Skip to main content


Welcome, the Hub connects all projects

Integrating STEM+C


All Topics for this Forum

Topic: "What criteria can we use to define developmentally appropriate goals?"

Topic Posts

Topic started by: Katie Rich on 4/11/16

CS has been traditionally taught at post-secondary levels, so one of the key challenges in bringing it down to elementary and secondary school is defining age-appropriate goals. While empirical testing is key, are there other criteria we can use to help us get started?

This archived topic is open to the public.

restart conversation
This topic has 6 posts, showing all.

Getting started

posted by: Katie Rich on 4/11/2016 10:06 am

I'm working with a few colleagues of several projects that are attempting to sequence learning goals in a way that is developmentally appropriate -- which I'm using to mean that they both are accessible to particular ages (What can a 3rd grader reasonably do?) and build on each other in a way that makes sense (If a student can write a simple if-then conditional, do we move on to if-then-else, or is there something that goes in between?)

Here are some things we've talked about using as a starting point for defining goal sequences:

1. Existing, isolated empirical studies, e.g., this paper says fourth graders were successful with this kind of loop, so we can use this as an anchor point for a sequence of loop goals.

2. Existing trajectories in other subjects, e.g., the CCSS-M says students should be thinking about symbols as representing an unknown as early as first grade, but not thinking about variables as changing quantities until sixth grade, so that gives us a start at a sequence of variable goals.

3. Level of concreteness, e.g. it's easier to conceptualize moving 4 steps than taking 1 step 4 times, so a move command should come earlier than a loop command.

Any other thoughts of other criteria we can use to think about these issues? Has anyone used these or different approaches?

What can a 3rd grader reasonably do?

posted by: Paul Goldenberg on 4/12/2016 11:50 am

I think your aim -- to sequence learning goals in a way that is developmentally appropriate -- is right on. I'm less convinced that the existing resources are faithful to child cognition and not (at least partially) artifacts of how we've already been teaching, and so I'd encourage a bold approach that assumes that, in this new context, kids may well be able to do things that didn't appear possible in old contexts. For example....

... the CCSS-M says students should be thinking about symbols as representing an unknown as early as first grade, but not thinking about variables as changing quantities until sixth grade, so that gives us a start at a sequence of variable goals.

The CCSS-M, alas, doesn't account for how the communication with the child takes place and doesn't start from where children's own linguistic sense-making starts. W.W. Sawyer and Robert Wirtz, ages ago, did understand how to communicate the idea of a "Pattern Indicator" to kids, and even second graders get it well.

Story from a second-grade inclusion class a bunch of years back: Kids had an incomplete horizontally-arranged table of pairs of numbers. The first pair had a 10 on top and a 2 on the bottom. The second pair, top to bottom was 8 and 0; the third pair, 28 and 20. The next six columns had only one number, top or bottom, in each: 18 on top blank on bottom, then 17 on top and blank bottom, then 3 on the bottom with a blank top, and so on. Copying exactly what Wirtz had done in 1964, we also wrote, in the final column, n on top and n - 8 on the bottom. No mention at all was ever made of that last column. Children had never before seen such notation, and had never had been given any explanation (or even mention) of "variables" or "letters standing for numbers." It was simply another entry in the table. Michelle raced up first to say she'd done it all. I asked how she figured out what to do. She said (as characterized all the kids) that she first saw 10 and 2 and just "recognized" that was "minus eight" (yes, she said minus, not take-away). Then she saw the 28 and 20 and then the 8 and 0, so she figured they must all be "take away" (this time using the other language).

Then she giggled and said, as if confiding that we had made some kind of colossal error, "and besides, it says that right here" pointing to the algebraic notation. She hadn't gotten that notation at home. She just did what young children, better language learners than we are, do all the time: she invested a meaning, based on context, in the language that went with that context. The notation looks like "minus 8" and the context was eight, so she decided that must be what that notation meant. She would not have understood that notation without having first made sense of the table; it would not have told her what to do with the table if there had not been the three fully filled-in columns; and that notation would not become general for her on the basis of this one example alone, but in context it was clear, and over the course of the year (this event was the very beginning of the year), most of the kids generated this kind of notation for very simple patterns.

