[From Fred Nickols (2001.08.16.1630 EST)] --
Rick Marken confesses to being unable to get his "little PCT brain around
it" and "it" seems to be that the examples of root causes he's seen looking
less than causes than something else and it is that something else he can't
get his head around. (Wow! I can't believe I just wrote that.)
Here's some unsolicited speculation on my part. One problem Rickus might
be having is that "root causes" (in the TQM context) are part of a linear
cause-and-effect chain. That chain is a linear sequence, an open
loop. PCT is concerned with control in a closed loop. In a closed loop,
the concepts of cause and effect are immediately seen for what they really
are, namely, as names for relative relationships, not absolute
factors. What causes behavior? Is it the reference condition? Is it the
perception? Is it the gap, if any, between perception and reference
condition? I dunno and I don't care; I like the whole loop and I know that
if I don't deal with it, or if I unhook it and string it out in serial
fashion I'm asking for trouble. So, I wouldn't be surprised if a
dyed-in-the-wool PCTer like Rick experienced a little cognitive dissonance
with the notion of root cause. It flies in the face of what we think we
know about human behavior. To search for the root cause in a complex
system is like searching for the stimulus that cause (or elicited) behavior.
Yet, as Rick also confesses, the notion of root cause has a certain
appeal. That, too, makes sense to me. It's kind of like saying, "Gee, if
I could engineer a human being's reference conditions -- and control the
disturbances offered by that person's environment -- I could practically
guarantee his or her behavior." Sure you could, but you can't. Ditto for
large, complex systems known as processes, organizations, plants, sites and
so on. "We" never get ALL that "stuff" COMPLETELY under control. But we
get enough of it under control that we can guarantee (within limits)
certain levels of performance. If we couldn't, we'd go out of business.
Now, speaking for myself as a dyed-in-the-wool human performance
technologist (HPTer), I'd come at the Rx problem from a couple of angles,
both of which would "fit" with PCT and Root Cause and TQM (and just about
any other quasi-religious belief system that I might have to deal with in
the course of doing something about the problem, namely, reducing the Rx
error rate). First, I'd want to capture all the data I could about the
"classes" of Rx errors. In short, what are the patterns in the
errors. Second, I'd want to know about the context in and conditions under
which those errors occurred. I'll bet you a lot of money that distractions
(a.k.a. "disturbances") are at the heart of a lot of them. Where that
leads to is engineering the conditions under which prescriptions are
written (and not to training the writers to stop making those damn
errors). I would also bet you a lot of money that not one of the
prescription writers wants to make those errors or does so intentionally
(insane, murderous physicians and pharmacists are excepted). The relevance
of what I'm saying here, Rickus, is that for you to focus on the conditions
under which errors occur is not "blathering" as you called it but a
legitimate and serious part of performance analysis. TQM folks will
readily agree.
The bottom line (izzat some kind of root?) here is that "root cause" is
both a technical term and part and parcel of the common language. In a
technical sense, a so-called root cause must satisfy three conditions:
First, it can be shown logically to explain the problem, that is, a
cause-and-effect sequence can be traced. Second, it is directly
controllable. Third, the effects of correcting it can be determined.
(That's from a paper I published a few years back when I was ingesting
large amounts of TQM stuff.) In the lay sense, "root cause" refers more
generally to what we talk about when we say, "Let's get to the bottom of
this." In the last analysis, and in both cases, the term refers to the
point at which you halt your investigation and turn your attention to
intervention.
So, Rick, your Rand colleague is still worthy of your respect, your PCT
standards haven't been violated or drawn into question, the sun will
probably come up tomorrow and, "Yes, Virginia, there are Rx errors and you
betcha, we can do something about them!" (And, frankly, Scarlett, I don't
give a damn whose view of the world we use to explain and account for our
success.)
In case any of you are wondering, I haven't "wigged out," I'm just getting
ready to retire.
Regards,
Fred Nickols
The Distance Consulting Company
"Assistance at A Distance"
http://home.att.net/~nickols/distance.htm
nickols@att.net
(609) 490-0095