- --�---jl

; ..-� (
��--��
'�"'·�-.��-���·��--·-��--���·---��-·--�-

� M �1-�_!_:_R �
(·

MANA GEMENT OF INNO VATION

'
l
i, ·;i
, ,, AND NEW TECHNOLOGY
i :. .· RESEARCH CENTRE
-.: . �.

( ' \. '
t ' ,
'

-,

; f"
:
.

\
THE BASADUR SIMPLEX CREATIVE
PROBLEM-SOLVING PROFILE INVENTORY:
DEVELOPMENT, RELIABILITY
· 1 ·:
AND VALIDITY
.• J'.
by

Min Basadur
,:_
•I , !

Management of Innovation and New Technology

Research Centre

WORKING PAPER NO. 83

: J

1 998

� I · MCMASTER
• U N I V E R S I T Y •

I n nis M I C HA E L G . D e G R O O T E

... SCHOOL OF BUSINESS

HD
45
.W657
n o.83
 THE BASADUR SIMPLEX CREATIVE
PROBLEM-SOLVING PROFILE INVENTORY:
DEVELOPMENT, RELIABILITY
AND VALIDITY

by

Min Basadur

Management of Innovation and New Technology

Research Centre

WORKING PAPER NO. 83

1 998

The Working Paper series is intended as a means whereby a researcher may communicate his or her
thoughts and findings to interested readers for their comments. The paper should be considered
preliminary in nature and may require substantial revision. Accordingly, this Working Paper should
not be quoted nor the data referred to without the written consent of the author. Your comments and
suggestions are welcome and should be directed to the author.
 TH E BASADUR SIMPLEX C REATIVE PROBLEM-SOLVI NG
PROFILE I NVENTORY: D EVELOPMENT, RELIABI LITY AND VALI DITY

ABSTRACT

The underlying theory and early development and internal consistency testing of the

Basadur CPSP (Creative Problem Solving Profile) inventory is reviewed, then new expanded

reliability and validity research is summarized. The four col umns, the four q uadrants, and

the two bi-polar scales of the C PSP demonstrated satisfactory reliability via test-retest,

independence of scales, and parallel split-half correlations.

The C PSP correlated with the Kirton Adaptation I nventory (KAI) and the Myers-Briggs

Type I ndicator (MBTI) in predicted ways. Also, the relative impact of creativity training was

moderated according to CPSP style in expected ways. Assessments of CPSP style by self

alone and in consensus with expert partner significantly agreed with the assessments made

by the i nventory itself. Predi ctions of field of endeavor (occupation) a nd of level of

organizational responsibility for strategic thin king by preferred C PSP style and preferred ratio

of ideation-eval uation polar opposites were strongly supported. Opportunities for improvi ng

psychometrics, especially reliability, a re identified.

 2

THE BASADUR SIMPLEX CREATIVE PROBLEM-SOLVI NG

PROFILE I NVENTORY: D EVELOPMENT, RELIABI LITY AN D VALIDITY

I NTRODUCTION

Basadur ( 1 979; 1 982; 1 987; 1 995) modeled organizational, g roup and individual

creativity and innovation as a continuous, ci rcu lar, multi-phase, multi-stage process of

thinking. Basadur, Graen and Green ( 1 982) and Basadur ( 1 994) showed that skills i n this

process, cal led Si mplex, can be i ncreased through training, practice and organizational

n u rturi ng. The process begins with generation and conceptual ization activity, then cycles

through optimization and i mplementation activity. Basad ur, Graen and Wakabayashi

( 1 990) published the Basadur Simplex Creative Problem-Solving Profile I nventory (CPSP) , an

instrument which helps i ndividua ls, teams and organizations discover their own unique

a pproach to the i nnovation process.

Basadur et al provided encou raging preliminary reliability and validity evidence on

the instrument and recommended fu rther research. The pri mary purpose of this paper is to

report research which fol lows up on that recommendation. A g rowing number of

researchers and practitioners of creativity, innovation and organizational development

worldwide are using the CPSP for various additional applications, including increasing

sensitivity to individual differences i n innovation styles; organizing more effective teams to

increase creativity by synerg izi ng individual similarities and differences; providing g roups

with a method to i mprove the q uantity and quality of their interactions; providing a process

model used by individuals and teams to generate and complete important projects;

desig ning special organizational i nfrastructures to encourage more innovative performance;

identifying training and development opportunities for individuals, teams and organizations;
 3

measuring i n novative style i n research; and diagnosi ng organizations and teams for

i nnovation ca pability and i n novative performance problems.

An objective of this paper is to fi rst document the underlying theory and the early

development and i nternal consistency testing of the instrument and then to summarize

ongoing reliability and validity research. This research includes test-retest and random

para l lel split-half correlations and polar, adjacent, and bipolar scale inter-correlations;

external validation by correlation to other established measures relating to creativity such as

the Myers-Briggs Type I ndicator and the Kirton Adaptation-Innovation I nventory; various

customized predictive validity experiments; and standardization studies linking various fields

of endeavor (occupations) and various organizational levels of responsibility for strategic,

non-operational thinking to different innovation styles and to different polar scale ratios.

The Si mplex process approach uses interchangeably the terms i nnovative thin ki ng,

creative problem solving, change making, creative thinking, creativity and innovation. The

Si mplex process is a system of synchronized attitudinal and behavioral thi nking skills which

can be lea rned, developed, shared, managed and continually applied by individuals, teams

and entire o rganizations to produce ongoing innovative results (Basadur, 1 982; 1 987;

1 988a; 1 992; 1 993; 1 994; 1 995; 1 997).

BACKGROUND

A starting point model defining organizational creative and i nnovative behavior as

a continuous, complete three-phase process of problem fi nding, problem solving and

sol ution implementation activity is shown in Figure 1 . This process is identified as a complete
 4
F i g ure 1

A "Complete C reative P ro b l e m Solving P rocess" E mphasizi ng

I d eatio n - Ev a l u ation as a Two - Step P rocess
in E a c h of Th ree C o nt i n u o u s P h ases

Problem
� Finding
Phase I

Problem
� Solving
Phase I I

� Phase Ill

�
 5

process of creative problem solving, which means that it is based on two fundamental

concepts. First, it has distinctly different phases: problem fi nding, problem solving a nd

sol ution implementation. Second, each phase contains a fundamental , sequential, two-step

mini-process of thinking cal led "ideation-eval uation." Ideation is defined as option

generation without evaluation (putting aside the judgment capability) . This is the divergi ng

aspect of the two-step process. Eval uation is the reverse. It is defined as the application of

judgment to the generated options to select the best one(s) . This is the converging aspect

of the two-step process.

This complete process approach considers converging as an important aspect of the

creative process. Cooper ( 1 993) and his colleagues have written extensively a bout the vital

importance of evaluative skills in creating successful new products. This a pproach also

emphasizes that there a re phases to the creative process beyond simply finding solutions to

a l ready identified problems. Skills in discov�ri ng new i mportant problems to solve and in

implementing new sol utions have both been identified as even more vital phases of the

creative process (than creating the solutions) by many researchers (Mackworth, 1 965;

Livingston, 1 9 7 1 ; Getzels, 1 975; Leavitt, 1 975; Simon, 1 960, Levitt, 1 963; Ackoff, 1 979).

In other words, there is more to the creative process than simply "brainstorming"

ideas to solve a presented, defined problem (contrary to much of the earlier research

l iterature on creativity, which was a lmost exclusively confined to brainstorming experiments) .

Osborn ( 1 963), and Parnes, Noller and Biondi ( 1 977) provided linear three- and five-step

creative problem-solving process models, beginning in fact finding and ending in gaining

acceptance.
 6

C ircu lar Flow

Gordon ( 1 956, 1 9 7 1 ) modeled learning and inventing as two opposite forces which

feed upon each other. Learning was characterized as a process of m aking new connections

while invention was characterized as a process of breaking old connections. When we learn,

we "make the strange familiar" (by making new connections between new phenomena and

current understanding) . This permits us to view new phenomena in old ways. In contrast,

when we invent, we "make the familiar strange" (by breaking old con nections that

compromise new understanding). This permits us to view old phenomena in new ways.

Thus, the processes of learning and inventing flow conti nually i nto one another in sequence.

O rganizational field research by Carlsson, Keane and Martin ( 1 97 6) produced a model of

the research and development process as a continuous, circular flow of manufactu ring new

knowledge.

Basadur ( 1 9 74; 1 98 1 ; 1 983) integrated the circular approach into the continuous,

complete process (see Figure 2) and evolved and developed the ful l Simplex process

(described later) beginning with the basic Osborn-Parnes three- and five-step linear models

using his extensive field application experience within Procter & Gamble, Ford and other

companies, and culminating in his doctoral dissertation ( 1 979) . Named Simplex™ in 1 981 ,

the process recognizes that, as new problems are sought and discovered and new solutions

subsequently i mplemented, new problems and opportunities a rise. For example, the

automobi le's i nvention provided not only a new solution to an old problem (improving

.i.------...i.-.i.:�-\t... . ..i. 1....---....l _ _,,, �r"t...I�....,..,. t�,.,JI, ....:,..... ,,. ... ,,.,.,... , ,.,,..,..;rlon+.,\ R,,.,,.,,..i , ,r fl• OOA\
11Ull:>f.JVI1u11u1 IJ uu1 u1 u11u-1 n:;vv tJ vu1v111;, \pv11v11v1 11 011-v-1�T1 "-'I'"""''""""""• llWI'/. ..,""'WI'""''-"'"'• \ , ,,. --r I

described his experiences in learning how to help such organizations as Procter & Gamble
 ..<.+!A!PV
6u!AIOS
we1q0Jd
..<.+!A!PV
uo!+o+uewe1dw1
uo!+n1os
..<.+!A!PV
6U!PLI!:1
wa1q0Jd
L
 8

and Frito-Lay to increase short- and long-term profitability. They did so by mainstreaming

innovative thinking by implementing the Simplex creative process to take advantage of

Mott's (1 9 72) organizational effectiveness resea rch. Mott showed that effective organizations

know how to synchronize two vital but very different characteristics: efficiency and

adaptability (innovation) . Efficiency means optimizing, stabilizing and perfecting current

methods and routines to attain the highest quantity and quality for the lowest possible cost.

High efficiency means mastery of routine (standard, prescribed methods by which the

organizational unit carries out its '!lain tasks) . Adaptability mea ns continually a nd

i ntentionally changing routines and finding new, continuous and better ways to do business.

Called opportunistic surveillance by Simon (1 977), adaptability means scanning the

environment to anticipate new opportunities and problems and changing methods to attai n

new levels of quantity, quality and cost: adaptability yields both new methods and new

products. High adaptability means a high rate of deli berate positive change of routine --

that is, innovation.

Application to Organ izational D evelopment

Basa d u r (1 997) repositioned the field of organ izational development (OD) as an

effort to mai nstream deli berate change in organizations and suggested that OD is a n

ongoing creative process, not a set of intervention techniques. Creativity, or innovation, i n

organizations can be conceptualized as a continuous process of deliberate change, i.e., a n

ongoing cycle of problem finding, problem solving and solution implementation. Problem

fi nding means continuously finding new problems to address, incl uding not just things that
 9

are goi ng wrong but current or future changes, trends, challenges and opportunities. This

incl udes the concept of "looki ng for the golden eggs" that is practised by leading Japanese

companies (Basadur, 1 992). Senge, Klei ner, Roberts, Ross and Smith ( 1 994) a lso emphasize

the i mportance of problem finding i n adaptable organizations. These organizations take the

time to explore background ca uses of i nternal problems to identify bigger, long-term issues

and recog nize the i nterconnectedness of decisions withi n the organization. This recognition

is the essence of systems thinki ng and the sta rting point for making long-term, permanent

i mprovements to the organization. Problem solving means developing new and useful

sol utions to identified problems. Sol ution implementation means making new solutions

succeed. I mplementation usual ly leads the organization to find new problems to solve. New

problems a rise as the system and its environment react to each newly i mplemented sohJtion.

Thus, organizational creativity is synonymous with continuous improvement, deliberate

change-making and adaptabi lity; it is a ci rcular process of conti nuously finding and solving

problems and i mplementi ng new sol utions which represent val uable changes that enable

the organization to succeed (Figure 2) . This definition also removes any distinction between

creativity and innovation (despite views of some researchers who distinguish between

creativity as the generation of an idea and innovation as its implementation) . Here,

creativity, i n novation and organizational development (OD) a re each defined as a multi­

phase, complete a nd conti nuous process. Each phase, i ncluding problem generation ,

solution generation and sol ution i mplementation, is a creative phase.

Specific OD i nterventions a re discrete tools used within this creative process. Some

of these tools a re used within the problem-finding phase, others within the problem-solving
 10

and implementation phases. For example, diagnostic interventions such as su rvey feedback

or the confrontation meeting a re tools for problem generation; used regularly, they help

antici pate and expose organizational issues that might otherwise have remained buried.

Group development (or team-building) process interventions help solve identified problems

such as slow project completion by improving the way people work together. When the

o rganization has identified employee development as an important problem to solve, it

might use structural interventions such as job enrichment. Structural interventions can a lso

hel p implement other i ntervention sol utions. For example, a new reward system like the

Sca nlon plan, i ntended to encourage employees to use their creativity by sharing the profits

from their productivity i mprovements, can ensure that employees permanently transfer

creativity skills learned in tra i ning to their jobs. Elsewhere, Japan's Toshiba Corp. places

newly h ired scientists and engineers directly into the sales department to reinforce the

i mportance of problem finding (learning customers' problems) before they begin developing

new products to solve those problems within research and development. Finally, individual

interventions such as counseling a nd coaching help in the i mplementation phase of the

organizational creativity process. For junior managers who have been trained in facilitative

leadership (i ndividual i ntervention) to work i n a flattened organizational design (structural

intervention), counseling and coaching by senior managers (individual intervention) can help

cement the training.

From this perspective, it is organizational creativity that provides the change-maki ng

dynamic for organizational adaptabil ity and effectiveness. This suggests that an organization

can achieve adapta bility if it mainstreams the creative process. For this to happen, two
 11

thi ngs a re necessary. Fi rst, employees must obtain new thinking ski lls (and reframe their

jobs, i .e. , become creative problem finders and solvers and solution implementers) . Second,

the organization must provide a framework for directing these creative thinking skills to

support its i mportant goals and objectives.

Training in a multi-phase, ci rcular, "complete" process (Si mplex™) has been shown

to enhance i n novative attitudes and thinking ski l ls i n a variety of organizations a nd

functional specialties (Basadur, Graen & Green, 1 982; Basadur, Graen & Scandura, 1 986;

Basadur, l 988a,b; Runco & Basadur, 1 993). Such training attempts to increase

appreciation of the value of a l l such multiple phases, and of separating and synchronizing

divergent and convergent thinking. If training is sufficiently i ntensive and lasts long enough,

participants develop not only this a ppreciation but the corresponding skills (Basadur, 1 979;

Basadur et al, 1 982). However, experience shows that most people feel less comfortable

within some phases of the "complete" process than in other phases. They also find that their

own preferences for various phases differ from those of others: styles of using a complete

process of creative problem solving differ from one person to another.

The Eight-Step Simplex C reative Process

Basadur's experience in implementing the circular process (Figure 2) suggested

breaking the three phases i nto eig ht steps as follows: 1 ) problem finding (anticipating future

problems and seeking current problems); 2) fact finding; 3) problem defining; 4) generati ng

potential sol utions; 5) eval uati ng potential sol utions; 6) action planning; 7) gaining

acceptance; and 8) taking action. Because the enti re process is circular and conti nuous, the
 12

eighth step is fol lowed by the fi rst step to begin a new cycle. Each action taken to implement

a new sol ution automatical ly leads to new problems, changes and opportunities for a new

round of creativity. Figure 3 displays the eight steps within the complete Si mplex creative

p rocess, begi nning with problem finding and prog ressing toward action in a never-ending

flow. One of the two key ideas upon which the Creative Problem Solving Profile (CPSP) rests

is this circular model of the "complete process." The other key idea underpinning the C PSP

is "dynamic tension," which fol lows.

Dynamic Tension

Parnes et al ( 1 977) suggested that creativity is a function of knowledge, imagination

and judgment. They provided a simple equation as a model: C = K x I x E. This equation

suggests that, to perform creatively (C) , one must begin with appropriate knowledge (K) . As

in a kaleidoscope, one's i magination (I) must transform what is known into new, different

combinations, cal led new pattern ideas, options or points of view. Final ly, evaluation (E) is

needed : one must exercise good j udgment to select the most appropriate patterns, ideas,

options or points of view for further development or implementation.

Let us first focus on the knowledge component of this kaleidoscope analogy. How

people gain knowledge (learn) is personal and individualistic. That is, the bits and pieces

of knowledge one absorbs and retai ns can be gained in contrasting ways. As discussed

above, one way is by direct, concrete experience (getting personally involved in the task and

"getting one's hands dirt-;"). An opposite \Atay is by detached, abstract thin king (standing

back, observing, analyzing and theorizing to understand). Some people have used the terms
 13

Figure 3

Organizational C reativity as a Conti nuous, Circu lar Process

of Eight Steps Across Three Phases

problem PHASE I
PHASE Ill finding
SOLUTION "fuzzy" PROBLEM FINDING
situation
IMPLEMENTATION

idea
finding

PHASE II
PROBLEM SOLVING
 14

"tacit" versus "explicit" knowledge to explain these two notions. For example, some

knowledge can be discovered only by experiencing first-hand new and unexpected things.

For example, Mi ntzberg (1 989) related how Japan's Honda Corp. stumbled across the

opportunity for its smal l scooter in North America just by being there. The company had

sent four marketi ng managers to Los Angeles to establish the market for Honda's new giant

"macho" motorcycles. To reduce taxi costs, the fou r managers rode errands a round the city

on Honda's small scooters. (Honda headquarters had assumed there would be little market

for such a small vehicle on the open roads of North America.) The scooters attracted

attention and finally turned i nto a giant market success for Honda, but only when the bigger

machines developed mechanical problems and the marketing managers pursued their newly

discovered opportunity. Thei r real mission, in retrospect, had been to learn whether they

could sell something - anything- i n North America. Rather than attempt to cover every base

early, the company resisted the l ure of over-rationality and came to America prepared to

lea rn by doing, rather than remai n in Tokyo using second-hand information.

Second, let us focus on the i magination and judgment components of the analogy.

How people use knowledge is q uite personal and individual istic. Two opposite ways of using

one's knowledge a re for ideation (to generate more options, ideas and diverse points of view

while deferring judgment) and for evaluation (using judgment to select from among those

options, ideas and diverse points of view) .

Each individual could thus be characterized as having a unique set of relative

preferences on these two i nformation processing dimensions (experiencing-thinking for

g aining knowledge and ideation-eval uation for using knowledge), as in Figure 4. This set
 15
F i g ure 4

Two D i men si o ns C ompri s i ng C reative Pro b l e m S olvi ng Process Activity

Gaining Knowledge
(Learning) by
Direct Concrete Experience

Using Knowledge Using Knowledge

for Evaluation for Ideation

Gaining Knowledge
(Learning) by
Detached Abstract Thi n king
 16

would describe the individual's preferred way of learning and usi ng knowledge. Each

person 's u nique set of preferences might then help identify h is or her unique style of creative

problem solvi ng.

Creative problem solving or innovation is considered a process involving both ways

of obtaining knowledge and both ways of using knowledge. An i ndividual's relative

preferences for these four concepts can portray his or her own blend of preferences. The

resulti ng blend is cal led the Creative Problem Solving Profi le, which provides a basis for

understanding preferences in using the Simplex process (see Figures 5 and 6).

RATIONALE

Fou r combinations of the opposing methods of gaining knowledge (experiencing vs.

thi n ki ng) and the opposi ng methods of using knowledge (ideation vs. evaluation) yield four

dominant styles of creative problem solving. The fou r styles correspond to the different

phases of the Simplex creative problem-solvi ng process as follows.

The Simplex problem-finding phase consists of two different aspects (Basadur,

El lspermann & Evans, 1 994). The first is the active sensing, seeki ng or anticipati ng of

problems. Referred to as "opportunistic surveillance" by Simon (1960), this aspect can be

termed problem generation. The second is the formulation of these problems, termed

problem formulation or problem conceptual ization. Al bert Einstein reputedly said that

merely formulati ng a problem is often far more essential than its solution. He said that,

g iven one hour to save the \Vorld, he would spend 55 minutes defining the problem and only

5 minutes solving it. Based on Ei nstein's statement (and John Oewey's quote: "A problem
 17
Figure 5

Examples of D ifferent Profiles of

Creative Problem Solving

(All Four Examples Below Have The Generator Style Dominant)

Generator style dominant

with all three other quadrants
small.

Generator style dominant

with Implementer style as
strong secondary.

Generator style dominant

with Conceptualizer style as
strong secondary.

Generator style dominant

with Conceptualizer and
Implementer as secondary
styles of significant and
equal strength.
 18
Figure 6

Examples of Different Profiles of

Creative Problem Solving

(All Four Profiles Below Have Different Styles Dominant)

Generator style dominant

with all three other styles
relatively small.

Conceptualizer style dominant

with all three other styles
relatively small.

Optimizer style dominant

with all three other styles
relatively small.

Implementer style dominant

with all three other styles
relatively small.
 19

wel l stated is half solved") , the proportions of Figures 2 and 3 bear modification. Thus, i n

Fig u re 7, the problem-fi nding phase represents half of the circular process; splitting that

phase to reflect its twi n aspects results i n a four-quadrant model, as in Figu re 8.

Figure 7

Problem Findi ng As One Half of the Circul a r Process

I mplementation
Problem
Finding
Problem

Figure 8

The Four-Quadra nt Model

Problem
I mplementation Generation

P roblem Problem
Solving Formulation
 20

The four quadrant styles were labeled Generator, Conceptualizer (Formulator) ,

Opti mizer, and Implementer to complete the Creative Problem Solving Profile (CPSP) . How

the CPSP may be overlaid on a four-stage Simplex wheel is shown in Figure 9.

Figure 9

C PS P Quadrant Styles Labeled

E ight Step Circular Model of a

Complete Process of Creative
Creative Problem Problem Solving with Ideation­
Solving Profile Evaluation Sequenced in Each
Step Grouped in Four Stages

8 l
Action Problem
'< > < >7
Finding
Quadrant I 7
Gain "-.. /Fad
(Generator)
< > < >
Acceptance"-.. / Finding

< >
Plan "-.. Problem
/
'< > 3
,Definition
/
/Evol. Idea "-..

< >
Select Finding

D EVELOPMENT OF THE I NSTRUMENT AN D SCALE CONSTRUCTION

Development of the I nstrument

The Creative Problem-Solving Profile inventory was developed as fol lows. First, 20

graduate business students in organizational change and development used divergent

thinking to generate a list of descriptors for each of the fou r concepts representing opposing

ways of gaining and using knowledge. Second, the group reached consensus on the 12

words from each of the four l ists that most appropriately described the fou r concepts. One

important criterion for selecti ng a word was its abi lity to be coupled with a word from the

opposing list. Final ly, 24 additional words were incl uded in the fou r lists (six words each)
 21

as distractors. These words were deliberately generated and chosen as either total ly

unrelated or opposite to the 12 initial words on that l ist. The result: fou r col umns, each

containing 18 words, i ncl uding six distractors i ntended to help i nventory users remai n

u n biased. (Users of such self-report instruments often follow a "best-answer'' pattern. )

Selectio n of Scale for Ran ki ng Items

Use of a forced-choice scale ensured that the inventory reflected relative preferences

for different phases of the complete creative problem-solving process (the CPSP is i ntended

to reflect a person's relative preferences within this complete process) . On this 4-3-2-1

forced-choice sca le, 4 denoted the most-preferred item and 1 the least-preferred within each

g roup of fou r items. The resultant inventory is shown in Figure 10. Col umns 1, 2, 3 and

4 represent the experiencing (X) , ideation (I), thinking (T) and evaluation (E) concepts

respectively. Sets of four words corresponding to items3, 4, 6, 7, 8, 9, 11, 12, 13, 15, 16

and 18 were compiled by pai ring words judged as the most a ppropriate opposites between

col umns 1 and 3, and between 2 and 4, respectively. The distractor sets correspond to items

1, 2, 5, 10, 14 and 17. These 18 item groups were presented i n a way that ensured

participants could complete the i nventory in an unbiased manner.

Selecting a forced-choice scale (also called an ipsative scale) requi res users to state

some preference level for each of the four quadrants of the complete process of creative

probl em solvi ng. This is consistent with the theory that, although each person has some

u ni que combi nation (profile) of relative preferences among the different phases, all phases

of the process are valuable and should be appreciated .

 22
Fig u re 1 0

Creative Problem Solving Profile I nventory

This i nventory is designed to describe your method of problem solving. Give a high rank to those words which best
characterize the way you problem solve and a low rank to the words which are least characteristic of your problem
solving style.

You may find it hard to choose the words that best describe your problem solving style because there are no right
or wrong answers. Different characteristics described in the inventory are equally good. The aim of the i nventory
is to describehow you solve problems, notto evaluateyour problem solving ability.

Instructions:

E ighteen sets of four words are listed horizontally below. In each horizontal set assign a 4 to the word which best
characterizes your problem solving style, a 3 to the word which next best characterizes your problem solving style,
a 2 to the next most characteristic word, and a 1 to the word which is least characteristic of you as a problem solver.
Be sure to assign a different number to each of the four words in each horizontal set. Do not make ties.

Column1 Column2 Column3 Column4

1. Alert Poised Ready __ Eager

2. Patient Diligent --
Forceful __ Prepared

3. Doing Intuitive Detached Selective

4. Experiencing __ Optimistic __ Objective __ Verifying

5. Reserved Serious Fun-loving __ Playful

6. Sensing __ Free Thinking __ Logical __ Experimenting

7. Feeli ng --
Alternatives __ Analyzing __ Evaluating

8. Action Divergence --
Abstract __ Convergence

9. Direct Possibilities Conceptual Practicalities

10. Quiet Trustworthy __ Irresponsible __ Imaginative

11. Involved Pro I iferati ng --

Theoretical __ Testing

12. Probing __ Projecting __ Structuring __ Examining

13. Immediate Gathering __ Understanding __ Confirming

14. Impersonal --
Proud __ Hopeful Fearful

15. Implementing __ Visualizing __ Modelling Decisive

16. --
Present-oriented --
Future-oriented --
Rational --
Detail-oriented

17. Sympathetic __ Pragmatic --

Emotional __ Procrastinating

18. Aware Childlike Orderly Realistic

 23

Using a forced-choice scale also permits respondents to weight their relative

p reference for gaining knowledge and using knowledge. Thus, not only does the profile

reflect relative preferences for the two opposing ways of gaining knowledge and the two

opposing ways of usi ng knowledge, but it more genera l ly reflects relative preference for

gaining or using knowledge.

P re limi n ary Screen i ng

The inventory was tested for face validity with attendees of a training conference

(n=30). The respondents u nderstood the inventory and its purpose, and reported that the

instrument was highly accurate in assessi ng their creative problem-solving process styles.

Quantitative reliability and validity testi ng required an appropriate selection of

psychometric tools because of the selection of a forced-choice (ipsative) scale to rank items

within each of the 12 item groups. For example, to make a prelimi nary test for i nternal

consistency of the fou r col umns, a special, non-forced choice (non-ipsative) Likert scale

format was created. Data were gathered using this non-ipsative version of the i nventory

using the same 18 item groups i ncluding the six distractor groups. Respondents were asked

to assig n a value (1 to 5) for each of the 72 items i n isolation (i ncl uding the 24 distractor

items). The 1-5 scale is shown i n Figu re 11.

Figure 1 1

Scal e Used to Rate Each Item i n Specia l Non-Forced Choice Format

�--------------t--------------t--------------t-------------4
2 3 4 5

Very little A little bit Somewhat Quite Very Much

Characteristic Characteristic Characteristic Characteristic Characteristic
of my Problem of my Problem of my Problem of my Problem of my Problem
Solving Style Solving Style Solving Style Solving Style Solving Style
 24

This non-forced choice format (non-ipsative) permits respondents to rate each item

independently withi n each of the 12 item groups. Because each item is independent of all

others in its g roup, a standard Cronbach Al pha can be appropriately calculated for each

col umn. By contrast, the forced-choice scale by defi nition yields non-independent ratings

for each item. It is expected that, despite individual differences, most people wil l give each

col umn a significant score as most people would regard favora bly all fou r quadrants and

both ends of both dimensions of using/gaining knowledge. Thus, the format selected must

encou rage respondents to choose from within each item group yet vary their pattern of

choices from one item g roup to the next. These two requirements are met with a well­

constructed set of horizontal item groups that offers variety among items in each vital column

p l us a forced-choice rating scale. Cronbach Alpha is apparently not an appropriate

measure of column reliability using the ipsative, forced-choice format, as it reflects the

simi larity of val ues a l lotted to a l l items in a column . The CPSP forced-choice format

encourages people to do the opposite. For example, one would expect that persons who

favour ideation and eval uation eq ually as a way of using knowledge would rank ha lf (6) of

the 1 2 item pai rs higher i n Column 2 than Column 4 and half (6) higher in Column 4 than

Col umn 2. The specific g roup of six likely would vary randomly from one person to the next

according to their u nique experience and interpretation of the words. This random way of

deployi ng equal weights (scores) to each column for each person in such a sub-group would

yield excessively low alphas for each col umn, but it would accurately assess relative

preference of each person.

 25

Thus, as discussed further i n the section on Reliabi lity and Validity, for the forced-

choice i psative sca le, the random paral lel split-half reliability method employi ng the

Spearman-Brown corrected correlation co-efficient was used to assess reliability for each of

the columns and the bi- polar scales rather than C ronbach Alpha. C ronbach Al pha would

be either i na ppropriate or excessively low for measuring the "complete process of creative

problem solving." At best, Cronbach Alpha would be an unnecessarily stringent

measurement of i nterna l consistency; at worst, it would be a tota lly inappropriate measure.

In using C ronbach Alpha and i nter-item correlation to screen the items selected for

a ppropriateness, the non-ipsative version of the CPSP inventory was used. First, on a base

n = 149 , (the n = 13 7 sample with 12 additional subjects) , a frequency distribution of the

responses to each item (including the distractors) was run for each of the 72 items. The

response distribution for each item was checked to ensure that it met the required desirable

proportion of 80%-20%. That is, for any item, the total percentage of respondents who

a nswered 1, 2 or 3 should fal l between 20% and 80%, as should the total percentage that

answered 3, 4 or 5. I nspection showed that, althoug h some of the distractors fai led to

survive, all of the regular items did.

Because very few people selected 1 on any item, all 1 and 2 responses were

combi ned i nto a single response. A computer program was written to recode responses 1

and 2 as 1, response 3 as 2, response 4 as 3 , and response 5 as 4. This recoded 1-4 scale

for each item was used throughout the remaining analyses.

The same work and results were conducted on a second sample (n= 107) . On each

of these two samples, using the recoded responses (1-4), a Cronbach alpha was calculated
 26

on each of the four columns of the non-i psative form of the inventory. All fou r col umns had

reasonably high alphas in both samples, ranging from . 76 to .83, as shown in Table 1. The

correlation matrix showing inter-item correlations and the Reliability Analysis for each

column a re displayed in the Appendix. Appendices 1 throug h VI I I refer to the n= 149

sample, and appendices IX through XVI refer to the n = 107 sample.

Table 1

I nterna l Consistency of Columns Usi ng N on-lpsative Format

C ronbach Alpha Col umn 1 Column 2 Col umn 3 Column 4

Sample 1 (n = l 49) . 78 .82 . 76 .83
Sample 2 (n=107) .76 .80 .80 .77

Reviewi ng Table 1 , the alphas for col umn 1 were .78 and . 76. All 12 items

correlated with the rest of the column l items taken as a group at least .20 or g reater in

both samples. Scanning the matrices, about seven inter-item correlations were

a pproximately zero in sample l, and 11 in sample 2. The remainder were above . l 0 a nd

positive, except sample 2, which had two items at -.12.

