Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] management with or without merit



On 09/12/2011 07:18 AM, Edmiston, Mike wrote:

.... I have been through nine different academic
deans in 33 years. (I have also been through three different
presidents.) Five of those deans pretty much loved me and thought I
was a great asset to the university. One of nine was oblivious to
me. The other three of the nine hated me, thought I was bad for the
university, and wanted to get rid of me.

First and foremost, let me say that those three were jerks and
idiots of the first water. Mike E. would be an ornament to the
faculty anywhere.

I don't particularly care for the present promotion, tenure, salary
increment system that I find myself in, because it is mostly based on
years of service as opposed to the value of my service. But I care
much less, and even fear, a system in which some administrator bases
these things on some sort of merit system. The systems in which
administrators exercise some degree of judgment over promotions,
tenure, salary are those very systems that permit cronyism, glass
ceilings (especially sexism, but also racism and personality
conflicts), age discrimination, etc.

I would say there is a perverse kind of consistency in these personnel
decisions, in the sense of "double or nothing". We have (a) the
decision of who to appoint as dean, and then (b) the decisions made
under that dean.
Scheme 1: If (b) lower-level decisions are going to be made without
regard to merit, then (a) in the /short run/ it /mostly/ doesn't
matter whether or not the dean is a jerk and an idiot. You can
appoint the dean almost without regard to his merit or even fitness.
This scheme is self-consistent.
Scheme 2: If (b) lower-level decisions are going to be made on the
basis of merit, with input from the dean, then (a) you'd better be
verrrry careful about who gets appointed as dean. You want this
appointment to be based on merit. That does not mean merely scientific
accomplishment, but also proven managerial skill. This scheme is
self-consistent, too.

Note! You cannot easily switch from scheme 1 to scheme 2! Suppose you
have a bunch of so-called managers (or deans) who were installed without
regard to management ability, and who have never actually managed. Then
suppose you suddenly ask them to start managing. I guarantee that the
result will be a disaster. The institution as a whole will be gravely
injured.

Let's be clear: Either scheme by itself works fine. Neither scheme is
to be feared. (For more on this, see below.) What should be feared,
deeply and rightfully feared, is a sudden switch from scheme 1 to scheme
2.

The [new] law throws out pay scales based on years of service, and
mandates that salary be determined by merit.

This is a recipe for disaster. If you want to switch from scheme 1 to
scheme 2, it requires years of preparation, inspired leadership, lots
of money to buy yourself out of the transient problems that will occur,
et cetera.

======================

As a consultant, I have visited lots of different companies. I have
seen companies that use scheme 1, and companies that use scheme 2.

The companies that use scheme 2 take promotion very seriously. I am
thinking of companies where the front-line engineers are pretty good,
the engineering supervisors are especially good, the district managers
are absolutely brilliant, and the corporate-VP types are world-class
awesome.

Such places have rules, and if you are screwing up they will enforce
the rules on you ... but the rest of the time, if there is something
that needs doing and the rules don't provide for it, you just do it
anyway and everybody is fine with that.

The same schemes apply to universities. Caltech is quite a special
place, and everybody there knows it. There are legends about how it
got to be that way, including lots of legends about the management by
Millikan in the early days, which was ruthlessly meritocratic. He
would figure out who was the top person in the world and do whatever
was necessary to recruit and retain that person. This process was
autocratic as well as meritocratic and there were undoubtedly a few
people who were treated unfairly, but the consensus was that it resulted
in ten or more steps forward for every step backward, and whatever he
was doing they wanted him to keep doing it.

I think judgment is a good thing. The first step is to find somebody
who has good judgment. The second step is to send him off to exercise
that judgment. (You skip the first step at your peril!)

=====================

This gets back to the long-running discussion of grading.

In sports, there is a proverbial question about recruiting: Suppose
there are two candidates whose records of performance are comparable.
Candidate A uses perfect technique, while Candidate B uses a wildly
imperfect technique. Which one should you hire?

The proverbial answer is B, because you can teach him the correct
technique and he will get better.

This illustrates the lunacy of making decisions based on objective
grades. Candidate A has the same grade as to raw performance, and
a better grade as to technique. Yet B is the better hire.

It must be emphasized that grades are called upon to serve many purposes,
and they cannot serve all of them well. The purposes conflict. In
this case, if the purpose is to inform the students how they are
doing, you *do* want to give B a lower grade as to technique, to
inform him that he should try to improve in this area. This conflicts
with the purpose of telling third parties whom they should hire, if
they are so foolish as to hire based on GPA.

This is one example (among many) where judgment comes in.

BTW this is how I explain affirmative action programs. Suppose you
have two candidates with comparable levels of achievement. Candidate
A has enjoyed every advantage, while Candidate B has been discriminated
against all her life. Whom should you hire? I'll put my money on B,
since if I give her a break she'll get better.

There is nothing perverse or contrarian about this. If a Candidate C
shows up with superior raw performance /and/ upside potential, so much
the better.

Over the years, I have spent quite a lot of my own personal money and
millions of dollars of other people's money based on judgment calls of
one kind or another. Sometimes the subjective judgment agrees with the
objective measures and sometimes it doesn't.

I don't know how to explain good management, just as I don't know how to
explain good teaching. In both cases, there is an extensive literature,
which is mostly but not entirely useless. One thing that does help is
to closely observe folks who it well. (You can also sometimes learn from
the mistakes of folks who don't do it well, but this is trickier.)

===================

To repeat: Anybody who thinks they can switch overnight from scheme 1
to scheme 2 obviously has seriously impaired judgment. They have an
agenda that does not include good governance. They should not be
entrusted with running a neighborhood bake sale, let alone serving as
governor.

ON THE OTHER HAND, this does not mean that scheme 1 is better than scheme
2. It's actually worse overall. The bottom line is that you do not want
*either* of these schemes to be set in stone by state law.

Real management is partly about goals but also about means. You need
to know where you want to go, but then you spend 90+ percent of your
time and effort managing the transitions, managing the /process/ of
getting there.