” and the director of Finding Nemo stated that “I won’t get it right the first time, but I will get it wrong really soon, really quickly.”

The benefits of this iteration process are often told through case studies in the design world. For example, in the book Art & Fear: Observations on the Perils (And Rewards) of Art Makingthere is a story about a ceramics class.  In this class, one group of students was graded solely on the quantity of their work (e.g., 50 pounds of pots is an A, 40 pounds is a B), while another solely on the quality of their work (the single best piece will be graded). At the end of the term, the students who made the most technically and artistically sophisticated work all came from the quantity condition. These students experimented and learned from mistakes as they created piece after piece. In contrast, students in the quality condition carefully planned a handful of  “flawless” pots across the course and with limited practice, showed limited improvement.

Photo from


I’ve blogged about the Marshmallow Challenge before, but in case you missed it – in a 2010 TED Talk, Tom Wujec discusses the same principle in terms of his design exercise, the Marshmallow Challenge. In this task, groups of four are given 18 minutes to produce the tallest freestanding tower using pasta, tape, string, and one marshmallow that must be placed at the top. Wujec has facilitated over 70 marshmallow challenges and one of his biggest findings is that teams that prototype iteratively over the 18 minutes have time to learn and adapt their towers and therefore perform better on the task. Teams that carefully build out one idea for 17:50 and then try to place the marshmallow at the top during the last 10 seconds almost always see their towers collapse.


Photo from my most recent Marshmallow Challenge with 7th and 8th grade students at Grand Street Settlement

Photo from my most recent Marshmallow Challenge with 7th and 8th grade students at Grand Street Settlement


In the design world culture, failure is simply a part of the journey towards innovation and is part of every day practice. While this “permission to fail” culture is often view negatively in education (the phrase is used to mean giving certain -often minority- students permission to do poorly), it is something that highly successful design firms such as IDEO pride themselves on and feel directly contributes to their successes.  Why is failure celebrated in these spheres, rather than taking on the same motivation-killing, gut-wrenching connotations that it does in academics?

I’m currently conducting research on a pedagogy known as “Invention with Contrasting Cases,”  (Invention for short) for middle school science learning. Invention is guided exploration where students analyze a set of data and are asked to “invent” an external representation or “index” of an underlying principle that runs through the data. This data takes the form of contrasting cases: examples of scientific concepts or phenomena with predesigned contrasts to highlight key features that will clue young inventors into the abstract ideas. Students are given some time (in our current study we give 15 minutes for each set of cases) to explore the meaning of these cases and invent their own principles, often with scaffolding from teachers or in small groups. Then, students are given a traditional lecture to explain the principles. Multiple studies of this pedagogy have demonstrated that it is an effective form of instruction, compared to traditional tell-and-practice methods, inquiry learning, or reading and summarizing text.

Invention involves what are called “productive failure” tasks, which are characterized by giving students novel problems to solve before giving them direct instruction (as opposed to traditional tell-and-practice pedagogy). Studies have shown that exploring and generating multiple solutions and grappling with the relevant concepts can be productive for learning when direct instruction is subsequently given, and learning gains are found even when problem solving leads to initial failure. In particular, Invention has been shown to foster deep understanding of science concepts and transfer to novel contexts.

However, anecdotally we have noticed that students often have difficulty generating ideas for new inventions after an initial invention fails, requiring high levels of teacher guidance. We have also noticed that sometimes students hit a wall motivationally when they first come across an invention task. I don’t blame them – asking a 13 year old to “invent an index using math” is asking a lot, and it is very different than what they are usually asked to do in a science or a math class.

Photo from


But what if these students could think like designers? or like the start-ups in Silicon valley? If we can get students to dive right into the challenge, expecting and aiming to fail early, learn from that failure, and continue on, I’d expect to see less hesitation and far more progress as they move through these invention tasks.

SO – I am currently developing an intervention where students can learn about design thinking, internalize its failure philosophy, and go through a short design task to experience the utility of failing forward. I’m interested in seeing how it affects their persistence and success on the invention tasks, but I also think this sort of thing could be generally useful for teaching and learning. For example, how would computer programmers benefit from the mantra? Or students writing essays that need multiple revisions? What other academic tasks require this willingness to fail at first and understanding that with iteration, you can improve and succeed?

I’ll keep you posted on my progress!



Babineaux, R., & Krumboltz, J. (2013). Fail Fast, Fail Often: How Losing Can Help You Win.  The Penguin Group, New York, NY.

Bayles, D., & Orland, T. (2001). Art & Fear: Observations on the Perils (and Rewards) of Artmaking. The Image Continuum, Santa Cruz, CA.

Wujec, T. Build a tower, build a team: Tom Wujec on | TED Blog. (2010). Retrieved from



%d bloggers like this: