Richardson Keynote

Thinking about Systems Thinking: How We Improve
Systems Thinking and Dynamic Modeling Conference Skamania, Washington, June 2004
George P. Richardson
Rockefeller College of Public Affairs and Policy, University at Albany
Origins of these Ideas

A long friendship with Barry Richmond, our most creative pedagogical thinker A long history of conversations with him about software, methods, and pedagogy A particular set of conversations we had in May 2002 Leading to thoughts about stages of map and model improvement
Barrys Recent Pedagogical Innovations

(An Introduction to Systems Thinking, isee systems)
A linguistic metaphor to guide model development

Noun-like quantities as stocks Verb-like quantities as ows Adverb-like quantities as modiers of ows
Start with everything being a stock or a ow and improve from there Subtext: Dont use word-and-arrow diagrams for conceptualization
Georges Recent Pedagogical Thinking

Early maps as half-baked modeling thoughts; half-baked is OK. Causal-loop diagrams sometimes essential for getting started A continuum from informal to formal maps Thinking more operationally enables movement along this continuum.
Some History: Maps and Models

Maps: unquantied word-and-arrow diagrams causal-loop diagrams, inuence diagrams, maybe with representations of stocks Models: formal, quantied representations, captured in diagrams and equations, which can be simulated by computer
The Historical Questions

Are maps by themselves are adequate to support insights about complex dynamic systems?
Some say yes; some say no.
Or, are formal mathematical models essential to ground insights in dynamic systems?
Some say yes; some say no.
Insights about building teamwork in a public school

Teamwork and Communication are self-reinforcing
+ Quality of communication + Trust + Risk taking (+)
Quality of communication within teams +
Teamwork + -
Resistance to teamwork
Isolation of teams, and punishing risk-taking inhibit the growth of trust

Quality of + communication between teams + Quality of (-) communication Quality of + communication within teams Trust (+) + + (+/-) Risk taking Teamwork + -
+ Positive responses to experiments ?
Individual + experiments
Resistance to teamwork
But longterm experience with teamwork can build communication

Quality of + communication + between teams + Quality of (-) communication Quality of + (+) communication + within teams Trust (+) + + (+/-) Risk taking Teamwork + + Cumulative experience with teamwork + + Team effectiveness
Resistance to teamwork -
Risk taking can enhance effectiveness, which can build trust

Quality of + communication + between teams + Quality of (-) communication Quality of + (+) communication + within teams Trust (+) + + (+/-) Risk taking Teamwork + + Cumulative experience with teamwork + + Team effectiveness +
Resistance to teamwork +
+ Personal learning
Average personal effectiveness

10
A team-player culture is self-reinforcing: an opportunity or a trap

Quality of + communication + between teams + Quality of (-) communication Quality of + (+) communication + within teams Trust (+) + + (+/-) Risk taking Teamwork + +
Attractiveness of the org to + team players
Cumulative experience with teamwork + + Team + effectiveness +
+ Fraction of staff who are team players
Resistance to teamwork +
+ Personal learning
Average personal effectiveness

11
Likely leverage points

+ Quality of Dialogue training Extent of + communication + Learning + between teams Organization + Quality of characteristics (-) communication present Attractiveness Quality of + (+) of the org to communication + + Trust within teams + team players (+) + + Cumulative Positive + Risk experience with responses to Teamwork + taking teamwork + experiments (+/-) + Fraction of staff who are + ? team players Individual + Resistance to + + experiments Team teamwork + Adm effectiveness teaching Understanding + role stages of + + Personal learning community Average personal building effectiveness +
12
Increasing Complexity
13
14
15
Urban Growth, Stagnation, & Decay

(or is it?!?)
Business land multiplier Business construction Land fraction occupied Business labor force multiplier Business structures Business demolition Jobs Labor to job Attractiveness from ratio jobs multiplier Labor force
Area
Housing land multiplier
Housing construction
Housing
Housing demolition Households to housing ratio
Population Outmigration Inmigration Net births Attractvieness from housing multiplier
Housing availability multiplier
16
Maybe yes, maybe no
17
Two Kinds of Insights

Structural insights
This structure (this loop, these few loops) has the potential to We should be wary of . We should try to exploit . We should try to control .
Behavioral or Dynamic insights

This behavior is produced by this structure.
18
A Feud: A Cautionary Tale about Maps

Hatfields + The more Hatfields, The fewer McCoys. So the fewer Hatfields killed, McCoys killed And thus the more Hatfields, by Hatfields So still fewer McCoys, And still more Hatfields. And so on and on... McCoys But you dont get more Hatfields by shooting them!
Hatfields killed by McCoys +
(R)
19
The Real Structure & Behavior of a Feud

Hatfields
Hatfields killed by McCoys
Hatfields
McCoys
McCoys
McCoys killed by Hatfields
The negative links in the CLD are outows, so the populations always decline!
20
An example from Barrys talk at the 2002 Systems Thinking in Education conference
Barry praised Senges use of this diagram to capture an insight
21
22
Barrys Thinking
CLDs mislead, can mislead catastrophically So always start with stocks and ows In fact, in his latest thinking,
Start with nothing but stocks and ows Add productivity converters to correct the units and the operational thinking for the ows Then revise and add operational clarity
23
Business structures Business construction
Business demolition
Housing Housing demolition
Outmigration
Population
Inmigration
Net births
24
Business structures Business construction
Business demolition
People inmigrating per person per year Housing Housing demolition Population
Outmigration People outmigrating per person per year
Inmigration
Net births People born per person per year
25
Businesses constructed per business per year Business structures Business construction Business demolition
Businesses demolished per business per year
People inmigrating per person per year Housing Housing demolition Population
Housing construction Housing built per house per year
Inmigration
Housing demolished per house per year
26
creating jobs Jobs per business structure Housing construction Housing built per house per year Housing Housing demolition
Jobs
Losing jobs
People inmigrating per person per year
Population
Inmigration
27
creating jobs Jobs per business structure Housing construction Housing built per house per year Housing Housing demolition
Jobs
Losing jobs
People inmigrating per person per year
Population
Inmigration
28
Jobs Jobs per business structure Housing construction Housing built per house per year Housing Housing demolition Outmigration People outmigrating per person per year Population People inmigrating per person per year
Inmigration
29
Barry: What distinguishes/denes Systems Thinking is a unique collection of thinking skills Filtering skills
10,000 Meters Thinking System as Cause Thinking Dynamic Thinking
Representing skills
Operational Thinking Closed-loop Thinking Continuum Thinking Nonlinear Thinking
Simulating skills
Quantitative Thinking Scientic Thinking
30
Barrys Model Improvement Phase

Think about the duration (time constants) of the stocks and replace with algebra (converters) if duration too short for time frame of system Think operationally throughout
Respect the difference between stocks and ows Respect unit consistency Think in terms of how things really work
(theres much more richness here: see Introduction to Systems Thinking)

31
Georges Thinking
Start almost anywhere, but with stocks and ows in causal paths if you can.
But sometimes trying to nd stocks and ows at the start can prevent getting started, so
Be exible, eclectic, nondogmatic. Do what works in the situation. Use intuitive word-and-arrow maps to make headway, if necessary, and see the beginnings of feedback insights. Purge forever the idea of separate kinds of diagrams, but rather employ a continuum of ever clearer, more operational, useful, and insightful diagrams. (Its like writing.)
32
Georges Map Improvement Phase

Rephrase concepts as variables Clarify foggy concepts or ambiguous causality by adding essential word phrases and links Identify units early on, even in CLDs Iteratively remap so all variables, causal paths and loops get clearer and make more and more operational sense General strategy: Work consistently toward a formal map a map that can be turned directly into a formal quantied model
33
We could even start with this level of conceptual confusion and move (gradually) toward a well dened problem and a formal map.
34
What do you really want?

Barry: Everyone an operational thinker George: Everyone a feedback thinker Well, no wonder we disagreed! We have different pedagogical goals. But its not that simple.
35
Goals
Barry wants operational thinking.
But Barry also wants an endogenous point of view (and more -- systems citizens!), so he also wants rich feedback thinking.
George wants feedback thinking.

But George wants extreme conceptual and causal clarity, so he also wants vivid operational thinking.
We both can get both.

36
Map/Model Improvement
Various starting places
Processes for improving maps:
Informal maps Various kinds
Stock-and-flow thinking Dimensional consistency Extreme causal clarity Operational thinking Nonlinear thinking
Formal maps
Formal models
37
An Example from my Childhood

The more I study, the more I learn. The more I learn, the more I know. The more I know, the more I forget. The more I forget, the less I know. So why study?
38
A rst attempt at a causal diagram

The more I study
Looks like a loop!
the more I learn
so the more I know
so the more I forget
and the less I know
Sounds like a stock
So why study?
39
What to do rst?
Picture the stock(s) and ows? Close the loop? Rephrase everything as quantities? ==> Probably that last one rst, but it doesnt really matter!
40
Closing the loop, replacing the question with a concept: motivation

the more I learn so the more I know so the more I forget
The more I study Motivation for studying
and the less I know
41
The more I know and the less I know are really the stock of Knowledge
+
the more I learn
Accumulated knowledge
-
42
Learning and Forgetting are its inows and outows

+
Learning
-
Forgetting The more I study Motivation for studying
43
Units help tell what we really mean

+
Bits learned per week
Learning
Bits of knowledge
Bits forgotten per week
Forgetting
44
The more I study, the more I learn

Learning
Learning
or
Studying
Accumulated studying
+
Studying
45
The whole story

Accumulated knowledge Learning Forgetting Accumulated studying
Studying
Motivation for studying
46
With link polarities

+
Learning
S S S
Forgetting Accumulated studying

+
Studying
S
47
The map with pipes for ows

Learning Accumulated knowledge
Forgetting
Studying
48
Heading toward quantication -becoming clearer and more operational

Productivity of studying (bits learned per hour) Accumulated knowledge
Learning (bits learned per week)
Forgetting
Studying (hours per week)
49
And wheres our original story?

Learning
Forgetting
Studying
50
Or if you prefer
+
Learning
S S
Forgetting
Studying
S
51
An aside: Barrys serious observation on motivation

+
Learning Recent rate of learning

S
Forgetting
Studying
S
52
The Result: A sequence of improving maps

The more I study
+
Learning
-
Forgetting
the more I learn
so the more I know
and the less I know
So why study?
Learning
S S S
Forgetting Accumulated studying

+
Learning
Forgetting
Studying
S
Studying
53
Thoughts about Process

Turn words into quantities Identify accumulations (stocks: draw boxes) Strive for causal clarity; add word phrases (new variables) in causal chains if necessary Identify units for quantities Make the units make sense Capture operational causal structure Revise, rene, expand, clarify, revise, revise, Tell insightful stories with evolving diagrams
54
Thoughts about Process

Move toward a formal map -- a map of such extreme causal, operational, and insightful clarity that it could be quantied and simulated But stop when the process has served its purposes
55
What Im still learning from Barry:

We must get good at processes for map and model improvement, get explicit about them, and get good at teaching them.
Various starting places
Processes for improving maps:
Informal maps Various kinds
Stock-and-flow thinking Dimensional consistency Extreme causal clarity Operational thinking Nonlinear thinking
Formal maps
Formal models
56
What are we striving for?

The ability to work together to solve real world problems The willingness to examine and change ones own assumptions and conclusions Openness to the mental models of others Patience and persistence in problem solving The willingness to be wrong and to learn from mistakes An acceptance that there is often no right answer
57
What are we striving for?

Seeing oneself as an integral part of a larger system with a shared responsibility for the common good Empowerment An extended time horizon An ability to relate the past to the present and the present to the future An internalization of all these principles that informs actions and interactions with others
58

Richardson Keynote

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Richardson Keynote

Uploaded by

Copyright:

Available Formats

Thinking about Systems Thinking: How We Improve

Origins of these Ideas

Barrys Recent Pedagogical Innovations

A linguistic metaphor to guide model development

Georges Recent Pedagogical Thinking

Some History: Maps and Models

The Historical Questions

Insights about building teamwork in a public school

+ Quality of communication + Trust + Risk taking (+)

Quality of communication within teams +

Isolation of teams, and punishing risk-taking inhibit the growth of trust

+ Positive responses to experiments ?

But longterm experience with teamwork can build communication

+ Positive responses to experiments ?

Risk taking can enhance effectiveness, which can build trust

+ Positive responses to experiments ?

Average personal effectiveness

A team-player culture is self-reinforcing: an opportunity or a trap

+ Positive responses to experiments ?

Attractiveness of the org to + team players

Cumulative experience with teamwork + + Team + effectiveness +

+ Fraction of staff who are team players

Average personal effectiveness

Likely leverage points

Urban Growth, Stagnation, & Decay

Housing land multiplier

Housing demolition Households to housing ratio

Population Outmigration Inmigration Net births Attractvieness from housing multiplier

Housing availability multiplier

Maybe yes, maybe no

Two Kinds of Insights

Behavioral or Dynamic insights

A Feud: A Cautionary Tale about Maps

Hatfields killed by McCoys +

The Real Structure & Behavior of a Feud

Hatfields killed by McCoys

McCoys killed by Hatfields

Business structures Business construction

Housing Housing demolition

Business structures Business construction

Outmigration People outmigrating per person per year

Net births People born per person per year

Businesses demolished per business per year

Housing construction Housing built per house per year

Outmigration People outmigrating per person per year

Housing demolished per house per year

Net births People born per person per year

Businesses demolished per business per year

People inmigrating per person per year

Outmigration People outmigrating per person per year

Housing demolished per house per year

Net births People born per person per year

Businesses demolished per business per year

People inmigrating per person per year

Outmigration People outmigrating per person per year

Housing demolished per house per year

Net births People born per person per year

Businesses demolished per business per year

Housing demolished per house per year

Net births People born per person per year

Barrys Model Improvement Phase

(theres much more richness here: see Introduction to Systems Thinking)

Georges Map Improvement Phase