On 04/23/2010 08:13 PM, K. K. Subramaniam wrote:
On Thursday 22 April 2010 07:33:25 pm Caroline Meeks wrote:
1. Software that assess students, track and displays results, quickly and
efficiently without using up a lot of instructional time.
2. Software and a content library that analyzes these results and gives
students the right learning objects/experiences for their current level and
learning style.
#1 is straightforward programming.
#2 is a grand challenge!
Not really. #2 is amenable to statistical methods. See www.assetonline.in, for
instance. Diagnostic tests are different from grading tests in that they do
analyze "wrong" answers too and report to teachers and parents about potential
areas of confusion.
They do have a drawback - they can detect confusion but not prevent them.
Getting it right the first time requires systems like Montessori that put the
learner in charge. All statistical methods come with outliers - 'exceptional'
or 'laggards'. Then you have a problem of dealing with them :-(.
Excellent points (I love this thread, thank you Caroline and all)
One hope is that an early system would take care of the inbetweeners,
who are the greatest number anyway, thus freeing up more time for the
teacher to do her magic on the (circumstantial and extreme) outliers,
besides also more time for each one in class since a lot of the routine
tasks are dealt with by the machine. I mean, if a computer can simply
take care of attendance and retrieving homework, that already gets me
back 5 to 10% of a high school class time, if it can deal with some of
the exercises to understand a concept, I get 50 or even 80% more to
spend in one-on-one follow up.
Also, the hope is that a further improved system can detect the specific
kind of confusion a kid has , to trigger a specific kind of
intervention, by the computer or the teacher, by pointing out the
specific need to the teacher, more efficiently, saving human time and
wear and tear spent in figuring what the current stress point is. I
actually envision such a system would be able to eventually also deal
with a lot of the reinforcing needed by some outliers, as well as the
extra content needed by others, thus a win-win for everyone. Alas,
still tied to the curriculum because that is the way things are. While
ideally this could be done by a human teacher, we know that there simply
are not that amount or level of human resources available able to
discern the needs and then follow up with the appropriate intervention,
and that is just going to get worse as time passes, as the statisticians
tell us.
My biggest selling point is that it would save work for the teacher.
The addition of high-quality content and individualized delivery might
be seen by many as gravy, though of course we know that that is the
heart and raison d'être of the concept in the first place.
Yes, Montessori might be best, but I know too well that requires such a
unique blend of skills and training that it just won't do, especially
for places where the very basis of it is unknown.
And in this concept the learner /is/ in charge. I mean, the very idea
is that the system would mold itself around the learner, following
curricular criteria of course, but shaping the delivery to the learner's
own path. The student would make "his" own choices as he proceeds
forward, with the machine gently giving hopefully the best and most
adequate and understandable material for his own quest to knowledge,
"Guide the child along his own way,..."
Automation can only take us so far. A teacher instructs and listens. The
former service can be done through a computer while the latter is difficult. For
many students with difficult backgrounds, the teacher is the only source of hope
and guidance. We are still a long way from empathic computers ;-).
And maybe it's right that is so. I accompany those who fear a
dehumanised learning system (/cf./ what you see in the opening scenes of
the Star Trek movie relating to Spock schooling), where the remaining
elements of interaction with peers and educators are bullying and
judgmental labels. I believe an Open Source system would never allow
that, enough checks and balances are the nature itself of our approach,
but such evil is not too hard to imagine if these folks
<http://www.p21.org/index.php?option=com_content&task=view&id=508&Itemid=192>
get the contract to be in charge of this initiative, as their for-profit
motive trumps all other considerations, as is the case too often right
now. All in all, I do hope that teachers will be even better able to
listen and support and help build character, eventually trained to be
their best at that, as other tasks are delegated to the machine.
Subbu
_______________________________________________
IAEP -- It's An Education Project (not a laptop project!)
[email protected]
http://lists.sugarlabs.org/listinfo/iaep
