Research Ed St Louis Conference Notes

These are raw notes taken in real time during the presentations. Some of my own thoughts are blended with those of the presenter. I make an effort to separate them, but I am not always able to. So, if you read something you don't like, don't automatically blame the presenter; it might be me your are upset with.

Research Ed Intro - It is evidence based, collaborative, cultural, practical to classroom the classroom, and for educators. The science of learning becomes an art when you are employing it with 28 8th graders who are running around and not listening to you.

Keynote: Inclusive Teaching: How Do You Teach 30 Students at the same Time? by Tom Sherrington

He has been at 53 ResearchEd conferences, starting with the very first one. 

If you don't understand the problem, you have no chance of being part of the solution. Knowing all of the research doesn't make someone a better teacher unless we can translate it into what it looks like in the classroom. 

We need to be certain that every single student in our classrooms is learning. "We aren't just doing our best with most of them." How do we support vulnerable learners? It's not primarily through taking them out to do special interventions. It starts with the way we teach the things from the very start. If you aren't doing that well, the kids with special challenges will suffer the most. 

The stuff that works isn't new and groundbreaking. We don't need to do new stuff. We need to do the stuff we know really, really well. Explicit instruction improves learning outcomes for everyone, especially for those with additional needs. Show them what to do. Don't ask kids who haven't learned yet to guess. Don't be afraid to give the answer to kids who don't know; don't just say, "Have a think about it." Most kids (but especially vulnerable learners) don't have the background knowledge it takes to make that work.

• Break complex things into simpler steps. 
• Use clear and unambiguous language. 
• Explicitly teach vocabulary. 
• Deliberately involve EVERYBODY in answering questions and checking for understanding. 
• Don't just call on the kid with his hand up or make kids compete to answer - plan in a way that you get an answer from the full spectrum - from the shyest child to the most confident child. The quiet kid's ideas don't matter less. The loudest kids ideas aren't always the best ideas.
• You can't say you believe in equity if you are reinforcing the culture of dominance.

An opinion doesn't matter if you don't have enough background knowledge to explain the thing you have an opinion about. We need to know things because it affects our understanding of the planet (Personal note: This is different from "using this in real life."). Knowledge matters to being informed citizens. 

Dan Willingham is the Godfather of evidence informed education and the most quoted at conferences. In Why Don't Students Like School, there is a chapter called "Why do students forget everything I say?" It's where he provides the model of the learning mind. 

Think Pair Share - If you want kids to talk about their learning, assign them a talk partner. It is the most efficient way; three is possible but requires additional structure. You also must have an accountability structure ("I will select people to respond, and it might be you."). Don't skip the "think" step. You choose which partner will go first. Ask them to generate. All students must practice, not just complete. Whatever method you use, always ask everybody rather than asking for volunteers.

In behavior management, communicate, "I love you dearly, but don't mess with me." Signal. Pause. Insist on compliance before moving on. 

Kids don't have poor attention spans; they have human brains. "Some of you, since we started today, have thought about other things. Outrageous." It is your job as a teacher to redirect their attention.

• Is everyone thinking? 
• Is everyone making sense of this?
• How do I know?

The ways NOT to do it - the veneer of learning

• Did that make sense?
• Any questions?
• Everyone good?
• Yeah? Yeah.
• Getting a response from only one student
• Asking them to think when there is nothing to think with (If they don't have the base knowledge, it doesn't matter how long you give them to think.) You can't think with knowledge you don't have.

You may have to create an experience for students who don't have prior knowledge to activate. If they have never seen coastal erosion, you may need to show a video. If they have never seen a curveball, you might need to have a baseball player throw one. Build from concrete to abstract deliberately.

You don't need them to practice getting it wrong (which is what happens when you ask them to speculate). You want them to practice getting it right after you have given them correct information to practice with. You cannot retrieve what you have not encoded.

Session 1 (103) Making Good Activities Even Better: The Power of Cognitive Alignment by Elida Laski 

Number Comparison: Give two numbers and ask which is more?

Number line estimation: Give them a blank number line and ask them to estimate where the number would be. When you give them a new number they cannot use their others for reference. 