The al phas for column 2 were .82 and .80. In both samples, all 12 items correlated

with the remai ning items in column 2 taken as a group a bove .20. Two individual inter-item

correlations were nearly zero i n sample 1 and three in sample 2. In both samples, the

remainder were above .10 and positive.

For column 3 , the Cronbach alphas were . 76 and .80. All of the items correlated

above .20 with all the other items taken as a g roup with one minor exception in sample 2

(.19). About 10 of the inter-item correlations were close to zero in sample l and about eight
 27

in sample 2. All the rest were above . 1 0 and positive except one item in each sample (-. 1 5

and - . 1 4 respectively) .

For col umn 4, the Cronbach alphas were .83 and 77 . . All of the items in both

samples correlated a bove .20 with the remainder of the items i n the col umn taken as a

g roup except one, which correlated . 1 9 and .09 in samples 1 and 2 respectively. About six

of the inter-item correlations were close to zero in sample 1 and nine in sample 2. All the

rest were above . 1 0 and positive except one in sample 2 (-. 1 2) .

Bi-pol a r Scales

Stil l using the non-ipsative scale, the i nternal consistency properties of the bi-pola r X-T

and 1 -E scales were i nvestigated. After subtracti ng each of the items of col umn 3 (T) from

its counterpart from column 1 (X), a single scale or factor, X-T, was developed. Likewise, a

second single scale or factor, 1- E, was developed from ·Columns 2 and 4. This provided

items in the two new scales with a range of possible val ues from -3 to +3. To create ordinal

scales, the quantity 4 was added to each new item in the two new scales, providing a val ue

between 1 and 7 for each new item.

The Pearson correlation coefficient between the two bi-pola r scal es was computed as

. 1 3 for sample 1 and .07 for sample 2. This indicated the independence of the two scales.

An i nternal rel iability analysis on the bi-polar X-T scale revealed Cronbach Al phas of

. 62 for sample 1 and . 73 for sample 2. With two relatively minor exceptions, all X-T item

pai r correlations with the remaining scale item pai rs as a group were quite strong. One of

the exceptions was an item pair that was weak in sample 1 (. 08) but was much stronger i n
 28

sample 2 (.30). This was item pair 1 5 (implementi ng-modeling). The other exception was

an item pai r that was slightly weak in sample 1 (. 1 6) but stronger in sample 2 (.22) . This was

item pair 1 8 (aware-orderly) . The rest of the item pairs were all above .20 in both samples.

The second scale, 1-E, had a Cronbach alpha of . 74 in sample 1 and . 7 6 in sample

2. Al l of the item pairs correlated above 20 with the rest of the item pairs in the scale taken

as a g roup in both samples. These data a re available in Appendices XVl l through XX, a nd

the Alphas a re summarized i n Table 2.

This internal consistency demonstrated on the non-ipsative inventory supported the

previous successful results experienced for face validity described above using the regular,

i psative, forced-choice i nventory format. The further reliability testing of the ipsative version

is reviewed in the Statistics section.

Table 2

I nternal Consistency of the Bi-pola r Scales Using N on-lpsative Format

C ronbach Alpha l nterscale

Correlation
X-T Scale 1-E Scale X-T� 1-E
Sample 1 (n= l 49) .62 . 74 .13

Sample 2 (n = 1 07) . 73 . 76 .07

SCORI NG, PLOTIING AN D I NTERPRETATION

Discovering One's Own C reative Problem-Solving Profile

As stated above, creative probiem soiving can be characterized as a continuous

circular process i nvolving two opposite ways of gaining knowledge and two opposite ways

of using knowledge. These two dimensions can be portrayed as two perpendicular axes with
 29

fou r q uadrants as shown in Figu re 2. I ndividuals can use the Creative Problem Solvi ng

Profile I nventory discussed below to plot thei r own relative preferences on each of four

scales: Experience, Ideation, Thinking and Eval uation. The shape of the profile a nd the

dominant quadra nt that result from connecting the fou r scores provide i nformation about

one's own� or profile of creative problem solvi ng. See Figures 5 and 6.

Quadra nt D escriptions

The Quadra nt I orientation toward creative problem solving is cal led generator. I n

a generator mode, one's dominant creative problem-solvi ng inclinations a re: (1 ) learning

by d irect concrete experience (sensi ng the environment; a bsorbing knowledge; experiencing

and gatheri ng i nformation personal ly) and (2) using knowledge for ideation (imagining

possibilities and sensing releva nce in a lmost everything; seeing many different points of view;

dreaming a bout what might be; wondering why things seem to be what they a re; speculati ng

about the future) . Combining these two i nclinations i ndicates a preference for problem­

sensing and fact-fi nding activities in the creative process. When operating as a generator,

one is a n initiator, or a proliferator of opportunities, problems, facts and feelings. Sensitive

to the environment, generators a bsorb diverse i nformation and possibi lities that might

pertain to their i nterests and goals. Generators a re comfortable with high ambiguity and

dealing with much i nformation and potential opportunity. In a generator mode, people love

to start things, are comfortable in the early phases of creative problem solving, and a ct as

probl em starters and challenge finders. They antici pate and sense new problems, changes

and opportun ities, and enjoy fact fi nding.

 30

The Quadrant II orientation toward creative problem solvi ng is cal led conceptualizer.

When one is operati ng in a conceptualizer mode, one's dominant creative problem-solvi ng

inclinations are: ( 1 ) using knowledge for ideation (as above) and (2) gaining knowledge by

detached abstract thinking (tryi ng to understand or explain a situation cognitively; being

detached and objective; making sense of things in the abstract) .

The conceptualizer's combination of these two i nclinations indicates a preference for

problem formulation and defin ition and idea generation. Conceptualizers tend to absorb

a wide range of seemingly unrelated facts or idea fragments and possibilities, and assimilate

them into an i nteg rated explanation, hypothesis, theory, q uestion, chal lenge, problem

definition or idea. They l i ke viewing the big picture and extracting and defining the essence

of the opportunity or problem, and generating ideas to solve it. They are problem definers

and idea developers and a re comfortable in the early to middle phases of creative problem

solving.

The Quadrant Ill orientation toward creative problem solving is called optimizer.

When one is operating in an optimizer mode, one's creative problem-solving inclinations

a re: (1) learning by detached dominant abstract thi nking (as above) and (2) using such

knowledge primari ly for evaluation (developing criteria for assessing alternatives, being

aware of possible pitfalls and deficiencies in potential sol utions, and looking for a single

optimum or best answer) . These two inclinations indicate a preference for being i nvolved

i n the practical sol utions of a well-defi ned problem, and planning and organizing concrete

staps for making them implementable. Thus, optimizers a re problem solvers, and a re more

comfortable in the middle to later phases of creative problem solving.

 31

The Quadrant IV orientation toward creative problem solving is called implementer.

When one is operating in an implementer mode, one's combi nation of inclinations toward

usi ng knowledge for evaluation and gaining knowledge by direct concrete experience

i ndicates a preference for implementation activity - gaining acceptance from others for new

sol utions and changes, and taking action to make sure those solutions and changes work

and stick. I n an implementer mode, one does not generally dwell on understanding the

theory behind a new idea or plan. I nstead, one wants to take it and "run with it," experience

it, work with it, show others how to use it, shape it and fit it to needs, adapt it to various

changing circumstances, and try and retry it (and not worry about why it didn't work the first

way) . I n a strong implementer mode, people wil l do whatever it takes (including alteration)

to implement the plan, idea, new product or new solution. They wil l become directly

i nvolved and experiment u ntil implementation is complete. Implementers a re problem

finishers and a re most comfortable in the later phases of creative problem solving.

Scori ng a nd Plott i ng

The inventory i n Figu re 10 is scored and plotted as follows: I n each col umn, all the

items are summed except items 1, 2, 5 1 10 1 14 and 17 (the distractors) to yield col umn

scores. Each column score is plotted on the appropriate axis of the Basadur Creative

Problem Solving Profile shown i n Figu re 10 below. Note that the col umn 1 score indicates

the individual's orientation toward gaining knowledge for creative problem solving by di rect

concrete Experiencing; the column 2 score indicates the orientation toward using Ideation

i n creative problem solving; the col umn 3 score indicates the orientation toward gaining

knowledge for creative problem solving by detached, abstract Thinking; the col umn 4 score
 32

indicates the orientation towa rd using Evaluation in creative problem solving. Connecti ng

the four points in sequence with curved lines makes an i rregular circle that represents one's

personal creative problem-solving profile. Four identical col umn scores would result in a

perfect ci rcle. This is unlikely to occur (but is a perfectly legitimate profile). The q uadra nt

in which the profile is most dominant indicates one's strongest orientation. The other

q uadrants represent secondary orientations. One's profile is one's unique blend of the four

q uadrants a nd, therefore, one's unique s1yle of creative problem solving.

In summary, the Simplex creative problem solvi ng profile (CPSP) is intended to

i ncrease understanding of the fol lowing points:

1. Creative problem solving may be considered as a "dynamic tension" between many

pai rs of seemingly opposing forces: freedom-discipline; convergence-divergence;

relaxation-alertness; patience-impulsiveness; thinki ng-feeling; perceiving-deciding;

learning-problem solving. The creative problem solving profile is constructed on two

of these dimensions: two opposite ways of gaining knowledge: ( 1 ) by direct

experience and (2) by abstract analytical, logical thinking; and two opposite ways of

using knowledge: ( 1 ) by ideation (making new possibilities, breaking connections,

diverging) a nd (2) by eval uation (testing and verifying new possibilities, making

connections, converging), as shown in Figu re 2 .

2. As a complete process, creativi1y is about more than just "getting ideas." It starts with

problem sensing and ends in action, and involves evaluation and convergent thin ki ng

as well as ideation and divergent thinking. 1II-1- ;r-,... rJ:�,..:nl:norl nr-1""\roec

1'1 Y \.AJV,._l t"" llll\iJY f"" '"""''"'VW.;;l'
nn
d
'-'fll
h nc CO\l.c.rn l
11""4""7 W""Y"-'1'-"1

different stages.
 33

3. Different people have differing skills i n the various stages of the creative process. We

can improve our ski l ls i n a l l stages (weak as wel l as strong) .

4. Organizational creativity requires teamwork, in which people of differing incli nations

and strengths complement each other to initiate, develop and implement

improvement ideas and resolve problems.

5. By diagnosing and explaining relative differences in creative problem-solvi ng

incl i nations and ski l ls, the CPSP enables individuals to understand how to team up

for synergy and how to combine their strengths for g reater creativity.

RELIABILITY AN D VALIDITY

Administeri ng the CPSP to hundreds of people during creativity training and

application sessions has demonstrated the instrument's excellent reliability and face validity.

People who take the i nventory more than. once usual ly report stable scores over time.

Changing scores a re usua l ly explained by such factors as job changes that require different

thinking and problem-solving modes. Most users state that the profile makes much sense

to them. It helps them better u nderstand the concept of a "complete process" of creative

problem solving and appreciate individual differences within such a complete process.

Excellent learnings result when individuals complete the CPSP within a group, in which they

discuss thei r styles and their relationship to thei r experience.

A more formal program to investigate reliabi l ity and validity psychometrical ly has

been under way for some time. This section reports findings to date.
 34

RELIABI LITY

1. Test-Retest Rel iability

To estimate test-retest rel iability, two samples, n=129 and n=40, representi ng a wide

variety of business and industrial organizations and job types were administered the CPSP

i nventory twice, one week apart. Correlations of the fou r columns and two bi-polar column

scores for the two occasions were then calculated, as shown in Ta ble 3.

Table 3

Test-Retest Correlation Coefficients for Two Samg.les

Scal e ( n = 1 29) (n=40)

x (Col umn 1) . 64 .66
I (Col umn 2) .60 .65
T (Col umn 3) . 64 . 64
E (Col umn 4) .58 .69
X-T (Col umns 1-3) .66 .71
1-E (Columns 2-4) .67 . 75

All correlation coefficients a re statistically significant at the p < . 001 level .

Thus, the correlations i n Table 3 range from .66 to . 75 for the bi-pola r scales, and

from . 58 to .69 for the columns, respectively. This represents satisfactory test-retest

reliability. The variations across the four columns, two bi-polar scales and two samples are

relatively small.

2. Solit-Half Reliability

The i nternal consistency of the instrument was investigated using the random,

para l lel, split-half method for four separate, i ndependent sa mples and one large sample
 35

which combi ned the fi rst three of the four samples with several additional samples. The 1 2

item g roups of the i nventory were randomly assigned to two parallel, split-half inventories

of six item g roups each. From each of the two parallel, six-item, split-half i nventories that

resulted , each partici pant's fou r q uadrant scores were computed, as were correlation

coefficients for each of the four quadrants. The Spearman-Brown corrected correlation

coefficient (rxx) was calculated for each q uadrant as the reliability estimate for the composite

1 2-item inventory. (The composite inventory was the sum of the scores of the two parallel

halves.) The results across the five samples a re displayed in Table 4.

Table 4

Spea rman-Brown Corrected C orrelation Coefficients (rxx)

for Quadra nt Scores Across Five Samples

Sample 1 Sample 2 Sample 3 Sample 4 Sample 5

CPSP n = 1 56 n=1 29 n = l Ol n = 1 67 n = 1 63 9
Quadra nt
rxx rxx � rxx rxx
Quad 1 : .65 .66 . 71 . 64 . 70
Quad 2 : .62 .66 . 73 . 72 . 64
Quad 3 : .62 .67 .67 .66 .69
Quad 4 : .63 .69 .71 . 71 .62

Al l correlations a bove a re statistically significant (p< .001 ) .

Thus, the Spearman-Brown corrected reliability estimates calculated above for the

four q uadrants of the Creative Problem-Solving Profile Inventory ranged across the quadrants

within each of the five samples from .62 to .65, .66 to . 69, .67 to . 73, .64 to . 72, and .62

to . 70, and within the four quadrants across the samples from . 64 to .71 , .62 to .73 , .62 to

. 69 and .62 to . 71 . This i ndicates satisfactory levels of consistency reliability within each

sample and across samples.

 36

3. I nter-Correlations of Sca les

Pearson correlation coefficients were calculated among the fou r q uadrants and

among the four basic scales (col umns) and between the two bi-polar scales for the five

samples above. (See Tables 5 and 6.)

I n Table 5, as expected from both theory and scale construction, the correlations

between the scores on the pairs of opposite q uadrants ( 1 and 3, 2 and 4) were strongly

negative across all five samples (-. 98, -.92, -.90, -.96, -.98, and -.97, -.92, -.87, -.97, -.97,

respectively) . Also as expected from theory, the correlations between scores on pai rs of

adjacent quadrants were low. I n sample 5, the range was - . 1 1 to .05; in sample 4, -.02 to

. 03; in sample 3 , . 02 to . 05; in sample 2, - . 23 to . 1 6; and i n sample 1 , - . 23 to .22 .

The strongly negative correlations between opposite q uadrant scores indicate that the

CPS P inventory in Figu re 1 0 discriminates between opposite process style concepts

represented by these q uadra nts. The low adjacent q uadrant correlations indicate satisfactory

i ndependence of adjacent quadrants despite the sharing of one col umn (scale) between each

adjacent q uadrant pair.

I n Table 6, as expected from theory, the correlations between the Bi-polar X-T

(column 1 -3 ) and the Bi-polar 1- E (col umn 2-4) scales were very low in all five samples (.06,

. 05, . 1 1 , -.07 and . 1 1 , respectively) . This indicates that the two scales are independent a nd

that the two dimensions of the CPSP -- gaining and using knowledge (see Figure 4) -- a re

separate constructs as intended. I n Table 6, as expected, a l l X:T and l : E correlations were

�.i.��� g l., ��� .i.:' "' I .t.. O ....,.f..--.A, - f.. 8 I roc

, _...,p or+ivol"\
�l l V l l • y l l 'Ci� 0 I I Vv \ - . v / I .t.. 7
-.v1 I
_ .t.. O .t,.r;,
. v ,r I - . v v , - . v v
_ _ AA a nrl 7Q.
, , ,.,,.. - . , '°"'I
_ _ AO
- . ,,,, I
I -.,
_ 7 QI _
• ·- __ , , ,. _ , , , .

