When I read Robert Cohen's recent posting about precision, and I got to
the part where he wrote, "(gasp!) human error," I had a strong
reaction. I'm not sure if it was a laugh or just a knowing shake of my
head. I guess it was mostly "Ha! I'm not the only one with this
problem."
I thank Robert for bringing this up, because it certainly is germane to
this conversation about precision, accuracy, and resolution. I am now
going to eat some of the words I said earlier about it not really being
important what is covered and what is not.
I sure wish that all the people who had the students before me would
either refrain from using the words "human error," or if they're not
the source, I at least wish they would begin the crusade to eliminate
this phrase from their students' vocabularies.
Even after I go into tirades about not using that phrase, students
still use it. It seems totally ingrained in them; it's like their
conclusion section isn't complete until they've added those magic
words. By the middle of the term I have to tell them that if I see
those words in a lab report it will be an automatic zero.
I've occasionally had someone tell me something like this, "If a human
cannot read the meter stick with sufficient resolution, or make
repetitive readings with sufficient precision, isn't that human error."
To which I respond emphatically, "No, that doesn't mean we need a new
human, it means we need a new measurement tool; perhaps a calipers or
micrometer."
Of course I guess we could say that trying to determine the thickness
of a wire with a meter stick rather than using a micrometer is a "human
error," but why in the world would anyone want to admit that in a lab
report? Should the student say, "Aw shucks, I made the human error of
using the wrong tool to do this experiment."
Actually, I do have students saying things like that in their reports.
But my response is quite easy, "What? You knew you goofed up and you
didn't go back and fix it? Why in the world did you turn in a report
when you knew you did the experiment incorrectly?"
Of course, at Bluffton College our physics labs are open and available
to students all day (as long as a professor is nearby). There is no
excuse not to retake data once the student realizes they have bad data.
In our setting there just isn't any legitimate reason for a statement
like "human error."
I realize some schools have limited lab time and students are unable to
take new data once they realize their original data are poor. I think
that's sad because real scientists don't work that way. But if
time/lab/money constraints are such that the students only get one
crack at data collection, then I would suggest teachers try to get the
students to explain more fully what happened rather than accepting the
cop-out "human error."
For example, wouldn't we feel much better about a student's procedural
error if that student wrote, "Analysis of our poor result has led us to
conclude that we should have used a micrometer to measure the wire
diameter rather than a meter stick. The one-millimeter resolution of
the meter stick allowed us to estimate the wire diameter to about
one-fourth millimeter whereas our goal for this experiment requires
that we know the wire diameter to within a hundredth of a millimeter."
In this regard I strongly agree with Cohen that it is important for the
student to have some ability to analyze what to expect (given the type
of instrumentation used) and whether things basically went right, or
whether they went wrong, and what needs fixed if something went wrong.
Or perhaps more important, if they are not yet sophisticated enough to
do a very good job of this, I would at least hope they realize it is
something scientists must do, and therefore they are willing to learn
it. Because... if they don't have the curiosity/desire to do this type
of thing, then perhaps they should be thinking of career choices where
this type of thing is less important. (Actually, right now I'm trying
to imagine what those other career choices might be.)
Michael D. Edmiston, Ph.D. Phone/voice-mail: 419-358-3270
Professor of Chemistry & Physics FAX: 419-358-3323
Chairman, Science Department E-Mail edmiston@bluffton.edu
Bluffton College
280 West College Avenue
Bluffton, OH 45817