[Edu-sig] Learning (some more) programming

[Paul D. Fernhout]

Paul D. Fernhout wrote:
> Arthur wrote:
>>One theme that seems to run through discussions here is related to this 
>>issue. Is it the educators' mission to find just the right motivational 
>>buttons and push them just right ???  Or rather focus on responding 
>>appropriately to those who come to the learning process with some 
>>critical mass level of  motivation???
>>
>>It seems to be one of the fault lines, in some of the discussions here.
> 
> You're right; this is very insightful.

Bad form to reply to myself, I know, but on reflection, I realize there is 
at least one other obvious alternative. One can also try to change the 
system to "lower the bar".

So, here are some options, if you think of learning to program in Python 
like learning to jump over a high bar.

1. Motivate kids to want to try really hard and jump over that bar 
(perhaps using operant conditioning like training dolphins, perhaps just 
making the value of jumping over the bar so compellingly obvious like the 
value of driving in US society, or perhaps by wrapping the subject matter 
in something a kid will find more interesting like a game or contest or 
story or artwork). This is conventional schooling and even conventional 
homeschooling at its best -- and is based on *extrinsic* motivation. (*)

2. Leave a bar around, and wait however long it takes until a kid really 
wants to jump over a bar, and then make any help available to him or her 
they want right then (just in time learning). This is the "unschooling" 
approach, and it is also the approach of "free schools".

3. Compel or coerce the child to jump over the bar by whatever unpleasant 
means is acceptable -- threat of the police or parents, threat of 
withdrawal of privileges, using peer pressure and related threat of 
humiliation, or alluding to some long term failure to find a job and 
succeed in life out of school. This is conventional schooling (or even 
conventional homeschooling) at its worst.

4. Lower the bar. This is the technology developer point of view (the one 
I usually have).

5. And as I reflect more on it, here is a fifth options. You could also 
change the task of jumping the bar into something the kid wants to do from 
*intrinsic* motivation. This merges somewhat into point #1, except it 
builds the feedback into the process directly. So, for example, you could 
put up all sorts of numerical feedback about each jump and plot a kids 
increasing jump height against on a big graph so they can see their 
progress jump to jump (with no praise per jump needed, except maybe for 
effort or progress). (Hard to know how to do this in a programming IDE? 
Perhaps length of programs? Or how often you go between syntax errors?) 
Or, for a high tech approach to jump training, perhaps you put up a smoky 
fog with a laser border shining on it to show a target area to jump though 
and then laser illuminate the area the kid actually jumps through (perhaps 
without the bar being there at first), or some such thing. (For 
programming, this might be some fancy system to give you metrics about 
programs you write: like on complexity or simplicity or elegance -- which 
are hard to think of, especially as aesthetic evaluation of programs is 
perhaps hard to formalize.) So, you are changing the nature of the task 
into one where there is continuous feedback of some kind the kids 
intrinsically wants to excel at even if they don't like the original task 
in its own right.

Great teachers in any setting tend to naturally think about their material 
in this fifth way, I would expect -- assuming they don't otherwise just 
convey enough excitement to get kids hooked on the subject matter in its 
own right out of intrinsic interest in, say, the feel of jumping well. In 
the case of our gardening education work, we could have written a 
"gardening curriculum" that kids had to study and answer multiple choice 
questions about, giving out gold stars (extrinsic motivation) for success. 
Or we could, as we did, make a virtual microworld where they could succeed 
or fail on their own, where motivation is *intrinsic* from seeing plants 
grow and harvesting the results, and which might then motivate them to 
read about gardening from other sources (books, peers, mentors, trial and 
error, the help system, etc.) in order to make their virtual plants grow 
better. Now, I think our garden simulator fails at being intrinsically 
motivating as much as we wanted it to be for various reasons (it's both 
too confusing from having too many features while also missing some other 
key features) but I think the *intrinsic* motivating approach was 
otherwise sound.

As primarily a software developer, I think of options #4 and #5 as the 
ones I explore. :-)

I think the fourth option of "lowering the bar" (as well as sometimes the 
fifth option of "changing the bar") is where I always get into 
disagreements with Kirby here. He maintains the bar is low enough and 
intrinsically interesting enough (that is, Python is adequate to use for 
kids to learn to program well). I maintain it would be nicer if it was 
even lower and more fun (and that Squeak shows some alternatives Python 
could perhaps benefit from). We are probably both right, and then it 
becomes a matter of politics in the sense of how to allocate time and 
money between curriculum and mentoring versus software development. Of 
course, since there is essentially no time or money beyond what we 
individually supply, Kirby mainly works on curricula and gives classes, 
and I mainly write software. :-) Which is probably just for the best.
As Howard Thurman said:
   http://www.gurteen.com/gurteen/gurteen.nsf/id/X000670AA/
"Don't ask what the world needs. Ask what makes you come alive, and go do 
it. Because what the world needs is people who have come alive."

--Paul Fernhout

(*) And if you are not careful, too much extrinsic motivation might create 
a "praise junkie". See:
   http://www.alfiekohn.org/parenting/gj.htm
   http://wik.ed.uiuc.edu/index.php/Praise
   http://www.msnbc.msn.com/id/12648314/
John Holt talks about how, ironically, children in progressive classrooms 
in affluent school districts (where praise is much given for various 
reasons) may actually live in fear of not getting praised in class, as 
much as a previous generation lived in fear of being whipped or spanked in 
class. Remember, the removal of continual praise is itself a form of 
punishment (deprivation). The short lesson for avoiding that is to praise 
in relation to effort or incremental progress ("You tried really hard!" or 
"You're getting smarter every day!"). From the last link: "Recent 
psychological study findings are quite straightforward and to the point — 
kids need praise to guide the development of such characteristics as 
self-control, self-discipline, frustration tolerance and perseverance. ... 
The results suggest that kids who are praised for effort and hard work 
begin to value learning opportunities, whereas children who are praised 
for their abilities value performance. The studies showed that praising a 
child for a personal characteristic such as intelligence (“Aren’t you 
smart. I can count on you for getting an A on your reports!”) can often 
backfire. The researchers noted that kids given praise that evaluates the 
child or their traits and abilities (known as person praise) were 
significantly more likely to display helpless reactions (cognitions, 
affects, and behaviors) when they were later challenged with more 
difficult tasks than children who received effort or strategy praise 
(“Wow, I like the way you looked at this problem from several angles and 
chose an unusual solution”)."
Note that what is important here is not a kid getting the right *answer* 
-- it is a kid getting the right *attitude*.