This supports the notion that opposing concepts a re being measu red at the poles of each of

the two major dimensions of the CPSP.

 37

Table 5

Pea rson Correlation Coefficients for Quadra nt Scores

Sample 1 Sam ple 2 Sa mple 3 Sample 4 Sam ple 5

Quadra nt Scales Correlated n = l Ol n = 1 38 n = 1 07 n = 1 67 n = 1 63 9

(1 ) Opposite Quadra nts

Quadra nt 1 : Quadrant 3 -.98 -.92 -.90 -.96 -.98
Quadrant 2 : Quadrant 4 -.97 -.92 -.87 -.97 -.97
(2) Adjacent Quadra nts
Quadra nt 1: Quadra nt 2 .22 .16 .05 -.02 . 05
Quadra nt 1: Quadrant 4 -.23 -.23 .02 .03 -.1 1
Quadra nt 2: Quadrant 3 - . 23 .00 .04 -.02 -.03
Quadra nt 3: Quadrant 4 .20 .13 .03 -.03 .05

Table 6

Pearson Correlation Coefficients for Basic a n d Bi-aolar Scales

Sampl e 1 Sample 2 Sample 3 Sample 4 Sa mple 5

Scal es Correlated n= l Ol n = 1 38 n = 1 07 _n= 1 67 n = 1 639
(1 ) Bi-pola r
X-T : 1 - E .06 .05 .1 1 -.07 .1 1
(2) Basic O pposite
x T - . 69 -.67 -.69 - . 65 - . 66
I E - . 78 -.69 - . 70 -.64 -.68
(3) Basic Adjacent
x I -. 1 8 - . 20 -. 1 0 -.3 1 -. 1 6
x E -. 1 6 -.24 -. 1 2 -.07 -.26
T I -.20 -. 1 2 -.30 -. 1 6 -.26
T E . 04 .02 . 05 -. 1 4 . 03
 38

The correlations between adjacent scales X and I were -. 1 8, -.20, - . l 0, - . 3 1 and -. 1 6,

and between adjacent scales X and E were -. 1 6, - . 24, -. 1 2, - .07 and -. 26, respectively. The

correlations between adjacent scales T and I , and T and E, were - . 20, -. 1 2, - . 30, -. 1 6, -. 26

a nd .04, . 02, . 05, - . 1 4, . 03, respectively. As predicted by theory, most of these correlations

a re quite low and represent satisfactory independence and internal consistency. The T and

E scales were the most independent of the adjacent pairings.

The low correlations between the bi-polar X-T and 1-E scales (ranging from -.07 to

. . 1 1 ), the strongly negative correlations between the opposite poles on the two dimensions

(ranges -.65 to -.69 and - . 64 to - . 78), and the relatively low correlations between the pai rs

of adjacent columns (XI, IT, TE and EX) indicate that ( l ) the two bi-polar scales and the four

colu mns taken in adjacent pai rs are satisfactorily independent; and ( 2 ) the opposing columns

within each of the two dimensions measure strongly contrasting concepts. This represents

good internal consistency for the CPSP instrument shown in Figure l 0. Furthermore, the

consistency of results across the five samples also indicates good parallel test rel iability for

i ntercorrelations of scales.

VALI D ITY

Five types of Va lidity were studied as described below.

1. External Validity by Correlation to Other Established Measures Related to C reativity

A. The Kirton Adaptation- Innovation I nventory Theory (KAI) defines and measures two

personal ity types relative to creative decision making (Kirton, 1 9 7 6 ) . Accoiding to the

Adaptation-Innovation theory, creativity styles measured by the KAI single-dimension

inventory range from highly adaptive to highly i nnovative.

 39

At one end of this conti n uum, adaptors characteristically use accepted definitions of

the problem and likely solutions in generating ideas. They examine these ideas closely a nd

proceed withi n the established theories, policies and practices of thei r organizations. Much

of thei r chan ge effort involves im provi ng and "doing better." This description is very

consistent with the description of the Optimizer stage of the CPSP (and opposite to the

Generator stage) .

Innovators are more l i kely to reconstruct the problem. By considering the problem

free of accepted thought, paradigms and customary viewpoints, they generate much less

expected, and probably less acceptable, sol utions. They think less about "doing things

better" than "doing things differently." This description is very consistent with the description

of Generators (opposite to Optimizers), who continually find new problems to solve and

initiate new projects.

Similarly, the I and E scales which measure opposing ways of using knowledge

a ppear consistent with the I nnovator and Adaptor KAI styles, respectively. The KAI

a pparently measures not different ways of gaining knowledge but different ways of using it.

As a result, the X and T scales of the CPSP are apparently unrelated to the KAI .

Because of their interest i n new ideas and their desire to correctly define the problem

and see the big picture, Conceptualizers are also consistent with the KAI I nnovator style.

I mplementers, who confine their activity to converting already developed plans and ideas

i nto action, a re consistent with the KAI Adaptor style.

Compa rin g the KAI I nnovator and Adaptor descriptions to the descriptions of the four

CPSP quadrants and to the I , E and 1-E bi-pola r use of knowledge distinctions leads to the

followin g hypotheses:
 40

Hypothesis # 1 : The Generator and Optimizer creative process styles should

correlate with the KAI, the former positively and the latter negatively.

Hypothesis #2 : The 1- E and I scale scores should correlate positively with the KAI

and negatively with the E scale.

Hypot hesis #3 : The X, T and X-T scale scores should not correlate substantial ly with

the KAI (because the KAI relates not to gaining knowledge but to usi ng it) .

Hypothesis #4 : The Conceptualizer and I mplementer process styles should correlate

with the KAI , as do the Generator and Optimizer styles, but to a lesser degree.

Testing. Two samples were administered both the KAI and the CPSP. The first

sample, n = 1 01 , consisted of manufacturi ng development engineers and the second,

n= 1 85, consisted of supervisors a nd managers from several busi ness and government

organizations and functions (and included the n = 1 1 0 sample described in the appendix).

Pearson Product Moment correlation coefficients were calculated for the following 1 1

variables and formed a correlation matrix for each sample: KAI score; total scores on each

of the fou r CPSP col u mns (X, I , T, E); scores calculated for each of the four C PSP Quadrants

(Quadra nt 1 , Generator style, product of col umns 1 and 2 scores; Quadrant 2,

Conceptualizer style, product of columns 2 and 3 scores; Quadrant 3, Optimizer style,

product of colu mns 3 and 4 scores; Quadrant 4, I mplementer style, product of col umns 4

a nd 1 scores) ; and scores for the X-T and 1-E bi-polar scales developed by subtracting T

scores from X scores and E scores from I scores.

The correlation matrices are shown here as Tables 7 and 8. The key findings noted

from scanning the tables are as follows. As predicted, Quadrant 1 , the Generator style,
 Table 7

Correlation of CPSP Columns a nd Quadra nt Scores with KAI Scores for Sample l, n = 1 01

Scale/Quadrant mean SD x I T E X-T 1-E KA I Xxl l xT TxE E xX

Experience (X) 30.2 5 .2

I deation (I) 28. 7 6.2 -. 1 8

Thinking (T) 30. 5 4. 7 -. 69 *** -.20 *

Evaluation (E) 30. 6 5.5 -. 1 6 -. 7 8 *** .04

(X-T) -0. 3 9.1 .93 *** .00 -.09 -. 1 1

( 1 -E) - 1 .8 1 1 .1 -.02 .95 *** -. 1 3 - . 94 *** . 06

KAI 98. 7 1 4. 5 .14 . 29 ** -.25 ** -.24 ** .21 * .29 **

Generator (Xx I ) 86 1 . 9 2 1 9. 1 . 55 *** . 72 *** -. 65 *** - . 7 7*** . 65 *** . 7 9 *** . 3 3 ***

Conceptualizer (I xT) 8 70. 1 208 . 9 -. 60 *** . 77*** .45 *** - . 68 *** -. 5 8 *** . 77 *** .12 .22 *

Optimizer (Tx E) 933.3 224.0 - .55 *** -. 7 1 *** . 66 *** . 77 *** -. 66 *** - . 78 *** - . 35 *** -. 98 *** -.23 **

I m plementer (ExX) 91 8.9 2 1 1 .0 . 60 *** -. 75 *** -.48 *** . 6 8 *** . 59 *** - . 76 *** -.08 - .23 ** -.9 7 *** .20 *

Coefficients marked *** a re statistically significant at p < .00 1 ; ** at p < .0 1 ; and * p < .0 5

.i::...
__,
 Table 8

Correlation of C PS P Col umns and Quadra nt Scores with KAI Scores for Sample 2. n= 1 85

Scale/Quad ra nt mean SD x I T E X-T 1-E KA I Xxl l xT T x lE ExX

Experience (X) 3 3 .5 5.1

Ideation (I) 29.2 5.2 -. 1 7 **

Thi nki ng (T) 2 8 .2 4 .3 - . 70 *** - .20

Evaluation (E) 29.2 5 .0 - . 2 7 ** - . 68 *** .04

(X-T) 5 .3 8. 6 . 9 3** .00 -. 91 ** - . 1 7*

(1-E) 0.0 9.4 .05 .92 *** -. 1 3 - . 9 1 *** .09

KAI 1 05 . 2 1 4 .8 .12 .44 *** -. 1 4 -.4 6 *** .14 .49 ***

Generator (Xx I ) 9 74.3 20 7 .4 . 6 1 *** . 6 7 *** -. 68 *** - . 74 *** . 70 *** . 7 7 *** .43 ***

Conceptualizer (I ><T) 82 1 . 8 1 88.4 - . 66 *** . 6 8*** . 5 7 *** - . 53 *** -. 6 7 *** . 66 *** . 25 ** .03 *

Optimizer (Tx E) 825 . 6 1 94.4 - . 67 *** -. 6 1 *** . 7 1 *** . 7 2 *** -. 74 *** -. 7 3 *** -.42 *** -. 9 7 *** .01

I m plementer (E xX) 9 74.5 1 95 .0 . 6 3 *** -. 7 0 *** - . 55 . 56 *** . 64 *** - . 6 9 *** -.28 *** - .09 ** -.9 7 *** .01

Coefficients marked *** are statistically significant at p< . 00 1 ; ** at p< .0 1 ; and * p< .05

�
""
 43

positively correlated with the KAI, in both samples (.33 and .43), and significant positive

correlations were found between KAI and 1 - E scores in both samples (.29 and .49) and

between KAI and I scores (.29 and .44) . Also as predicted, the Quadrant 3 (Optimizer style)

correlation with KAI was significantly negative i n both sa mples (-.35 and -.42) . Quadra nt

2, the Conceptualizer style, also correlated positively with KAI in both samples, but, as

predicted, at lower levels (. 1 2 , not significant, and .25) . The Quadrant 4 style, I mplementer,

also correlated negatively at lower levels (- . 08, not significant, and -.28). The correlations

between KAI and Quadrant 2 (Conceptualizer) and Quadrant 4 (Implementer) lend some

support to Hypothesis #4. As expected, there were low correlations between KAI a nd the

X-T, X, and T scales. The sample 2 correlations were lower (. 1 4, . 1 2, and -. 1 4) than

sample 1 (. 2 1 , . 1 4, and -.25) respectively. General ly, all four hypotheses were supported.

B. The Myers-Briggs Type I ndicator (MBTI) 1 is a personality inventory based on J ung's

theory of psychological types that is widely used in organizational psychology. The MBTI

produces fou r bi pola r scales measuring Extroversion-I ntroversion (El), Sensing-Intuition (SN) ,

Thinking-Feeling (TF) and J udgment-Perception (JP) . Many of the characteristics assessed

by these scales (Myers, 1 994) appear highly relevant to the CPSP quadrant styles.

The El scale i ndicates how strongly an i ndividual focuses internally or externally. The

Extrovert type focuses on external events, people, things and experiences, and prefers to

learn by doing. The I ntrovert type focuses on the in ner world of concepts and ideas, and

prefers to learn by reflection and thought. Extroverts a re said to be sociable and expressive,

and more i nterested and comfortable when they are working actively with people or thi ngs.
 44

I ntroverts are said to be private and contained, and more interested and comfortable when

thei r work involves ideas and thought.

The SN scale defines preferences for "fi nding out." Sensing types are said to be

factual , concrete and oriented to today, to val ue practical applications, and to trust

experience. They rely on what their senses tell them about their environment. I n contrast,

l ntuitives a re future-oriented, abstract and theoretical, concerned with the big picture, and

a ble to see patterns and meaning in facts; they a re also said to enjoy defining and solving

new, complex problems. They use thei r intuition to gain meaning, propose relationships and

possibilities beyond their physical senses, and consider what they might do about a situation.

The TF scale defines an individual's approach to decision making. Thinking types are

said to rely on logic, analysis and cause-a nd-effect reasoning when making decisions a nd

solvi ng problems, and to seek a detached and impersonal standard of truth . Feeling types

a re said to make decisions based on affective and person-centered values rather than on

objective standards of logic. They incorporate anything that matters or is important to them

or others without requiring that it be logical.

The J P scale measu res orientation to the outer world. Judging types seek to control

and regulate life, to make clear decisions and move on; they plan extensively, and a re

systematic, methodical and organized . Perceiving types a re said to be spontaneous, flexible

a nd cu rious, seeking to understand, experience and adapt rather than to control life; they

prefer to keep thei r options open rather than restrict themselves to fixed plans or schedules.

The MBT! Form G (I-Ayers & Myers, 1 984 ; 1-llyers & McCaulley, 1 985) is a 1 26-item

inventory produci ng four pai rs of preference scores. MBTI practitioners usual ly convert each

of the fou r pai rs of scores to a dichotomous preference, then assign an individual to one of
 45

1 6 personality types. A summary of the four bipolar MBTI type preferences is provided in

the MBTI Users Manual by Consulti ng Psychologists Press, I nc., and is shown in Tables 9 a nd

1 0. Several personality researchers (Hicks, 1 984; 1 985; McCrea & Costa, 1 989; Stricker

& Ross, 1 964) have suggested that rather than disti nct MBTI types, the MBTI really measures

fou r i ndependent and contin uous personality traits. Following this line of reasoning, a

research a lternative to dichotomizi ng the MBTI scores was adopted. Four continuous trait

scores, one for each dimension (shown in Table 1 1 ) were obtained by subtracting the two

scores representing the opposing poles of the dimension, and resetting the origin to a

neutral point of zero. MBTI conti nuous scores are scored in the direction of the second ­

named pole. For example, a score g reater than zero on the El scale represents a preference

for I ntroversion (I) and a score less than zero a preference for Extroversion (E).

A correlational study was performed using the MBTI and CPSP. The total number of

partici pants was 1 34 MBA students, most of them enrolled i n the co-operative education

program. The resu lts are shown in Tables 1 2, 1 3 and 1 4 .

The correlations i n Table 1 2 show that the Generator style was associated significantly

with a l l fou r MBTI scales, most strongly with Extroversion (.35) and Perception (.30) (both

p < . 0 1 ) but also with I ntuition (.25, p< . 0 1 ) and Feeling (.22, p < .05). Generators may,

therefore, characteristically focus externally rather than internally and prefer to employ their

Perceivi ng fun ction rather than thei r Judgi ng function. This is consistent with the idea that

they excel in the initial phase of the Simplex creative problem-solving process, which involves

scanning the external envi ronment to identify new facts and new problems to work on. This

is also consistent with the Feel ing characteristic of accommodating any relevant points no
 Table 9 46

Bi-Polar MBTI Type Preferences

EXTROVERTS I NTROVERTS
Like variety and action. Like quiet for concentration.
Tend to be faster, dislike complicated Tend to be careful with details, dislike
p roced ures. sweepi ng statements.
Are often good at g reeting people. Have trouble remembering names and faces.
Are often impatient with long slow jobs. Tend not to mind working on one project for
a long time uninterruptedly.
Are interested in the results of their job, in Are interested in the idea behind their job.
getting it done and in how other people do it.
Often do not mind the interruption of Dislike telephone intrusions and interruptions.
answering the telephone.
Often act quickly, sometimes without Like to think a lot before they act, sometimes
thinking. without acting.
Like to have people a round. Work contentedly alone.
Usual ly communicate freely. Have some problems communicating.