Wirtz also used "pattern indicators" for a word game in fourth grade: after a small number of examples of the game, they gave patterns like MTTM, and children were to find words that matched, like TOOT or PEEP or ANNA. They cleverly also used patterns like ABBC, giving a couple of examples like EMMA, MEET, and TOOT, making clear that the A and C in the pattern could be different letters but didn't have to be.

And kindergarten teachers are teaching this all the time. From a STAR, MOON, MOON, STAR, MOON, MOON, pattern, they have the kids write ABBABB and then come up with some other ABBABB pattern, perhaps using color or notes or.... The point is that A stands for anything, not just STAR. (Alas, unlike algebra, the B seems always to be required to be something different.) The ABBABB, by the way, uses A and B not as symbols for an unknown (see again the CCSS-M) but more like a variable (though with a bit more restriction than algebra gives it), with the entire expression ABB being a pattern indicator.

"Variables as changing quantities" is, indeed, a very adult concept but largely because it is phrased (and conceptualized) that way. But the concept is not alien to kids: a toy block can be a truck or an airplane or table. And, communicated in context (and without a lot of confusing blather with technical jargon like "variable" and "quantity") even the notation is palatable to little kids.

To my mind, this is another place where giving something a special adult name gets in the way of the idea that the name (later) comes to represent.

Level of concreteness, e.g. it's easier to conceptualize moving 4 steps than taking 1 step 4 times, so a move command should come earlier than a loop command....

I wonder, here, too, whether this is a research-verified truth about cognition or an artifact of context and communication. I doubt a child will find it hard to understand what's meant by "I want you to jump (or write your name, or take one small hop forward) four times." But why would anyone call that a "loop"? The term "loop" dates from a time when the coding really looked like that. But REPEAT commands have no "feel" of a loop: conceptually, they specify a thing or a set of things to repeat, whence the (better!) language. MOVE (or some other single act like RING BELL) does still need to come first, but that's not (imho) because the concept of repeating is harder but because there's nothing to repeat until you have something to do. Until a sensible context for REPEAT exists, it is hard to understand.

Bold approach needed to defining developmentally appropriate goals

posted by: Irene Lee on 4/14/2016 10:43 am

I, too, hope researchers and others take a "bold approach that assumes that, in this new context, kids may well be able to do things that didn't appear possible in old contexts."

--Irene

just a place to start

posted by: Katie Rich on 4/18/2016 11:51 am

Hi all,

Thanks for the responses.

I agree with the idea that we should not think of any of these criteria as limitations. I genuinely believe that teaching CS will allow students to do things they've never been expected to do before. We're on the same page there.

On the other hand, though, starting with the idea that kids may be able to do just about anything at any age is very overwhelming, for developers and for kids. My intent was just to throw out some ideas of how we might distribute ideas over K-5. That is, even if we start with the bold assumption that every Kindergartner that work with every aspect of CT, should we try to pack that much into one year? And if not, how can we use what we know to make a good start at distributing content across the grades in a way that's best for kids?

CT

posted by: Trevor Takayama on 4/12/2016 11:38 am

There's a good article in Make magazine about CT vs. programming.
http://makezine.com/2016/04/05/stop-teaching-programming-start-teachin g-computational-thought/

I would be very interested in seeing a curriculum map of skills that specific grade levels can handle.

In my experience, kinder+ can sequence and understand symbols/logos.

Loops are probably for 2bd/3rd grade+ (or simple loops for 1st graders like counting how many times to complete a step move-move-move = 3 moves).

Or could be taught early too, like 2nd grade.


:-) You said it better and briefer

posted by: Paul Goldenberg on 4/12/2016 11:52 am

Yay