Using adaptive apps instead of teachers

For most parents (but not kids!!!), the most radical and hard to accept part of gt.school is the use of adaptive apps instead of teachers. Parents find our basic premise, that gt.kids learn faster when you get the teachers out of the way, to be troubling and non-intuitive. They have concerns like:

• How can an app teach as well as (insert their favorite childhood teacher)?
• What if my child gets stuck?
• Who “makes” my child focus and learn? Won’t they just goof around if not monitored?
• I’ve tried to get my kid to use apps and they didn’t like them.

These are all great and natural concerns. We built gt.school as an easy way for parents to try it out and prove to themselves that “apps without teachers” works. We have hundreds of students who have used apps without teachers for years as proof points. You can hear from one of these students on her blog, AustinScholar.com. The entire blog is worth reading, but these two stand out as someone using adaptive apps for years:

• ‘What school do you go to?‘
• ‘A High Schooler’s Guide to Reaching the 99th Percentile’

How can adaptive apps teach better than a teacher?

There are a myriad of reasons, but let’s start with the most important one – your Gifted & Talented child. You already know that your GT child is either bored, disengaged or crushing the standard school curriculum. Your child is already learning a lot on their own rather than relying on class time with the teacher to learn. You know they can learn faster. Apps are the way to unlock their learning potential.

Let’s get into the science – the learning science. For over 50 years, there have been countless papers written that highlight better and faster ways to learn (click here for a sample of them). Every graduate school of education has collectively had thousands of graduate students perform studies and write dissertations on how to improve education. This field of learning science is a well understood area. Mastery based learning, adaptive learning, spaced repetition, and testing effect are all well known examples of improving learning.

And if you do dive in and read the papers, you will find the magnitude of the improvement to be HUGE! These aren’t improvements that have 5% better outcomes. They show 2x better, 3x, better, 1 sigma better, 2 sigma better! These are huge improvements in learning. And our adaptive apps implement them. Your child will be able to take advantage of these, not just while at gt.school, but for the rest of their learning life.

One question we always get is “if these are so well known, why don’t standard schools use them?” This is an easy answer – the “teacher in front of a classroom of students” model does not work with these learning techniques. Given that, they can’t use them and are stuck with the old, slow educational model.

Let’s look at some examples.

Mastery-based vs. Time-based Learning

Here are two good videos from Sal Khan (of Khan Academy fame) describing mastery:

The reason a standard school can’t implement mastery is because standard school is time based. It is the end of 5th grade and the students need to move on to 6th grade. They can’t force everyone who didn’t get 100% on the test to remain in 5th grade. So they do what they can, which is give them a grade and push them forward. And as Khan states, it eventually catches up with you. And remember, studies have shown that mastery based education can deliver 1 standard-deviation improvement in performance over standard school.

The Testing Effect

This technique relies on knowledge of how the brain works. If you are reading or listening to a lecture the material is being lightly imprinted on the brain. The best way to deepen the imprint, is to force the brain to “recall” the data. One of the best ways to do that is with a little quiz. A simple multiple choice quiz will deepen the imprint and improve recall later. It is also important that the quiz give immediate feedback to the student on correct/incorrect answers.

It is obvious why this can’t be implemented in a standard “teacher in front of the classroom model.” A teacher can’t be handing out pop quizzes every 15 minutes and then collecting them all, grading them and handing them back to the students. But with adaptive apps, they can quiz the students every 15 minutes and give immediate feedback on right/wrong answers. Various studies have shown this learning technique to be a 2-5x improvement in longer term recall of facts.

Adaptive Software

One of the most famous papers in learning science is Bloom’s 2 sigma paper from 1984. It shows that if you give one on one tutoring to a student, you get 2 sigma better performance than standard teacher in front of a classroom. And for those of you who haven’t kept up with your statistics, 2 sigma implies that students who get one on one tutoring will test at the 95th percentile. It is intuitive that if every student had their own personal tutor with unlimited time, they would score at the top of their class. It is also obvious why this can’t be implemented in standard school. It is impossible for each student to have their own tutor.

Let’s take an example of a standard class of 30 where the teacher is teaching fractions. There are a set of 3-5 kids who already know the material and sit around bored for the 60 minute class. There are 15 students who know division and are ready to learn fractions. And there are 10-12 kids who still don’t know division well enough to be working on fractions. The teacher has to gear the material to the middle 15. What really needs to happen for the 10-12 who are behind is for the teacher to stop the class and go back and teach a lesson on division until the students understand it. Without knowledge of division, sitting in a class trying to learn fractions is an exercise in frustration. But the teacher can’t stop the class. They have to teach to the middle 15, leaving part of the class lost, and part of the class bored and disengaged.

But, technology solves this problem. Adaptive apps act just like personal tutors. The apps can tell if the student knows the material, and if so, accelerates them through it to harder problems. So the 3-5 gt kids who know fractions are going to be given harder problems. And the app is able to stop the fraction lesson and go back to division. The app is also patient – it continues to teach the division lesson until the student masters it and is ready to move on to fractions.

Adaptive Apps with a mastery-based approach implement well-known learning science that improves learning speed tremendously. And it can only be done with apps, not with teachers in front of a classroom. It is not the teacher’s fault; it is a problem created by the system they operate in.

Learning to learn

One of the first life skills we teach in bootcamp is “learn to learn.” A stated goal of many schools is to build a lifelong learner. At gt.school, we start that journey immediately. There is no way to succeed at gt.school without learning how to learn. This means that it is far more than a slogan to us, it is a foundational element that we have mastered and is a key enabler for our school.

The first phase of learn to learn is “what do I do when I am stuck on a problem in an app?” We teach them an easy 5 step process:

1. What do you know? First, ask yourself, “what do I already know about this problem/question?” If it helps you, write it down!
2. Use the in-app help functions. Many students avoid pushing the help button because of the impact on their streak of correct answers. Trust us, it is faster to push the button and learn the correct way to work the problem. That knowledge will help you speed your streaks later. Always use the in-app help.
3. Google! use online search databases for quick information. Google, Wikipedia, or YouTube. Everything you want to learn is on the internet.
4. Make use of the great supplemental apps available to you – WolframAlpha, Socratic, Khan Academy, or Lexia PowerUp.
5. Ask a gt.friend – If you can’t find the information on your own, ask a gt friend. A lot of times, your friends can explain a missing component to you better than anyone else.

One non-intuitive learning we have had as we teach kids to learn to learn: take the teacher out of the equation. We used to have a 6th step – ask a teacher. But we found that it became the default answer and disempowered the kids. Before coming to gt.school, kids are trained to ask for help whenever they get stuck. We have found that if you take the teacher out, the kids assume ownership and agency of their learning. They know they have to solve their own problems and not rely on an authority figure to learn. This is one of the key life skills needed to unlock a gt.kid’s potential. They need to understand that they can learn anything they want and they don’t need a teacher to be there to help/bail them out.

Kids intuitively know this, just look at how they become experts in areas outside of academics. They don’t have an adult telling them what to learn and answering questions. They do it all themselves. gt.school extends that to their academic learning as well.