THI N KI N G TYPES FEELING TYPES

Do not show emotion readily and a re often Tend to be very aware of other people and
uncomfortable dealing with people's feelings. their feelings.
May h u rt people's feelings without knowi ng Enjoy pleasing people, even in unimportant
it. things.
Li ke analysis and putting things into logical Like harmony. Efficiency may be badly
order. C an get along without harmony. disturbed by office feuds.
Tend to decide impersonally, sometimes Often let decisions be influenced by their own
payi ng insufficient attention to people's or other people's personal likes and wishes.
wishes.
Need to be treated fai rly. Need occasional praise.
Are able to repri mand people or fire them Disl ike telling people unpleasant things.
\Vhen necessary.
Are more analytically oriented-respond Are more people-oriented-respond more
more easily to people's thoughts. easily to people's values.

Tend to be firm-minded. Tend to be sympathetic.

 Table 1 0 47

Bi-Polar MBTI Type P references

SENSI N G TYP ES I NTUITIVE TYPES

Dislike new problems unless there a re Like solving new problems.
standard ways to solve them.
Like an established way of doing things. Disli ke doing the same thing repeatedly.
Enjoy usi ng ski lls already learned more than Enjoy learning a new skill more thanusing it.
learning new ones.
Work more steadily, with realistic idea of Work in bursts of energy powered by
how long it wi l l take. enthusiasm, with slack periods in between .
Usually reach a concl usion step by step. Reach a conclusion quickly.
Are patient with routine details. Are impatient with routine details.
Are i mpatient when the details get Are patient with complicated situations.
complicated.
Are not often i nspi red, and rarely trust the Follow their inspirations, good or bad.
inspiration when they a re.
Seldom make errors of fact. Frequently make errors of fact.
Tend to be good at precise work. Dislike taking time for precision.

J U DG I N G TYPES PERCEPTIVE TYPES

Work best when they can plan their work Ada pt well to changing situations.
and follow the plan.
Like to get things settled and finished. Do not mind leaving things open for
alterations.
May decide things too quickly. May have trouble making decisions.
May dislike to interrupt the project they a re May sta rt too many projects and have
on for a more u rgent one. difficulty finishing them.
May not notice new things that need to be May postpone unpleasant jobs.
done.
Want only the essentials needed to begin Want to know all about a new job.
their work.
Tend to be satisfied once they reach a Tend to be curious and welcome new light on
judgment on a thing, situation, or person. a thing, situation, or person.
 m
c...
c -I (/)
?S.
c... :::r
- · CD
:l
a
co :l <
� t-3 - ·
en en CD l:Ij
=i= •
3 '1'
:l Cil
- ·
CD :l
:l c.c c.c
-+ 0
:l
...-.... ...-.... ...-... ...-....
I I I I
...._... ...._... ...._... ...._...
-I
:::r
CD
-n
0
c
...
�
C:J
:::!
""'CJ �
...
CD Q::
0 0 0 0
� CD
... ......
CD
::l ......
n
(!)
0
3
CD
::l
(/l
0
::l
(/l
...-.... ...-... ...-.... ...-....
+ + + +
...._... ...._... ...._... ...._...
""'CJ :l
CD .....
..,
11 :l ..,
n CD ..... c
- ·
CD CD c <
ti--4
"'C
- ·
-+
�
:l
'Tj - ·
-+ z CD
0 (ii
- ·
0 c.c :l
:l 0
8P' :l
 49

Table 1 2

Pearson Correlations Among MBTI a nd C PS P Quadra nt (Style) Scores

MBA Students (N = 1 34}

C PSP Styles
MBTI Sca l es
Generator Conceptual izer Optimizer Implementer
(Quadrant 1 ) (Quadrant 2 ) (Quadrant 2) (Quadrant 4 )

EI Pearson Correlation - . 35** .01 . 3 7** - .02

Sig. (2-tailed) .00 .90 .00 .82
SN Pearson Correlation . 25** .38** - . 24** - . 36**
Sig. (2-tailed) . 00 . 00 .01 . 00
TF Pearson Correlation . 22* - .07 - . 1 7* .07
Sig. {2-tailed) .01 .44 . 05 .41
JP Pearson Correlation .30** . 20* - . 3 1 ** - .2 1 *
Sig. (2-tailed) . 00 .02 .00 .01

* Correlation is significant at the 0.05 level (2-tailed) .

** Correlation is significant at the 0.01 level (2-tailed).

Table 1 3

Pearson Correlations Between MBTI Scores and C PSP Bipolar Scores

MBA Students (N = 1 34)

CPSP Bipolar Scales

MBTI Sca l es
I- E X-T
EI Pearson Correlation - . 23** - . 25**
Sig . (2-tailed) .01 .00
SN Pearson Correlation .45** -.13
Sig. (2-tailed) .00 .15
TF Pearson Correlation . 08 . 1 8*
Sig. (2-tailed) .39 . 04

JP Pearson Correlation .36** . 05

Sig. (2-tailed) . 00 .5 7

* Correlation is significant at the 0.05 level (2-tailed) .

** Correlation is significant at the 0.01 level (2-tailed) .
 50

Table 1 4

Pearson Correlations Between MBTI Scores and CPSP Column Scores

MBA Students (N = 1 34)

C PS P Colu mns
MBTI Scal es
Col u m n l Col u m n 2 Col u m n 3 Col u m n 4
Gaining Using Gaining Using
Knowledge by Knowledge for Knowledge by Knowledge for
Experiencing Ideation Thi n ki ng Eval uation
(Physical (Creating (Mental (Choosing
Processing 1 ) Options) Processing) Options)

EI Pearson Correlation - .25** -.16 . 20* . 27**

Sig. (2-tailed) .00 .07 .02 .00
SN Pearson Correlation -.10 .38** .13 - .44**
Sig. (2-tailed) . 26 .00 .13 .00
TF Pearson Correlation . 1 9* .04 -.14 - .09
Sig. (2-tailed) . 03 .61 .1 1 .28
JP Pearson Correlation .04 .30** - .05 - .36**
Sig. (2-tailed) .61 .00 . 60 .00

* Correlation is significant at the 0.05 level (2-tailed) .

** Correlation is significant at the 0.01 level (2-tailed).

matter how il logical, and with the I ntuitive characteristics of enjoying new problems and of

being oriented toward the future and the big picture.

The Optimizer style correlations were exactly opposite to the Generator style. All four

correlations with the MBTI were significant (I , .37, p < . 0 1 ; J, . 3 1 , p < . 0 1 ; S, . 24, p < . 0 1 ; a nd

T, - . 1 7 , p < . 05) . Thus, the MBTI preferences most strongly associated with the Optimizer

style are I ntroversion and J udging. An introverted focus is consistent with the Optimizer style

preferences for mental testing of ideas, worki ng on one project at a time, and for dealing

with thi ngs rather than people. Their preference for J udging is consistent with their
 51

preference for the evaluati ng, selecti ng, planning a n d organizing stage of the Simplex

creative problem-solving process. The Thi nking preference is consistent with the Optimizer's

reliance on logic, analysis and reasoning in problem solving, and the Sensing preference is

consistent with the orientation toward practical solutions and the aversion toward add ressi ng

many new problems at the same time.

The results also show that Conceptualizers are characterized by a clear preference

(.38, p < . 0 1 ) for I ntuition. I ntuitive individuals seek to grasp patterns, and try to understand

relationshi ps and make connections between facts; they a re also said to do wel l at seeing

new possi bil ities and different ways of doing things. This is entirely consistent with the

thinking style of the Conceptualizer, and the Conceptua lizer's preference for the problem­

definition and idea-finding steps of the Simplex creative problem-solving process. There was

a lesser but significant (.20, p < . 05) correlation with the Perception type, consistent with the

Conceptualizer style's seeki ng to understand and keeping options open.

The I mplementer style (the opposite of the Conceptualizer style) was associated a lmost

equal ly strongly with the opposite types -- that is, Sensi ng (.36, p < . 0 1 ) and Judging (. 2 1 ,

p < . 05) . People who prefer Sensing are said to be factual, concrete and practical , a nd to

val ue experience and application -- characteristics that seem consistent with a preference for

real-world i mplementation. The Judging orientation is consistent with the I mplementers'

desire to move on, get things finished, and follow through thei r plans without interruption.

The results suggest that preferences for performing successive stages of the Simplex

creative problem-solving process are related to preferences for different MBTI functions. At

different stages, the poles of various dimensions predominate. The first stage, Generating,

is related to Extroversion and Perceiving and also to I ntuition and Feeling. Generators are
 52

sensitive to their envi ronment, scanning for new problems to solve, avoid formi ng

concl usions or closing off options, and focus on absorbing a variety of sti muli. They see

opportunity in al most everything, and enjoy the ambiguity inherent in peoples' problems

which involve feelings and val ues. In the next stage, the I ntuitive function hel ps discern

patterns in data, to make new connections and to conceptualize new insights into how the

problem should be defined and modeled. The Conceptualizer begins to consider the future

by looking at new possibilities, hatching new ideas and keeping them open. Next, the

Optimization stage is governed by I ntroversion and J udging, and to a lesser extent, by

Thinking and Sensi ng. By focusing inward, the Optimizer subjects solutions and ideas to

sustained and systematic mental scrutiny using concrete, logical criteria, before making a n

opti mal choice. The practicality of that choice is governed by the realities of the situation,

not a plethora of intuitive speculation. Finally, during the implementation stage, the Sensing

function predominates, with Judgment supporting. The I mplementer deals with the concrete

and current and focuses on converging upon a successful finish, not on diverging to open

up new opportunities.

2. Predictive Validity: Moderation of Traini ng Effects

A field experiment was conducted to examine how individual creative problem-solving

style affects training in creative thinking. This intensive hands-on training emphasized the

Simplex CPS three-phase process which synchronizes divergence and convergence in

problem fi nding, problem solving and i mplementation . T\.YO attitudes associated V'1ith

divergent thinking were measured before and after training. The sample (n= 1 56) consisted
 53

of both managers (n= 90) and non-managers (n= 66) and a variety of functional specialties,

hierarchica l levels and types of busi ness organizations.

The primary hypotheses were that, compared to the other three styles (generator,

conceptual izer and implementer) , participants with an opti mizer style who received trai ning

in the three-phase process wi ll show:

H10 : Greater increases in preference for ideation (active divergence) ; and

H1b : Greater decreases in tendency to make premature critical evaluations of ideas

(premature convergence).

A six-item preference for ideation scale and an eight-item tendency to make premature

critical evaluations of ideas scale were used to measure "active divergence" and "premature

convergence" attitudes. The items a nd scales a re ful ly descri bed in Basadur and Finkbeiner

( 1 985) . The major finding of this study was that, as predicted by hypotheses H10 and H1b ,

practice with creative techniques and processes was particularly effective for people with a n

opti mizer style of creative problem solving. For the entire sample, optimizers made g reater

gains than the other three styles on both attitudes. All four creative problem-solving styles

improved significantly on the premature convergence attitude, but the improvement on the

active divergence attitude was statistical ly significant only for the optimizers. For the non­

manager subsample, the advantages for the optimizers were the same as those of the entire

sample; only the advantage over conceptualizers was not statistical ly significant. For the

manager subsample, the advantages for the optimizers were the same as those fou nd for

the non-manager subsample on premature convergence. On active divergence, there were

no advantages for the optimizer. Thus, hypothesis H 1 b was strongly supported for the entire
 54

sample and hypothesis H 1 0 (active divergence) was strongly su pported for the entire sample

and the non-manager subsample, but not for the manager subsamples.

Table 1 5 presents the mean differences in the two attitudes for the four creative

problem-solving process styles for the entire sample and for the manager and non-manager

subgroups separately. For the overall sample and the active divergence attitude, only the

optimizer style showed a significant gain, t(30) = 2. 1 , p < .05, fol lowing training. Tukey's

tests indicated that this gain was significantly g reater than the average of the other three

styles and significantly g reater than for either the conceptualizers or implementers (p < .05) .

For the premature convergence attitude, all four styles showed significant gains.

For the non-manager subsample and the active divergence attitude, only the

optimizer style showed a significant gain, t(l 0) = 2.3, p < . 05, following training. Tukey's

tests indicated that this gain was significantly g reater than the averages of the other three

styles and the implementer style. For the premature convergence attitude, all fou r styles

showed significant gains fol lowing training. Here again, Tukey's tests following the ANOVA

indicated (p< . 0 1 ) that the gain by the optimzer style was significantly g reater than the

average of the other three styles, and significantly g reater than the implementer style

(p< . 05).

For the manager subsample, the gain in the active divergence attitude fol lowi ng

training was not significant for the opti mizer, implementer or conceptualizer styles. Only the

gain for the generators was significant, t(9) = 2.6, p < .05. None of the comparisons of the

gains of the four styles vvas significant. For the premature convergence attitude, a ! ! four

styles showed significant gains (p< .0 1 ) following training. Tukey's comparisons indicated
 Table 1 5 55

Means. Standard D eviations (in 12arentheses) and Differences in

D ivergent Thi n ki ng Attitudes for the Four P roblem-Solving S:!yles
Active Divergence Premature Convergence
t
Style Before After for chan9e Before After for chan9e
Total Samele n = l 5 6)
Generator
( n =30/29) 2 1 .0 23.1 25.3 1 8.4b 3 .9**
(4 . 1 ) (4.5) (5 .4) (5 .4)
Conceptualizer
( n =4 7/46) 2 1 .5 23.l 2 3 .4 1 7.4 3 . 7**
(3.6) (4. 0) (5.2) (5 .4)
Optimizer
(n = 3 1 /30) 20.7 23.6 2.1 * 24.8 1 5.9a,c 4 .5**
(4 .2) (4.3) (5. 7) (5 . 7)
Implementer
( n =48/47) 20.4 2 2 .2 24 . 7 1 9.?b 3 . 6**
(4 . 1 ) (4.2) (5 .0) (5.2)
(ANOVA F) (ns) (ns) (3.8)* (ns) (5.2)**

Non-Managers {n = 6 6}
Generator
( n =20/l 9) 2 1 .8 23.0 26.6 1 8. 9 3 .6**
(4.3) (4 .4) (6 .0) (5.8)
Conceptua I izer
(n= 1 5/1 4) 2 2 .4 23.7 24.5 1 7.8 3.6**
(3. 9) (4 .2) (5 . 4) (5. 3)
Optimizer
(n = l l /1 0) 2 2 .0 25.3a,c 2 .3* 24.4 1 4.5a,c 4 . 7**
(4 . 1 ) (3.9) (5.6) (5.5)
Implementer
( n = 2 0/ l 9) 2 1 .6 2 2 .3 b 25.6 20.3b 3.5**
(4.3) (4. 1 ) (5.2) (5 .4)
(ANOVA F) (ns) (3.6)* (3.8) * (4 .9)* (6. 1 )**

Managers {n =90)
Generator
( n = l 0/9) 1 9.4 23.2 2 . 6* 22.8 1 7 .4 3 .4**
(4 .5) (4.2) (6.2) (5 .8)
Conceptualizer
( n = 32/3 1 ) 21 .l 22.8 22.8 1 7 .2 3 .4**
(4 . 1 ) (4 .0) (5.5) (5. 7)
Optimizer
(n=2 0/l 9) 20. 1 22.7 25. l l 6 . 7a 4 .0**
(4. 1 ) (3. 8) (5.2) (5.5)
Implementer
(n=28/2 7) 1 9.6 22.1 24.1 l 9.2b 3 .2 **
(3.9) (4 .0) (5.2) (4. 9)
(ANOVA F) (ns) (ns) (ns) (ns) (5.7) ** (5 .5)**
Note: Values denoted .9. are significantly different from those denoted )2 at the p< .05 level. s indicates that Optimizers are different from
the average of the other three styles at the p < .05 level. Tests were concluded based on Tukey's method following ANOVA.
* and ** denote significant changes at p< .05 and p < . 0 1 respedively comparing After-Before means based on t tests for the
matched sample (Kerlinger, 1 973).
ns - Non-significant
 56

that the gain by the opti mizers was significantly greater than the average of the other three

styles, and significantly g reater than the gain by the implementers (p< .05).

