Scratch Jr: Introducing Programming to Preschoolers
March 1, 2012 at 7:26 am 5 comments
How cool! I’m interested in the changes that they’re recommending for Scratch Jr and the rationale that they’re offering.
In focus groups with teachers and children, the Scratch Jr research team has also noticed that younger children struggle with the number of blocks needed to create a program. “The relationship between cause and effect needs to be clearer for this age group,” Bers said. The idea is to reorganize the program so kids can focus on only one thing at a time.
Younger children also have trouble distinguishing between the colors in Scratch, (Scratch Jr will be redone in bright, primary colors), and they struggle with how Scratch moves from top to bottom (Scratch Jr will move from side to side.)
via Introducing Programming to Preschoolers | MindShift.
Entry filed under: Uncategorized. Tags: K-12, Scratch.
Computing Ed Research - Guzdial's Take 
1.
nickfalkner | March 1, 2012 at 7:40 am
Very interesting. Do you think this is related to the number of choices, as expressed by the variety of blocks, as a Hick’s Law effect? I’d also be interested in what it is about the colours that is causing trouble – is this environmental (their exposure to colours is more confronting and higher saturation) and new (changes in modern media), or more based on a simpler eye and it’s always been there?
Lots of food for thought!
2.
Alan Kay | March 1, 2012 at 10:10 am
I’m hoping they incorporate the early “Button Box”-Tortis- work by Radia Perlman when she was a Seymour Papert student in the early 70s.
Also, according to the article it looks as though they are passing up a great opportunity to start helping children learn how to read (they should pair icons and text, etc. and have the blocks be able to say their names, etc.)
Radia’s work was before Mitchel came on the scene, but he is aware of it. Radia had a host of really good ideas about how to deal with 3-5 year olds, and most of them worked quite well.
At PARC, we duplicated her setup and re-ran her experiments. We also made a screen version of -Tortis- to see how the lack of tactile feedback from the buttons affected learning.
The children could make procedures by having the Tortis “remember” a sequence of actions, first in a single “thought-cloud” and then she introduced a box that could handle 4 thought clouds, each in a different color.
Also in Japan in the last decade, an elementary school brought in (MS) tablet computers for all the children and also taught Etoys to the earlier grades.
Despite the UI on the tablets being not good in general, and Etoys had not been tweaked to make this easier, the children could do quite a few things that older children can do, including making animations using the Holder, “object look like other object” etc.
However, cognitively, less was going on that appeared. For example, they could do animation readily, but really didn’t understand it. Even though they learned to program it — they then just used it and could not adapt what they’d learned about this kind of looping to anything else. (This is just what you would expect from this age using the tool in this form.)
My feeling was that they could have done a lot more if they had used the follow-on to -Tortis- that Danny Hillis did (when he was a student there at the same time). It was kind of like lining up physical blocks (in the form of cards with icons and labels on them) and putting them into slots — so it was a physical editor for actions. (There are many cognitive reasons why this is a good idea for very young children ….)
It would be great to have a study done that tries to measure the “percentage?” of “magic” and “superstition” in computer users, and especially programmers of all ages and levels of sophistication. Neil Postman pointed out that people today have to accept much more on faith than they did in the Middle Ages — faith based religion is still around and most of them also haven’t learned science, math, and engineering, governance, etc. — i.e. about how some important things work.
As an example, using physical Tinker-Toys doesn’t often lead to superstitions, but e.g. the pattern matching “StageCast” end-user programming system done at Apple did.
Part of the trick of teaching real cause and effect thinking is to find ways to avoid the use of “magical causes” – this makes engineering of some kinds and math of some kinds good things to teach, but makes both science and computing difficult.
Cheers,
Alan
3.
Alfred Thompson (@alfredtwo) | March 1, 2012 at 2:56 pm
I see some similarity between what they have learned and things that Kodu already does. Kodu instructions are in a list but the parts of an instruction are left to right with the structure of a statment being When {some thing or things happens} Do {some thing or things}. So cause and effect is right out there in the clear.
Also the programming is done with very kid friendly icons with an image and a single word discription of the object. The icons are less subtle than Scratch shapes.
4.
Garth | March 1, 2012 at 3:43 pm
For K – 3 I would really like to have something very basic. I would love to have something like Light-bot with a lot of simple puzzles to solve. Light-bot presently ramps up too fast on the difficulty level and the graphics are too small for young kids.
5.
Computer programming for preschoolers « Gas station without pumps | March 2, 2012 at 12:27 pm
[…] Scratch Jr: Introducing Programming to Preschoolers (computinged.wordpress.com) […]