Kids who do well on the comparison task do not necessarily do well on the number line task. Yet, a preschooler's performance on a number line estimation is predictive of their math performance for the future. The number line task has more context and more relationship. A preschooler's relational reasoning maps out to their algebra performance later.

When we plan a task, we need to :

1. Identify the desire mental representation. (e.g. Build a mental number line)
2. Select material that provide instances of the properties of that desired representation. (e.g. 0-10 board game rather than two number card game) Make sure you materials more closely align with the mental representation (e.g. a spiral game board doesn't match as well as a square with lines)
3. Structure the instruction and procedures to promote the processing of the key features of the materials. (e.g. Is the way they play the game forcing them to process the numbers? If they spin 3 while they are on 15 and they just count 1,2,3 to move, it won't develop their number sense as much as if they have to count from where they are 16, 17, 18. It leads to more encoding because it requires more thinking about meaning.)

Learning materials will be most effective if their format, procedures, etc. are aligned with the mental model and the learning goal. The teacher's comments should also align. In the case of comparing numbers Don't say, "You are on 87 now." It's not comparative. Instead, say, "You are two rows ahead of me. You are 5 squares behind her." These are comparative statements. The more aligned the whole experience is to the learning goal, the more likely the goal will be achieved. 

How do you go from concrete to abstract in a task?

• In a pattern, rather than saying colors in a pattern (red, green, green), abstract it in a way that will fit any pattern (A, B, B)
• Teacher gestures support equivalence. It aligns abstract processing to concrete behavior.

None of this requires no resources or monetary investment. It's about thinking deeply and intentionally about how to represent what is happening in their minds while learning. 

Session 2 (100) - Routine, Retrieval, and Relationships by Amber Haven

STEM teacher 5th-8th graders

Intentionally minimize distractions to prepare young minds for learning - Don't make your welcome slide distracting with extraneous information. Make your agenda slide show your agenda. What is the purpose of a welcome slide? To draw attention to the objective or agenda or give an instruction. Making it busy just strains their working memory. There are already so many distractions you cannot control; for the ones you can, don't compete with yourself.

Audit your opening routine. Do you have your desks facing you? Is your decor a distraction or an aid to learning? Do kids know where to sit? Do they have some things that should be automatic by the end of week one?

"When students consistently find themselves in a predictable learning environment, they can let their guard down and engage."

1. Identify the learning objective (the What)
2. Call upon what they have already learned (Retrieval)
3. Plan the lesson (the How)
4. Gesture where appropriate and have students gesture too.
5. Repeat the same activity with a variety of examples.
6. Remove scaffolds as it becomes appropriate
7. Change context as it becomes appropriate

End your lesson in a way that you can confirm students have met the learning objective.

• Exit ticket (WITH verification to prevent encoding whatever they did even if it was wrong)

Relationships are not built before instruction; they are built THROUGH instruction.

Consistent success through well designed instruction will increase your credibility with the student, making them have a relationship with you. Playing hacky sack with them is not relationship building. Doing your job with them leads to success, trust, engagement, feeling good about progress, and then relationship.

Session 3 (107) - Masterful Modeling: Making Thinking Visible by AJ Pettway

Showed a video from Schitt's Creek of trying to follow a recipe with no knowledge of the what the term "fold in the cheese" means.

Albert Bandura - "Learning would be exceedingly laborious and hazardous if they had to rely solely on the effects of their own actions. Fortunately, most human behavior is learned observationally through modeling . . . later , this coded information serves as a guide for action."

Don't be afraid to show students how to do things rather than having them figure it out for themselves (my note: and encoding all of the misconceptions they have along the way).

Mondeling conditions

• Attention - Are student attending to the salient features of the model?
• Retention - Is the model understandable and memorable?
• Reproduction - Students need an opportunity to make the model for themselves.
• Motivation - Are students willing to take the action? Do they see its purpose? Are they seeing that their actions lead to their success?
• Trust - Are you a credible source for your students?