3. Predictive Validity: Accuracy of I nventory Assessment of CPS Profi le as C ompa red

to Self-Assessment an d Consensus with Expert Partner

The ability of the CPSP i nventory to predict a subject's profile compared to the

estimate of the subject's profile made by consensus with an expert partner (one who knows

the subject well) was tested on a base of n = 24 managers from industry.

Each consensus pair ranked the fou r q uadrants in the order they agreed should

describe each subject best. This ranking was compared to that provided by the inventory

itself for all 24 sets of scores as fol lows. Because there are four quadrants to rank order,

there a re four "slots" for each subject to compare for agreement as shown in Figure 1 2 (total

96 slots for 24 subjects). For each slot, there would be 1 00% agreement (same quadrant

picked for this slot), zero agreement (opposite q uadra nt picked for this slot) or partial

agreement (an adjacent q uadrant picked for this slot) . Then a one-way table chi-squared

test containing three cel ls was performed.

The form used for ran king is shown below in Figu re 1 2. Table 1 6 shows the results

of the degree of agreement test between prediction by "i nventory" and prediction by "expert

partner and self-consensus" for this sample.

A chi-square val ue was computed as follows:

X2 -
_ (4i:;;: _ l")..t)2 /l") A
.J L..... I L .....
+ / 1 7_
\ I , 1") A )2/'J4
L....,. L
..L
I
( 'l A _ A Q.) 2 /A Q.
V""'T -r\J I """T ....,
- 4.
�
'JA J:\ ')
"""T . '-' L

The result is significant at the p < .001 level with d.f. = 2.

 57

Fig u re 1 2

Form Used by Each S u bject for

Ran k Ord eri ng of Quadra nt Scores for Self

Scale: •
a score of 4 indicates the quadrant ranked highest (most like me)
•
a score of 3 indicates the quadrant ranked 2"d highest (second most like me)
•
a score of 2 indicates the q uadrant ranked 3rd highest (third most like me)
•
a score of l indicates the q uadrant ranked lowest (least like me)

Ran k Ord ering

Ql Q1 Q3 ill.
Ran ki ng Method
1 . My prediction of my profile is

2. My partner's prediction of my
profile is
3. Our consensus of my profile is
(by discussion after separate
p redictions i ndependently
made)
4. The CPSP i nventory's prediction
of my profile is

Table 1 6

Agreement Comparison: Actua l vs. Random Chance (n = 24)

Agreement Agreement Expected

Observed by Random Chance
1 00% Agreement 45 (47%) 24 (25%)
(Same Quadra nt picked for slot)
0% Agreement 1 7 ( 1 8%) 24 (25%)
(Opposite Quadrant for picked slot)
Partial Agreement 34 (35%) 48 (50%)
(Adjacent Quadrant picked for slot)
Total Slots 96 ( 1 00%) 96 ( 1 00%)
 58

Discussio n

The null hypothesis i s that there will be random chance (25%/25%/50%) distribution

in the three cells as fol lows:

Expected Expected Expected

25% 25% 50%
% of "slots" agree % of "slots" disagree % of "slots" partial ly agree
(have identical quadrants) (have opposite quadrants) (have adjacent quadrants)

e.g . : Ql vs Ql e.g. : Q l vs Q3 e.g. : Q l vs Q2 or Q3

Q2 vs Q2 Q2 vs Q4 Q2 vs Q l or Q3
Q3 vs Q3 Q3 vs Q2 or Q4
Q4 vs Q4 Q4 vs Q l or Q3

If the inventory is invalid, the agreement wi ll be a purely random chance distribution

across the base i.e., 25%/25%/50%. But, if the inventory is valid, we should get significantly

more than 25% in the first cell (agreement) and significantly less than 25% in the second cel l

(disagreement) .

For example, for subject #1 :

Most 2 nd Most I 3 rd Most Least

Consensus Ranking is Ql Q4 Q3 Q2
I nventory Ranking is Ql Q4 Q2 Q3

Then % of slots agree is 50%; % of slots disagree (opposite) is 0%; and % of time slots

partially agree (are adjacent) is 50%. This provides a 50%/0%/50% distribution vs the

random chance distribution of 25%/25%/50% for this subject.

I n Table 1 6 , we have a 46.9%/ 1 7. 7%/35.4% distribution. To test whether we may

reject the null hypothesis, the Chi 2 test is performed as follows:

 59

For each cell, (Observed Frequency - Expected Frequency)2 is divided by (Expected

Frequency) then summed .

I n the n = 24 sample, the sum is:

(45 - 24 ) 2
Cell 1 :
=

1 8. 3 75
24

( 1 7 - 24 ) 2 =

Cell 2 : 2.04 2
24

( 34 - 4 8 ) 2 =
Cell 3 : 4 . 083
24
Total 24.52

Total d.f. = 2

X2 = 24.2 > . 01 (al pha = 9. 2 1 ) and the null hypothesis can be rejected (p< .00 1 ) .

Predi ctive Val id ity: Rating of D egree of Fit of C PS P rofile as Compared to Subject's Own

Perception of Own Actual C reative Problem-Solving Style

The following table shows a "goodness of fit" between the CPSP styl es i nventory

assessment a nd the participants' own perceptions of their creative problem-solving process

styles. Participants used a 1 - 1 0 scale to indicate their estimates of "goodness of fit."

In Table 1 7, the mean ratings of how wel l the CPS Profile matched individuals'

perceptions of their own creative problem-solving process style are 7 . 1 and 8.3 for

u ndergraduates and managers respectively, based on a 1 0-point scale. The fit was rated

as 7 (good) or higher by 72 .4% of the undergraduates and 92. 9% of the managers.

Overa l l , a high proportion of the undergraduates a nd managers considered the

profile as a good predictor of their personal process style of creative problem solving.

Managers' ratings were even higher than those of undergraduates.

 60

Table 1 7

Summary of Rati ngs of Degree of Fit of C PS Profil e With

S ubject's Perception of Own Actual C reative Problem Solving Style

Rating Frequency (%) Average Rating

Study # 1
Busi ness 10 (3 .3%)
Undergraduates 9 (8.9%)
(n= l 81 ) 8 (32. 6%) 7. 1
7 (27.6%)
6 or below (27. 6%)
1 00.0%
Stu dy # 2
Managers 10 (2 1 .4%)
(n = 1 4) 9 (28. 6%)
8 (28.6%) 8.3
7 ( 1 4 .3%)
6 or below (7. 1 %)
1 00.0%

4. P redictive Val idity: Optimal I deation-Eval u ation Ratios Theory

Basadur et al ( 1 982) suggested that an interesting line of research would be to

explore the relative contri butions of ideation and evaluation during each of the three phases

of the process of Figure 1 . For exa mple, do these relative contributions differ by task or field

of work? Perhaps in high-pressure jobs oriented strongly toward implementation, eval uation

(convergence) is relatively more important than ideation (divergence). Perhaps there a re

opti mal ideation-evaluation ratios that differ by phase for any job or organization.

The ideation-evaluation thinking process in no way implies a discarding of judgment.

On the contrary, Figure 1 models a process i n which people alternate thei r judgmental and

non-judgmental thinking throughout a multi-phase process. In an applied ind ustrial

 61

environment focused on profits and quick action, increasing creative behavior would l i kely

involve i ncreases in both ideation and evaluation skill. However, on a day-to-day basis, this

environment would probably require more of a converging, judgmental (evaluational)

attitude than a diverging, perceptive (ideational) attitude. Environments that a re less ti me­

sensitive, such as pure research, might favor divergence and perceptiveness (ideational) at

the expense of convergence and j udgment (evaluational).

Training in a "complete creative problem-solving process" emphasizing ideation­

eval uation at each phase (see Figu re 1 ) encourages participants to balance ideation a nd

evaluation. This balance is i ll ustrated in the educational film, "The Dot and the Line"

(Norto n , 1 96 5) , which shows the pitfalls of the two extremes of total rigidity and total

flexibility, and promotes an opti mal concept cal led "disci plined freedom."

Second, such training might move individuals from either extreme on an ideation­

evaluation spectrum toward some optimum (depending on the type of work or field of

endeavor i nvolved) . The location of this optimum may differ for work requiring more

problem finding (e.g . , pure research), more problem solving (e.g., applied research) or more

sol ution implementation (e. g., manufacturing). This thinking is depicted in Figure 1 3 .

The n otion of different optimal ratios of ideation and evaluation in varying phases is

ill ustrated in yet another way i n Figure 1 4. Figure 1 4 represents a revision of the basic

model in Figu re 1 . The i deation and evaluation symbols a re of equal length in Figure 1

withi n and among the three phases. I n Figure 1 4, however, the lengths are similar only

within Phase I I ; the ideation symbol is longer than the evaluation symbol in Phase I and vice

versa in Phase I l l . These ratio differences in Figure 1 4 illustrate differences in each phase

in the opti mal balance between ideation activity and evaluation activity.
 62
Fig u re 1 3

Possi b l e Opti m a l I d eati o n - Eva l uation P reference Ratios

for Di fferent Fields of E nd eavor

Field of Endeavor

Emphasis: Problem Finding

(e. g . , Pure Research)
x-i
( Phase I*)
Preference
for
Opti mu / Preference
for
Evaluation I deation

Emphasis: Problem Solving

i--����-x��
(e.g., Applied Research)
(Phase I I*)
Preference
t
Optimum Preference
fo r for
Eval uation I deation

Emphasis: Sol ution I mplementation

(e.g., Manufacturing)
�x -----i.

(Phase I l l *)
Preference -Optimum Preference
for for
Evaluation Ideation

* See Figu re 1
 63
Fig u re 1 4

Revisi ng Fig u re 1 i n t h e Light of Opti m u m I deat ion-Eva l u ation Ratio Theory:

A C omplete C reative P ro b l e m Solvi ng Process E mphasizi ng I deation- Eva l u ation
in E ach of T hree P hases B ut in D i fferent Ratios

N ote :

The three q uadrilateral figures

representin g the three phases
a re a l l equal i n area. This
represents equal ti me or equal
activity. The ratios of ideational PROBLEM
and evaluational time or activity FINDING Phase I
a re different i n the three
quad ri laterals.

PROBLEM
Ph ase I I
SOLVING

Ph ase I l l
 64

Third, throughout the 11complete process" (Figure 1 ), the relative desirable emphasis

on eval uation with i n a given field of endeavor may increase within each phase but from

different starti ng points (and vice versa for ideation). This concept is illustrated in Figure 1 5.

Perhaps for most decision-oriented problem-solving professions, such as law,

medici ne and engi neering, the opti mum will tend toward the E side. Training in a 11complete

creative problem-solving process" emphasizing ideation-evaluation wil l promote a

preference for such an opti mum. The opti mal ratio between time devoted to ideation activity

a nd time devoted to evaluation activity (l/E) would then be relatively lower than the optimal

ratio for more pure research professions. This optimal ratio (l/E) would be even l ower i n

implementation fields such as manufacturing. In contrast, for other pursuits emphasizing

problem fi nding and with fewer time restrictions, such as pu re resea rch, the opti mal l/E

ratios would be relatively higher. In the former cases, the range of 11good judgment"

decisions might be much tighter than in the latter cases, requiring more attention to

eval uation. An a rchitect or writer might face diverse acceptable problems and sol utions.

For a physician or manufacturi ng engineer, the number of possible good problems a nd

sol utions may be more heavily constrained by the short-term needs of the patient or

organization.

I ntuitively, the 11 ideation-eval uation ratio" seems an appealing notion. A possible

direction for future research is to determine whether optimal ideation-evaluation ratios exist

on the spectrum for each phase of the 11 complete process of creative problem solving" (see

Figure 1 and Figure 1 3) for various fields of endeavor. Another research route is to

determine whether optimal ideation-eval uation ratios change with increasing eval uation at

each succeeding phase of the 11complete process of problem solving" (see Figure 1 and
 Figu re 1 5

Possible Opti m u m I deation-Eva luation Preference Ratios for Each of Three Phases of
"A Complete Process of C reative Problem Solving" for Different Fields of Endeavor

In Manufacturing In Applied Research In Pure Research

(Requiring Relatively More (Requiring Relatively More (Requiring Relatively
Solution Implementation) Problem Solving) More Problem Finding)

Preference Preference Preference Preference Preference Preference

for for for for for for
Evaluation Ideation Evaluation Ideation Evaluation Ideation
Problem
Finding I I I *
(Phase I*) I
I x� I x:-1 I
E I
E
\ J
E \ 1 /
Optimum Optimum Optimum

Problem
Solving � -x 1 1 x --{
(Phase I I *) J . � -x� J
E
"a ptimum I E "" .
I E
/ I
Optimum Optimum

Solution
I mplementation L _
"'
I I X 1
(Phase I l l *) 11
I /
rx I ' x "'
E I E
Optimum O ptimu m
Opt im um
I
0-
():J
 66

Figu re 1 4) and whether these ratios would differ for various fields of endeavor (see Figures

1 3 and 1 5) . The movement of various samples of trai nees from extreme positions of E or

I toward some optimal center 0 should be fully explored .

Research has already supported the usefulness of such future exploration. Basadur,

Wakabayashi and Graen ( 1 990) showed differential impacts of training on individua ls

whose d iffering creative probl em solving styles reflect varying preferences for ideation­

evaluation. Such different styles may relate to field of work. Kirton ( 1 987) found that

departments more concerned with implementation (such as production) have more adaptive

creative styles. One could speculate that they favor problem-solving activity si nce they prefer

to address problems by staying within given paradigms (problem definitions) . Departments

that focus on finding new long-term opportunities (such as research and development) have

more innovative creative styles. It could be said that they favor problem-finding activity,

si nce they prefer to deal with problems . by breaking (redefining) given paradigms.

I nterestingly, and in support of this line of speculation, engineering departments -- which

work with R&D in translating new concepts (problems) into new designs (solutions) a nd work

with production in translating new designs (solutions) into manufactured products

(implementation) - had mean KAI scores roughly halfway between those of production and

R&D . This supports the usefulness of exploring the concept of optimum l/E ratios, and also

indicates it may be possible to measure these ratios.

I n Table 1 8, research data on relatively sma l l base sizes showing l/E ratios for various

fields of endeavor tests this theory. The respondents were extracted from the various

samples used in this report. For some occupations, the base sizes were too sma l l to report

and wil l be supplemented in future work. Table 1 8 includes a col umn that predicts, based
 67

on judgment and experience, which of the three phases (problem fi nding or solving or

solution i mplementing) would typify the work of each field of endeavor.

Table 1 8

Mean l/E Ratios by Field of E ndeavor

Expected Means (SD)
Field of E ndeavor E mphasis Base �
Professor/Researcher P.F. 23 1 .5 1 (.40)
Training & Development P.F. 71 1 .34 (.48)
Marketing P.F. 48 1 .20 (.47)
Market Research Associate/Mgr P.F. 13 1.15 (.42)
Person nel/H uman Resources P.F. 41 1.14 (.44)

Quality Control/Assurance P.S. 36 1 .04 (.46)

Mfg. Development Engineering P.S. 1 01 1 .0 1 (.42)
Gov/t Administration P.S. 14 .97 (.22)

Manufacturi ng Production S.I . 259 .95 (.30)

Sales/Accou nt Manager S. I . 1 80 .95 (.27)
C ustomer Relations S.I. 28 .94 (.32)
Manufacturi ng Mai ntenance S.I. 38 . 94 (. 23)
Finance S. I . 64 .91 (.27)
Social/Health Services Worker S.I . 20 .90 (.20)
Distribution/Logistics/Warehousi ng S.I. 59 .90 (.22)
Secretarial/Admin. Support S.I. 50 .89 ( . 24)
Accounti ng S.I. 53 .88 (.24)
Market Research Tabulation S.I. 8 .80 (. 1 9)

D iscussion

The data in Table 1 8 suggest that l/E ratios can be meaningfully used to describe the

relative effect of various fields of endeavor on different phases of a complete process of

creative problem solving. I n virtual ly every case, the predicted emphasis was consistent with

the relative l/E ratio.

