[Edu-sig] Observations from the Northwest Science Expo

Wed, 25 Apr 2001 15:13:51 -0700

Wow!

I have been away for a few days, and there are follow-on comments I haven't
read yet as I clean up a 1000-posting Python backlog, winnowed to the 100
that I intend to read.

This is awesome.

First, I learned statistics in college, and yet it wasn't until, many years
later where I took on being a statistician for a 29-week training project in
which people managed their progress and achievement of their results through
statistics that my relationship changed from a fun game with mathematics and
numbers to an experienced measure of something that was happening in the
world.

I affirm that the *mechanical* use of computers is not appropriate in
education or *anywhere* *else*.

I tutored high-school algebra last summer, and I am putting together a team
of 10 to provide more for the same school (the one my entire family
attended) using fellow alumni from the class of '57.  One of the things that
I noticed is that the use of calculators and decimal notation leaves kids
completely at sea when we get to rationals again as part of algebra (where
decimal simplifications are unavailable).  I don't mean to disparage the use
of calculators.  What I see is missing is that intimate relationship with
numbers and arithmetic in which we begin to develop a sense of how these
things work and can be confident in the manipulations and relationships that
the algebra over standard arithmetic then abstracts and makes systematic.

I tutored my 11-year-old grandnephew, a new 6th grader earlier this school
year, and I saw how all he wanted to do was get an answer the teacher would
accept.  There was nothing about how we learned to deal with our fallibility
as arithmeticians, being able to demonstrate our results to others and,
indeed, being able to coach a fellow student.  (I recently noticed that this
is all in Polya's "How to Solve It", but when I examined it as a college
freshman I didn't get it.)

Getting back to the key point.  Having an intimate relationship with the
data and having a way to visualize the data's story born from that
familiarity is crucial.  When I worked for engineers at Boeing as a
19-year-old engineering aide, it was amazing to see their sense of the data
and the automatic way their attention was drawn to things that looked off
and needed to be explored more deeply.  There was usually an error in the
calculations, in some input data, or the rough model we were perfecting.  I
had the same experience working with Paul Horst, a serious statistician at
the University of Washington at that time.

Don't let go on what you observed.  It is critical and much to be cultivated
if we are to encourage the future engineers, managers, and leaders who we
are going to entrust with shepherding the technologies of the future.

The biggest concern I have is the lack of experience teachers may have with
this themselves and how they are being thrown into a world of computer use
without any way to get grounded in where the power really is.  It does not
empower them to teach and it does not give them a sense of the wonder and
power of scientific approaches.  More than anything, there is a missing
compassion for human fallibility and the use of tools that mask that rather
than give us mastery of it.

Thank you for your thoughtfulness and careful observation.

-- Dennis

-----Original Message-----
From: edu-sig-admin@python.org [mailto:edu-sig-admin@python.org]On
Behalf Of Thomas O'Connor
Sent: Thursday, April 19, 2001 15:39
To: edu-sig@python.org
Cc: toconnor@hpvcpto.vcd.hp.com
Subject: [Edu-sig] Observations from the Northwest Science Expo

On March 12 of this year, I participate as a middle school judge for
the Northwest Science Expo held on the campus of Portland State
University, in Portland Oregon.  I was assigned to the team evaluating
middle school behavioral science projects along with 8 other
scientists and engineers.  We evaluated 21 student projects.

This was a great group of kids and I want to be clear that this is in
no way a criticism of any of their work.  I believe that the judges
took much effort to value and encourage the students.  But as is often
the case, a world view can be shattered by an example of something
outstanding, which then shows all else to be mediocre.  I had one of
those experiences at NWSE, and it relates to the use of computers in
education.

I'd be interested in any comments anyone in this group has to offer.

The outstanding example in this case was a project put together by
young lady in the fourth grade (who was, by the way competing against
6-8th graders).  Her project investigated the effects of varying
light/dark cycle periods on bean plant growth.  Her experimental
design and investigation was reasonably thorough, particularly for
someone her age, but not outstanding by comparison to those she was
competing with.  What struck me as outstanding however, was how she
related to her data.