Strong modeling - an expert demonstrating something to novices

• Unpacks expert thinking
• Makes the abstract concrete
• Sets the bar for quality work
• Provides cognitive support that prepares students for independence

Four ways to use models

1. Worked examples - a completed or partially completed example, showing the steps involved
2. Live modeling - reveals expert thinking while demonstrating in front of students
3. Comparing models - Show students multiple model responses and have them compare and evaluate which is better
4. Student models - displaying student work to show how it meets success criteria

Preparing to model

• Clearly define success criteria
• Carefully select CLEAR examples/non-examples
• Anticipate common mistakes
• Script your thinking "The first thing I do is . . . I make sure I have considered . . . I don't want to forget to . . ."

Delivering a model

• Communicate broader context
• Signpost what to notice - direct attention
• Talk through your thinking out loud "I know I'm on the right track when . . . I used to find this part challenging, but I overcame that by . . . "
• Link to success criteria

Faded example - Start with a fully worked problem on the first one. Leave the final step out on the second one (the same type of problem). Leave out a couple of more steps on the third one (same type of problem). Students solve the fourth problem (same type of problem) on their own.

Using student models helps build efficacy because students see themselves represented in their peers. "Someone like me can do this" is a powerful motivator.

One of the challenges of student modeling is building a culture of error - Make sure you are creating conditions where mistakes are honored, considerate behavior is primed and sharing how the student's example succeeds.

Session 4 (213) - Behaviour by Design: Designing Scaffolds to Improve Student Decision-Making by Richard Wheadon 

Sometimes, even when we know better, we just make stupid decisions. How do we get them to make better ones.

Judges make more lenient decisions early in the morning or just after a food break. They are harsher when they are hungry (a study of 1000 parole hearings). Conclusion: External factors influence our behavior and decision making.

There is a gap between knowing and doing. Insights matter, but there needs to be more to produce effective change. Srah Cottinghatt wrote about the Adaptive Expertise Framework. 

Understand. Do. Decide. Improve

Behavior Question - Why did they do that?

Behaviorism -  (Positive reinforcement) - based on Skinner's work with rats - Pressing a lever produced a food pellet reward, reinforcing the behavior of pressing the lever. A consequence is an event hat occurs after a behavior and affects the likelihood that the behavior will happen again. (We associate the word consequence with negative penalty, but it is actually about efficacy.) 

Punishments stop behaviors, but they don't start new behaviors. You need a reinforcer for that.

What should teacher action look like if we want to change behavior? Often, calling out poor behavior reinforces poor behavior by giving it attention. Calling out good behavior reinforces good behavior. (Note: This is not the same as the advice we used to get about ignoring poor behavior; it is about drawing the attention of the misbehaving student to the right thing to do, not ignoring them.)

• Student is talking. Teacher shouts students name - Student stops talking but doesn't engage in task. 
• Student is talking. Teacher praises the engagement of the student near them. Student returns to task.

Evolutionary Biology - Animals that were trained using Skinner's method resulted in some animals not behaving as expected. Conclusion was the reinforcement doesn't always override innate, biological behaviors. This is even more complex with humans because they will have their own values and concerns. In 2003, an experiment was done in which a game was played between two computers and human (the human didn't know they weren't playing against other humans). The computer was programmed to exclude the human after a few minutes. The social exclusion activated the part of the brain that is associated with physical pain. 

Getting praise from other students is more likely to reinforce behavior than anything a teacher does. If a student doesn't turn in homework, gets praised by peers and punished by teacher, they are still more likely to neglect homework again. The peer acceptance is more powerful than the teacher punishment because it is more biologically driven. 

Observing a veteran teacher with great behavior management is likely to be ineffective because there is so much that is invisible making strategies work. 

Engagement - Kids see someone else do something; they are likely to do it too if they believe they can do it. Modeling matters.

Kids saw adults smash a doll. They put the kids in the room with the same doll, and they smashed it too.

If the kids don't think they can do it (efficacy), they won't even start. You must provide the sweet spot of giving them a reasonable chance of success. 

• Mastery - Start with an easy win and build difficulty. 
• Vicarious - Seeing someone similar to you doing it makes yo ubelieve you can do it
• Social persuasion - Everyone else is doing it

Sessions 5 and 6 (103) were my presentations, so I have no notes for the here. However, if you would like the slides, you can find them in pdf form at my website thelearninghawk.com