The short time horizons of accounting, manufacturing, production and

distri bution/logistics/warehousing range below . 90 (. 78 to .89), while longer ti me horizons

 68

of endeavors that emphasize understanding of needs and problems (research, people

development, marketing) range above 1 . 1 0 ( 1 . 1 7 to 1 . 5 1 ). The other fields range between

. 90 and 1 . 1 0 and are consistent with medium time horizons of solving problems, helping

others solve problems, and incremental ly improvi ng products and processes.

5. Predictive Validity: Preferred C PSP Styles by Field of E ndeavor

The reasoning on opti mal l/E ratios above can easi ly be extended to predict the

preferred C PSP styles of people in various fields of endeavor. Table 1 9 shows mean

quadrant scores of individuals in various fields of endeavor. The fields of endeavor are

grouped in descending order by dominant q uadrant score. By inspection, a l l of the

groupings make reasonable sense. For example, the q uadrant 4 fields a l l emphasize short­

term implementation work: manufacturing, sales, secretarial. The q uadrant 3 fields,

i ncluding engineeri ng, finance, accounting, systems development, quality assurance and

progra mmer/analyst, all i nvolve solving problems with precision and eval uating a nd

optimizing procedures. I n ma rket research and information systems jobs requiring

consultation with clients to define problems accurately, quadrant 2 would tend to dominate,

as supported by Table 1 9 . Quadrant 1 dominance is represented by fields such as

marketing, advertising account management, professor/researcher, teaching and training

and development, which require people to generate new big ideas, projects, opportunities

and problems to research further or to stimulate others' interest. Such fields as counseling

and personnel/human resources require the ability to operate with ambiguity, helping

people in u nstructured situations in which the problem is far from defined.

 69

Table 1 9

Mean Quadra nt Scores by Field of E ndeavor

Mean Quadra nt Scores

Domi n a nt
Quadra nt Base Field of E ndeavor Ql .Q1 Q3 ill.
1 8 Counselor/Psychologist/Psychiatrist/Minister 1 279 81 2 592 908
1 18 Teacher (Elementary, High School) 1 1 82 963 660 772
1 23 Professor/Researcher 1 1 77 937 664 805
1 71 Traini ng & Development 1 01 8 976 786 803
1 48 Marketing 983 908 826 872
1 77 Non-Profit Org ./Univ. Admin. 967 849 830 930
1 41 Personnel/H u man Resources 936 912 858 879
1 29 Advertising Agency Account Mgr. 918 908 882 87 1

2 9 Market Research/Research Assoc. 900 959 89 1 83 1

2 45 Senior Systems Consultant 882 955 900 846
2 4 Market Research Senior Mgr. 94 1 949 857 845

3 27 Engineering/Engineering Design 772 89 1 1 029 894

3 64 Finance 805 862 997 919
3 53 Accounting 806 84 1 988 949
3 23 R&D/Basic Research 809 915 986 868
3 54 Programmer/Analyst 830 882 967 902
3 24 Mgr./Di r. Systems Development 856 930 934 866
3 1 01 Mfg. Development Eng. 862 870 933 919
3 14 Government Administration 866 873 921 91 7
3 36 Quality Control/Assurance 872 889 920 900

4 23 General Management - Small Co. or Div. 954 758 832 1 038

4 8 Market Research Tabulator 805 768 99 1 1 023
4 59 Logistics/Distri bution/Warehousing 893 775 889 1 02 1
4 27 Purchasing 832 772 950 1 020
4 1 80 Sales 927 781 859 1 01 5
4 20 Social/Health Services Case Worker 878 798 897 1 000
4 50 Secretarial/Admin. Support 866 79 1 924 1 000
4 25 Technical Customer Service 82 1 789 975 998
4 28 Customers Relations 888 806 901 983
4 259 Manufacturing Production 890 81 5 892 983
4 11 N u rse 929 820 853 968
4 38 Mfg. Maintenance 859 854 933 94 1
 70

6. Predi ctive Validity: Mean C PSP Quadra nt (Style) Scores by Organ izational Level

of Responsibi l ity for Thi n ki ng Strategica l ly

I n Table 20, mean quadrant scores are tabulated for a large sample (n= 1 639) by

respondents' level of organizational responsibili1y for strategic thinking rather than

operational thin king. The i ncreasing levels are defined as non-management, fi rst-line

s upervision, middle management, upper management and technical/professional . The

higher in the hierarchy, the less important it becomes to implement and optimize day-to-day

operational tasks and current methods, products and services, and the more important to

generate new di rections to pursue, to think strategically about the future, to conceptualize

the "big picture" and to create appropriate goals for others to achieve. The category

selected as the highest i n the hierarchy includes "non-hierarchical" jobs for which highly

specialized technical and professional people, including economists, scientists and planners,

are paid pri mari ly to thin k and to help senior management to thi nk strategical ly.

Accordi ng to this logic, as responsibilities increase, the mean scores for the generator

and conceptualizer q uadrants should i ncrease and those for the implementer and optimizer

should decrease. The data in Table 20 and Figure 1 6 support this prediction, especially for

conceptualizing and i mplementing scores.

Ta ble 2 1 and Figure 1 7 show that l/E ratios increase in a similar fashion as

organizational levels gain more responsibili1y for strategic thinki ng rather than operational

thinking.
 71

Table 2 0

Mean Quadra nt Scores by I ncreasi ng Level of

Organ izational Responsibilify for Strategic Thi n ki ng
Rather than Operational Thi n ki ng

Mean Quadra nt Scores

Organizational
Level Base Generator Conceptua I izer Optimizer .lr!m.lementer

N on-
management 1 91 888 808 896 989

First Line
Supervision 494 893 81 3 897 980

Middle
Management 232 899 843 893 946

U pper
Management 84 898 875 895 913

Technical/
Professional 474 929 896 873 884

Table 2 1

Mean l/E Ratios by I n creasi ng Level of

O rgan izational Responsibil ify for Strategic Thinking
Rather tha n Operational Thi nki ng

Means (SO}
O rgan izational Level Base lLE Ratios

Non-management 1 91 .94 (.28)

First Line Supervision 494 .95 (. 29)

Middle Management 232 1 .00 (.36)

U pper Management 84 1 .04 (.40)

Technical/Professional 474 1 .1 1 (.43)

 Figure 1 6

Mean Quad rant Scores by Increasing Level of Organ izational Responsibility

for Strategic Thinki ng Rather than Operationa l Thinki ng
Mean
Quadrant
Scores

1 OOO I mpleme ntin g

• • • • • • • • • • •
• • • • •
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•

Optimizing
900 • • • • • • • • • • • • • • • • • •
•
• • • • • • • • • • • • • • • • • • • • • • • •
• •
• • •
Generating • •
.,... .,... � r:• .,... ;.I
• ..-.--
• •
• •
--
--
--
--
--
--
--
--
,,,,.
--
--
--
,,,,.
--
--
,,,,.
Co ncept ua I izi ng --
--
--
------ - -�1--�����-��-+-��+-�
8 QQ 1-

Non-management First Li ne Middle Upper Tech nical/

Supervision Management Management Professional

"""
t-...)
Level of Organizational Responsibility for Thinking Strategically Rather than Operationa l ly
 Fig ure 1 7

Mean l/E Ratios by I ncreasing Level of Organ izational Responsibility

for Strategic Thi n ki ng Rather than Operational Thi n king
Mean
l/E
Ratio

i .2 IT----1---+---�---_J

1 .0 r-�r-�����-1-��-::--::t;�Oll-�:;;ootl�===-��+-���_J

.80 1--�1--�����-��-+��+-���-l

Non-management Fi rst Line Middle Upper Technica l/

Su pervision Management Management Professional
'-I
w
Level of Organ izational Responsibility for Thinking Strategica l ly Rather than Operationally
 74

SUMMARY

This paper presents evidence of adequate psychometric properties for the CPSP but

there is considerable room for reliabil ity improvement. The validity evidence in support of

the CPSP has been presented from many directions and indicates quite strongly that the

i nstrument measures what it says it measures and can predict accurately.

The underlying theory and early development and internal consistency testing of the

Basadur CPSP (Creative Problem Solving Profile) inventory is reviewed, then new expanded

rel iability and validity research is summarized. The four columns, the four quadrants, a nd

the two bi-pola r scales of the CPSP demonstrated satisfactory reliability via test-retest,

i ndependence of scales, and para l lel split-half correlations.

The CPSP correlated with the Kirton Adaptation I nventory (KAI) and the Myers-Briggs

Type I ndicator (MBTI) in predi cted ways. Also, the relative impact of creativity training was

moderated according to CPSP style in expected ways. Assessments of CPSP style by self

a lone and i n consensus with expert partner significantly agreed with the assessments made

by the inventory itself. Predictions of field of endeavor (occupation) and of level of

o rganizational responsibility for strategic thinking by preferred CPSP style and preferred ratio

of ideation-evaluation polar opposites were strongly supported. Opportunities for improving

psychometrics, especially reliabi l ity, a re identified.

 75

NOTE

1 . MBTI is a registered trademark of Consulting Psychologists Press, Inc., Palo Alto California
which publishes a users manual for the instrument.

REFERENCES

Ackoff, R. L. ( 1 9 79). The future of operational research is past. Journal of Operational Research
Society, 30, 9 3 - 1 04 .

Basad u r, M.S. ( 1 997) . Org anizational development interventions for enhancing creativity in the
workplace. Journal of Creative Behavior, 3 1 (1 ), 59-7 2 .

Basadur, M.S. ( 1 995). The Power of Innovation. London, UK: Pitman Professional Publishing .

Basadur, M.S. ( 1 994) . Managing the creative process in organizations. In Problem Finding, Problem
Solving, and Creativity. (Editor: Runco, M.J .), New York: Abl ex.

Basadur, M.S. ( 1 993). I m pacts a nd outcomes of creativity in organizational settings. In The

Emergence of a Discipline: Nurturing and Developing Creativity, Vol ume I I . (Editors: Isaksen,
S.G., Murdock, M.C., Firestein, R. L., and Treffinger, D.J.), New York: Ablex, December.

Basadur, M.S. ( 1 992) . Managing Creativity: A Japa nese model. Academy of Management
Executive, 6 (2), 29-42.

Basadur, M.S. ( l 988a). Improving union-management bargaining using a special process of applied
creative thinki ng . McMaster University Faculty of Business Research and Working Paper
Series, No. 300, Hamilton, Ontario, Canada, L8S $M4.

Basad ur, M.S. ( 1 988b) . The new creative thinking skills today's purchasing professionals must have
to be successful . The Cincinnati Purchaser, Cincinnati, OH, P.O. Box 39376, 45230-0376,
November/December.

Basad ur, M.S. ( 1 987) . In Frontiers of Creativity Research : Beyond the Basics. (Editor: Isaksen, S.G.,
Chapter 1 8.) Buffalo, N.Y. : Beerly.

Basadur, M.S. ( 1 983). Employee involvement creative problem-solving workshop. In Ford Education
a nd Training Catalog, Ford Education a nd Personnel Resea rch Department, Dearborn, Ml,
September, 1 1 5 .

Basadur, M.S. ( 1 982 ) . Research in creative problem solving training in business a n d industry.
Proceedings of Creativity Week 4 . Greensboro, NC: Center for Creative Leadership.

Basadur, M.S. ( 1 9 8 1 ). Training in creative problem solving and measuring i mprovement.

Engineering Digest, 2 7 (3), Toronto, Ontario.
 76

Basad ur, M.S. ( 1 979) . Training in creative problem solving : Effects on deferred judgment and
p roblem finding and solving in a n industria l research organization. Doctoral Dissertation,
University of Cincinnati, December.

Basadur, M.S. ( 1 97 4). Think or sink. The Deli berate Methods Change Bulletin, J uly-September,
Procter & Gamble Management Systems Division, Cincinnati, Ohio.

Basadur, M.S., Ellspermann, S.J ., a nd Evans, G.W. ( 1 994) . A new methodology for formulating ill­
structured problems. OMEGA: The International Journal of Management Science, 2 2 , 627-
645 .

Basadur, M.S., & Finkbeiner, C.T. ( 1 985) . Measuring preference for ideation i n creative p roblem
solving training. Journal of Applied Behaviora l Science, 2 1 ( 1 ), 37-49.

Basadur, M.S., Graen, G.B., & Green, S.G. ( 1 982 ) . Training i n creative problem solving: Effects on
ideation and problem finding in a n a pplied research orga nization. Organizationa l Behavior
a nd Human Performance, 30, 4 1 -70.

Basad ur, M.S., Graen, G.B., & Scandura, T.A. ( 1 986) . Training effects on attitudes toward divergent
thinking a mong manufacturing engineers. Journal of Applied Psychology, 7 1 (4), 6 1 2 -6 1 7 .

Basadur, M.S., Graen, G . B., & Wakabayashi, M. ( 1 990). Identifying i ndividual differences i n creative
problem solving style. Journal of Creative Behavior, 24 (2) .

Basadur, M.S., Wakabayashi, M., & Graen, G.B. ( 1 990). Attitudes towards divergent thinking before
a nd after training : Focusing upon the effect of individ ua l problem solving styles. Creativity
Research Journa l, 3 (1 ) , 2 2-32 .

Carlsson, B., Keane, P., & Martin, J . B. ( 1 9 76). R&D organization as learning systems. Sloan
Management Review, Spring.

Cooper, R.G. ( 1 993). Winning at New Products: Accelerating the Process from Idea to Launch.
Reading, MA: Perseus Books.

Getzels, J .W. ( 1 9 75). Problem finding a nd the i nventiveness of solutions. The Journal of Creative
Behavior, 9 ( 1 ) .

Gordon, W.J .J. ( 1 9 7 1 ) The Metaphorical Way. Cambridge, Massachusetts: Porpoise Books.
.

Gordon, W.J .J . ( 1 956). Operational a pproach to creativity. Harvard Business Review, 9 ( 1 ) .

H icks, L E . ( 1 985). Dichotomies a n d topologies: Summaries and implications. Journal of

Psychological Type, 1 0, 1 1 - 1 3 .

H icks, L E . ( 1 984) . Conceptual and empirical analysis of some assumptions of a n explicitly

topological theory. J ournal of Personality a nd Social Psychology, 46, 1 1 1 8- 1 1 3 1 .

Kirton, M.J . ( 1 987). KAI Manual. Occupational Research Centre, P.O. Box 1 09, Hatfield, Hertsford,
AL 1 0 9AB, U . K.
 77

Kirton, M.J. ( 1 9 76) . Adaptors and innovators : A description and measure. Journal of Applied
Psychology, 6 1 , 622-629.

Leavitt, H.J. ( 1 9 75) . Beyond the analytic manager. California Management Review.

Levitt, T. ( 1 963). Creativity is not enoug h . Harvard Business Review.

Livingston, l.S. ( 1 9 7 1 ) . Myth of the wel l -educated manager. Harvard Business Review, Jan-Feb, 79-
89.

Mackworth, N.H. ( 1 965). The American Psychologist. 20, 5 1 .

McCrea, R.R., & Costa, P.T., J r. ( 1 989). Reinterpreting the Mye rs-Briggs Type I nventory from the
perspective of the five-factor model of personality. Journal of Personality, 57, 1 7-40.

Mintzberg, H . ( 1 989). Mintzberg on Management: I nside Our Strange World of Organizations.

Monroe, LA: The Free Press.