When I asked what conclusions she drew from her experiment, she made
an expansive hand motion, one of those "it should be clearly obvious
to the most casual observer," gestures towards her hand drawn data plots.
With wide eyed excitement, she explained how her data clearly proved
her original thesis incorrect.  It was clear that she had fully
internalized her data.  In this, she stood apart from here peers.

By contrast, most other student I spoke with related to their graphs
and charts as fuzzy abstract representations of their data.  Some
examples:

* More than half of the students produced three-dimensional bar
  charts to display data.  Unfortunately, only one student was able to
  accurately read their bar charts.  The others failed to account for
  the parallax between the bar and the back of the graph which
  represented the X scale.

* One young man was analyzing the effect of color on mood.  But when
  he reported his results, his graph was color coded for color.  In
  other words, his red bar represented the color black, the pink bar
  represented yellow, the yellow bar represented green.  When asked
  about this, he said his software wouldn't let him pick the colors of
  the bars in his bar-chart.

* An 8th grader noted that her results did not vary significantly
  from what would have been predicted from a purely random sampling.
  This was a very enlightened analysis, and the judges all gave her
  credit for attempting to utilize good statistical analysis.  But the
  statistical theory she was attempting to utilize was clearly well
  beyond her knowledge and skill level.  When asked why so much of her
  her project display was focused on the statistical analysis of her
  data when she admitted that she didn't really understand it, she
  stated, "well, Excel did all the math for me."

* A sixth grader did a project to determine which parent the gene
  for red hair was inherited from.  Her knowledge of genetics was
  impressive, particularly for her age.  But the computer generated
  graphs she displayed conveyed no meaning to any of the judges.  Had
  she displayed and focused on the genealogy charts she had stashed in
  the back of her project notebook, she probably would have won an
  award.

* There were two students who had a very good project on luminescent
  solutions and the effects of solution temperature on luminescence.
  While their presentation was very good, what I found most notable in
  speaking with them was the rate of change graph they had scribbled
  on the back of a piece of yellow legal pad to explain the cascade
  effect they had observed.  Their scribbled graph conveyed more
  meaning than all their beautiful computer generated charts.

Questions to ponder:
--------------------

* Does the use of the computer to generate charts enhance the
  synthesis of data, or might it actually hinder it?

  In this case, the fourth grader who was determining environmental
  effects on plant growth collected her data by drawing points on her
  graph each day during the experiment.  At the science fair, she was
  able to articulate the meaning of that data better than other middle
  school students I spoke with.  The only other students who showed
  equivalent understanding did so by utilizing a scribbled graph drawn
  on the back of a legal pad.

  So is there a correlation between the manual plotting of data and
  the cognitive interpretation of that data?

* Gerald E. Jones in "How to LIE with CHARTS," on the "old
  fassion" approach to charting mused:

    Having such crude tools might have forced those early
    chart-makers into slower thought processes.  It is
    conceivable that they actually pondered carefully
    composition--maybe even the content!--of those
    pathetically simple charts and graphs.  Can it be that
    in their technological poverty they achieved a higher
    level of consciousness?  Did they actually come to grasp
    the meaning of their graphic creations? [1]

  With the proliferation of three-dimensional bar that even the
  creators can not accurately read, is the real meaning of the data
  being masked by the visual presentation?

* I know from experience that I can take someone who is generally
  computer phobic but who has a good understanding of statistics, sit
  them down in front of Excel, and within a few hours they can be using it
  productively.  On the other hand, as I observed, an 8th grader who
  knows Excel inside and out, cannot do meaningful statistics.

  So why do so many people insist that schools teach Excel, or Word,
  or Powerpoint for example?  Shouldn't the real focus be on teaching
  number theory, problem solving, language composition and cognitive
  perception?

I'm beginning to wonder how much benefit computers actually add to
the learning process?  I'd welcome a discussion on the topic by those
who are effectively utilizing computers in their curriculum today.

Thanks for the bandwidth.

Tom O.

----- Thomas O'Connor toconnor@vcd.hp.com
      Hewlett Packard, Vancouver Washington
      Phone:  (360) 212-5031
      Telnet: 212-5031

[1] "How to LIE with CHARTS," Gerald E. Jones, Sybex, 1995, p. XVII.