KnowRe’s data-driven adaptive learning tools for teachers and kids

KnowRe uses bite-sized behavioral data to help kids learn math and help teachers step up their game

Over the past ten years, slews of new companies have promised to revolutionize the way that kids learn and teachers teach. If data collection and analytics technologies can shift paradigms in industries like transportation, manufacturing, and media—the argument goes—why shouldn’t they do the same for education? iPads flowed into classrooms and learning management systems (LMS) multiplied, vied, and died. Many products seek to take offline teaching tools and put them online, to take the hassle out of recording grades and tracking attendance. And some of the most ambitious are seeking to reinvent the way an individual learns—to get inside the heads of students.

Among them, KnowRe has emerged as a tenacious and creative player, using systematic data collection and analysis to drive “adaptive learning” through its math instruction products.

 

Data in math instruction

Teaching math skills to kids is messy. When a child attempts to solve an algebra problem, say 3x2=114, she is in effect performing a number of different small operations (some sequential and some simultaneous). When the student arrives at an answer, whether right or wrong, she may have taken several paths to get there—some of which are more effective than others. There essentially are, for each problem, dozens of ways to get the wrong answer and a few ways to get the right one. The student can mess up the order of operations. She can mess up the division. She can mess up the exponent. Or she can do all three of those steps correctly and then miscopy one digit in the final step. What a bummer.

This is why, when we were kids, our teachers always forced us to “show our work”—so that they could collect data on where we were failing and how well we understood the intermediary steps. The problem, of course, is that it takes tons of time and effort to look through every student’s work, assess areas of strength and weakness, and then prescribe the corresponding practice in a systematic way. And it’s really easy to get it wrong. Enter KnowRe.

 

Value creation and capture

The KnowRe founders realized that every problem solving attempt, for every student, could be a rich source of data to determine real underlying understanding and needs. The algebra problem above can be broken down into its component operations and placed alongside other problems that have the same operations (or, even better, one different operation). This allows for nearly endless systematic comparison and controlled experimentation. Put hundreds of problem attempts together, with systematic variations, and it’s possible to draw statistically significant conclusions about the cause of right or wrong answers for a student or class. If a kid consistently gets exponents wrong, he should get more practice with exponents, not with division.

In KnowRe’s early days, the team spent months on end codifying hundreds of math problems, breaking down the component parts and classifying how the pieces fit together. Armed with a huge library of problems, they built an automated product that presents students with lessons that adapt to build the right skills. They recently added handwriting recognition technology to their products, expanding the data they can collect. The more problems a student completes, and the more of their work they “show,” the more KnowRe (and the teacher) knows. Using the data wisely allows teachers, schools, and districts to improve outcomes, which improves their funding and creates huge value. Teachers can better tell what they are doing right and wrong. KnowRe currently charges the schools to license the tools, which integrate with the existing grading and LMSs the teachers currently use. Eventually they may offer the product direct to parents.

 

Differentiation and future prospects

KnowRe has differentiated itself from competitors by taking “adaptive learning” further than other companies can. While it’s relatively easy to classify the difficulty of a problem, it’s much more challenging to create an operations-based taxonomy like KnowRe has done. They have a huge head start in doing so, and are likely better equipped than anyone else to continue to grow the library.

The long term question remains as to whether or not KnowRe will be able to continue to scale its library of math problems efficiently, and whether it will be able to expand into trickier math disciplines (e.g. calculus). However, in the short and mid-term, there are plenty of schools with young kids who are struggling to learn basic math, and the more data KnowRe collects, the stronger they become. The task at hand, therefore, is to knock its early school launches out of the park, prove the value, and keep the data flowing.

Previous:

DoorDash; Delivery Made Better by Data

Next:

Hudson’s Bay: Big Data for the Perfect Fit

Student comments on KnowRe’s data-driven adaptive learning tools for teachers and kids

  1. Thank you for your post, Won’t Not. KnowRe certainly appears to be doing interesting work. However, I wonder how useful “big data” is for the elementary and middle school math problems KnowRe is addressing. If the goal is to identify areas for improvement, software can determine where a student should focus (i.e. exponents instead of division) within a handful of problems.

    While there are obvious benefits to having a large library of problems, traditional textbook publishers already have extremely large libraries. I wonder how much new value KnowRe can create on a standalone basis – perhaps it’s a simple tool that can be stapled onto a publisher’s existing virtual offerings.

    I wonder how extensible KnowRe’s approach is. Most immediately, as you suggest, it would be interesting to see whether the product can deal with higher-level math. Mistakes in these classes are often tougher to decompose into piece-parts like exponents or division – more often, they relate to a misunderstanding of how the piece-parts interact. It would require a complicated algorithm to identify patterns of mistakes in this context – perhaps the “big data” approach you outline would have more value here.

    Further out, I wonder if KnowRe can do what it does outside of math. Math is the most analytical discipline and it is easier to work on specific skills there than in other disciplines. However, a tool like this might also have value in strengthening, for example, specific grammar skills. The further it is extended beyond analytical subjects, however, the more likely it is to devolve into a commoditized flashcard solution that simply helps with memorization.

  2. Love this, and the idea of splitting up math into smaller pieces. This makes me think of the process of studying for the math component of the GMAT — because the GMAT is an adaptive online test, many of the training programs can break down a precise area that you need to focus on (e.g. you keep making fraction mistakes across various types of questions). I remember thinking that that kind of focus would have been helpful earlier in my math education, and that sounds exactly like what KnowRe is doing.

    I disagree with AJ that expanding to other disciplines would make it commoditized… In fact I think if you could expand into language skill and pinpoint repeated mistakes, be they punctuation, grammar, conjugation, etc., this could be of tremendous value, especially if paired with other online learning tools (i.e. Khan Academy).

Leave a comment