Mott, P.E. ( 1 9 72 ) . The characteristics of effective org anizations. New York: Harper and Row.

Myers, l . B. ( 1 9 94). I ntroduction to Type. (5th ed ., European English Version.) Oxford : Oxford
Psychologists Press.

Myers, l . B . , & McCaulley, M.H. ( 1 985). Manual : A Guide to the Development and Use of the Myers­
Briggs Type I ndicator (2"d ed.) . Palo Alto, CA: Consulting Psycholog ists Press.

Myers, P.B., & Myers, K.D . ( 1 984) . MBTI Form G. Palo Alto, CA: Consulting Psychologists Press, I nc.

Norton, J . ( 1 9 65) . The Dot a nd the Line (Film). Wilmett, IL: Fil ms Incorporated, 1 1 44 Wil mett Ave.

Osborn, A.F. ( 1 963). Applied I magination. New York, NY: Charles Scribner's Sons.

Parnes, S.J., Noller, R. B., & Biondi, A.M. ( 1 977). Guide to Creative Action. New York, NY: Charles
Scribner's Sons.

Runco, M.A. a nd Basadur, M.S. ( 1 993). Assessing ideational and evaluative skills a nd creative styles
a nd attitudes. Creativity a nd I nnovation Management, 2 (3), 1 66 - 1 73.

Senge, P., Kleiner, A., Roberts, C., Ross, R., & Smith, B. ( 1 994) . The Fifth Discipline Fieldbook. New
York: Doubleday.

Simon, H .A. ( 1 977). The New Science of Management Decisions. Englewood Cliffs, NJ : Prentice­
Hal l .

Simon, H .A. ( 1 960) . The New Science o f Management Decision. Englewood Cliffs, NJ : Prentice­
Ha l l .

Stricker, L.J . , & Ross, J . ( 1 964). A n assessment of some structural properties o f the Jungian
personality topology. Journal of Abnormal a nd Social Psychology, 68, 62-7 1 .
 78

SAMPLE D ESCRI PTI ONS

This paper has reported data from various participant samples. Following is a description

of each of these samples. The data were usually gathered in the field d u ri ng training,

development and application activities with organizations seeki ng to improve understanding

of the creative problem-solvi ng process and organizational performance.

Sample D escription

n = 1 07 This consisted of managers, supervisors and professionals from a large

consumer goods company and a large public works facility, university

g raduate students in business (M. B.A.), and a small group of people

with miscellaneous managerial and professional positions from four

different organizations.

n = 1 49 /1 3 8/1 1 0 This consisted of managers, supervisors and professionals from a large

consumer goods company, first-line supervisors and foremen from a

large industrial company, and professionals, lead hands and first-l ine

supervisors from a government mi nistry of mi nes and natural resources.

Sample sizes varied because varyi ng logistical constraints meant that

not a l l partici pants fil led out all questionnaires.

n = l Ol This consisted of engineers, engineering managers and engineering

technicians in product and process development in a l a rge ti re

manufacturer.
 79

n = 1 29 This consisted of members of a wide variety of business and industrial

organizations and job types, including manufacturing and headquarters

managers, supervisors, professiona ls, and clerical and shop floor

employees from a large consumer goods company; marketing and

salespeople from a mid-sized outdoor advertising company; and the top

management team from a mid-sized consumer goods company.

n = l 56 This consisted of partici pants representing a variety of fu nctional

specialties and a variety of organizations. The partici pants were

classified as managers and non-managers. The managers were middle

managers and di rectors (top management) and the non-managers were

engineers, first- line supervisors and technical specialists drawn from the

same organizations as the managers.

n = 40 This was a small group of partici pants representing a variety of job types

and organizations.

n = 1 67 This was a mixture of members of a variety of business and industria l

organizations includi ng managers, supervisors, professionals and

clerical employees from manufacturing, marketing, sales, information

systems and other miscellaneous kinds of jobs.

n = 1 639 This sample combi nes all the above samples (n =849) plus several

additional samples (n= 790) .

n = 1 85 Similar to and including the n = 1 1 0 sample.

 80
APPENDIX

The Appendices referred to throughout the paper a re exhibited in this section in

numerical order.

The items i n the columns for the correlational analyses in Appendices I through XX a re
n umbered as shown below.

x I T E
Column 1 Column 2 Column 3 Column 4

1. Alert 1 9. Poised 37. Ready 55. Eager

2. Patient 20. Diligent 38. Forceful 56. Prepared

3. Doing 21 . Intuitive 39. Detached 57. Selective

4. Experiencing 22. Optimistic 40. Objective 58. Verifying

5. Reserved 23. Serious 41. Fun-loving 59. P layful

6. Sensing 24. Free Thinking 42. Logical 60. Experimenting

7. Feel ing 25. Alternatives 43. Analyzing 61 . Evaluating

8. Action 26. Divergence 44. Abstract 62. Convergence

9. Direct 27. Possibilities 45. Conceptual 63. Practicalities

1 0. Quiet 28. Trustworthy 46. Irresponsible 64. Imaginative

11. Involved 29. Proliferating 47. Theoretical 65. Testing

1 2. Probing 30. Projecting 48. Structuring 66. Exami ning

1 3. I mmediate 31 . Gathering 49. Understanding 67. Confirming

1 4. Impersonal 32. Proud 50. Hopeful 68. Fearful

1 5. Implementing 33. Visualizing 51 . Modell ing 69. Decisive

1 6. Present-oriented 34. Future-oriented 52. Rational 70. Detail-oriented

1 7. Sympathetic 35. Pragmatic 53. Emotional 71 . Procrastinating

1 8. Aware 36. Childlike 54. Orderly 72. Real istic

 APPENDIX I

Corre l at i o n Matiix - Inter- I t em Corre l at i ons

Column 1 (X)

Non- Ips ative S c al e , n=1 4 9

I T E M

I tem 3 4 6 7 8 9 11 12 13 15 16 18

3 1. 0000
4 . 2894 1. 0000
6 . 0618 . 2834 1. 0000
7 . 0967 . 3204 . 5557 1. 0000
8 . 4843 . 19 6 9 . 13 69 . 1294 1. 0000
9 . 2927 . 1167 . 0600 . 0177 . 4549 1. 0000
11 . 1848 . 1153 . 14 9 2 . 0665 . 3489 . 3 17 1 1. 0000
12 . 07 7 6 . 2186 . 2865 . 1128 . 2205 . 2528 . 4489 1. 0000
13 . 4113 . 2 2 64 . 18 15 . 0858 . 5322 . 3773 . 3073 . 2281 1. 0000
15 . 3439 . 19 7 0 . 15 1 6 . 0169 . 5125 . 2905 . 3263 . 1627 . 4 099 1 . 0000
16 . 15 3 8 . 17 3 1 . 2547 . 15 3 0 . 2433 . 22 67 . 18 6 7 . 2 184 . 3 132 . 19 2 5 1 . 0000
18 . 0038 . 18 2 4 . 2 168 . 18 2 1 . 2250 . 2496 . 3 19 4 . 2479 . 14 6 7 . 2 2 05 . 1900 1 . 0000

co
--'
 APPENDIX I I

Corre l at i o n Matrix - Inter- I tem Corre l ations

Column 2 (I)

Non-Ipsat ive S c a l e , n= l 4 9

Item

I tem 21 22 24 25 26 27 29 30 31 33 34 36

21 1 . 0000
22 . 2254 1. 0000
24 . 17 3 4 . 2 8 57 1. 0000
25 . 2028 . 2625 . 4356 1. 0000
26 . 0966 . 19 7 0 . 4 050 . 5 0 12 1. 0000
27 . 14 8 9 . 3578 . 3968 . 4525 . 53 13 1. 0000
29 . 3500 . 0671 . 4351 . 2 2 65 . 3 6 07 . 3 105 1. 0000
30 . 2849 . 1610 . 2406 . 2280 . 18 8 9 . 2437 . 4051 1. 0000
31 . 0640 . 1677 . 17 3 2 . 2616 . 34 56 . 2371 . 2388 . 3474 1. 0000
33 . 3 155 . 3 3 98 . 3410 . 3 118 . 2 6 19 . 3228 . 3434 . 4290 . 1389 1 . 0000
34 . 2 07 5 . 2776 . 3 528 . 2989 . 3170 . 3 814 . 2950 . 2431 . 2 6 14 . 4831 1 . 0000
36 . 2564 . 2075 . 2454 . 18 4 7 . 17 5 9 . 09 16 . 2 658 . 2 187 . 12 19 . 2 113 . 174 1 1 . 0000

co
t-..)
 APPENDIX I I I

Corre l at i on Matr ix - Inter- I tem corre l ations

Column 3 (T)

Non- Ips ative S c al e , n= 1 4 9

Item

I tem 39 40 42 43 44 45 47 48 49 51 52 54

39 1. 0000
40 . 2830 1. 0000
42 . 16 6 9 . 2898 1 . 0000
43 . 1808 . 2363 . 6753 1. 0000
44 . 2329 . 06 3 1 - . 1497 . 0067 1. 0000
45 . 1211 . 2 14 3 . 13 2 3 . 2469 . 4992 1. 0000
47 . 03 4 1 . 15 9 2 . 17 2 0 . 3 042 . 4455 . 5001 1. 0000
48 . 0986 . 2 3 04 . 2759 . 3877 . 08 19 . 2 155 . 2230 1. 0000
49 . 0069 . 19 4 7 . 16 2 6 . 3 14 3 . 08 0 2 . 0952 . 0975 . 2504 1. 0000
51 . 02 8 3 . 08 9 8 - . 0088 . 14 7 9 . 3 129 . 3528 . 2935 . 1259 . 2744 1 . 0000
52 . 1083 . 24 7 5 . 4 3 15 . 3834 . 0016 . 2737 . 19 7 5 . 3 57 0 . 3 4 50 . 2638 1 . 0000
54 . 1783 . 3209 . 2468 . 3 068 -. 0 2 18 . 1562 . 03 7 5 . 2439 . 1893 . 14 2 6 . 3825 1 . 0000

co
w
 APPENDIX IV

Corre l ation Matrix - Inter - I tem corre l ations

Column 4 (E)

Non- Ipsat iv e s c al e , n= 1 4 9

I tem

Item 57 50 62 61 62 63 65 66 67 69 70 72

57 1 . 0000
58 . 3820 1 . 0000
60 . 0048 . 0668 1 . 0000
61 . 3328 . 5101 . 13 4 8 1 . 0000
62 . 3032 . 3 637 - . 0055 . 3800 1. 0000
63 . 2791 . 3 8 67 . 0104 . 2775 . 3945 1. 0000
65 . 2 6 17 . 2661 . 4513 . 2788 . 0948 . 19 17 1. 0000
66 . 3 140 . 4676 . 2753 . 4869 . 3206 . 3448 . 6064 1. 0000
67 . 3261 . 5111 . 0839 . 4598 . 3132 . 3258 . 3725 . 5 194 1. 0000
69 . 15 3 5 . 3 2 52 . 12 8 9 . 4051 . 2 09 1 . 14 15 . 1768 . 2971 . 27 19 1 . 0000
70 . 2975 . 4481 . 0651 . 4670 . 2 625 . 1534 . 14 5 0 . 4042 . 3896 . 14 7 2 1 . 0000
72 . 2989 . 3591 . 03 5 1 . 3477 . 2 63 1 . 4903 . 19 8 0 . 2863 . 3370 . 2 04 3 . 2 158 1. 0000

co
.i::...
 APPEND I X V

Re l i abi l i ty Analys i s

Column 1 ( X)

I t em Corre l ations Wi th Rest o f s c a l e and c ronbach Alpha

Non- Ip s ative S c al e , n= l 4 9

Scale scale . Corrected

Mean Var i ance I t em- Squared Alpha
I f I t em I f I t em Total Mul t i p l e I f I tem
I t em Deleted Deleted Correlation Corre lation Deleted

3 29 . 4899 3 0 . 0759 . 3953 . 3 3 68 . 7673

4 2 9 . 87 2 5 29 . 8553 . 3862 . 2271 . 7684
6 3 0 . 07 3 8 29 . 6229 . 3945 . 4014 . 7 676
7 2 9 . 9 0 60 3 0 . 5857 . 2809 . 3796 . 7804
8 29 . 3691 28 . 8426 . 5881 . 4974 . 7492
9 29 . 6376 29 . 1786 . 4 3 17 . 2891 . 7636
11 2 9 . 3 9 60 3 0 . 1597 . 4565 . 3288 . 7622
12 29 . 8322 29 . 5054 . 4039 . 2935 . 7 666
13 3 0 . 0134 2 7 . 8 106 . 5372 . 3879 . 7512
15 2 9 . 7 5 17 2 9 . 7 69 0 . 4556 . 3523 . 7616
16 3 0 . 1342 29 . 5900 . 3825 . 1672 . 7691
18 29 . 6174 3 0 . 8729 . 3561 . 2084 . 7 7 10

Re l i ab i l ity C o e f f i c i en t s - 12 I tems

Alpha = . 7803

co
U'I
 APPENDIX VI

Rel i ab i l i ty Analy s i s

Column 2 (I)

I t em Corre l at i ons With Re st o f S c a l e and cronbach Alpha

Non- Ips ative S c al e , n= l 4 9

Scale S cale Corrected

Mean Vari ance I tem- squared Alpha
I f I t em I f I t em Total Mul tiple If Item
I tem Del eted Del eted Corre l at i o n Correlation Del eted

21 2 6 . 7203 31. 0902 . 3539 . 2325 . 8 153

22 2 6 . 4895 31. 0826 . 3938 . 2656 . 8 1 07
24 2 6 . 97 2 0 29 . 2809 . 54 3 8 . 3770 . 7969
25 2 6 . 7 692 30. 4 182 . 5292 . 3777 . 7989
26 27 . 2937 30. 3920 . 5248 . 4424 . 7992
27 2 6 . 8531 30. 4 2 19 . 54 9 1 . 4235 . 7975
29 2 7 . 3 3 57 30. 7 4 57 . 52 1 1 . 4060 . 8 000
30 27 . 0000 30. 2 67 6 . 4631 . 3397 . 8 04 6
31 2 6 . 9231 31. 7616 . 3 608 . 2461 . 8 129
33 2 6 . 5 664 30. 1769 . 5587 . 4115 . 7965
34 2 6 . 6084 29. 9301 . 5 157 . 3339 . 7996
36 2 8 . 0070 32 . 6408 . 3329 . 1569 . 8 14 1

Re l i ab i l ity C oe f f i c i ents -. 12 I tems

Al pha = . 8 174

OJ
°'

..
 APPENDI X VI I

Re l i abi l i ty Analys i s

column 3 (T)

I tem Corre l at i ons With Rest o f s c a l e and Cronbach Alpha

Non- Ips at ive s c al e , n= 1 4 9

s c al e Scale Corrected
Mean vari ance I tem- squared Alpha
I f Item I f I tem Total Mul t i p l e I f I tem
I tem Del eted Del eted Correlation Corre l at i o n Del eted

39 27 . 0940 2 6 . 0587 . 2401 . 19 3 4 . 7 627

40 25 . 7987 24 . 3781 . 3902 . 2378 . 7486
42 2 5 . 5168 2 4 . 3 055 . 4002 . 5509 . 7475
43 25 . 6711 22 . 9925 . 5514 . 5719 . 7297
44 2 6 . 8322 2 5 . 0865 . 2690 . 4398 . 7626
45 2 6 . 0805 2 2 . 9 12 4 . 4978 . 4255 . 7353
47 2 6 . 6913 23 . 8095 . 4379 . 3860 . 7431
48 2 6 . 2.2 1 5 2 4 . 052 0 . 4248 . 2 3 10 . 7447
49 2 5 . 7 3 15 2 5 . 04 9 1 . 3425 . 2 4 18 . 7535
51 2 6 . 3 69 1 2 4 . 1534 . 3492 . 2496 . 7541
52 25 . 7047 2 3 . 54 7 3 . 5199 . 3954 . 7345
54 25 . 9732 24 . 3641 . 3639 . 2449 . 7517

Re l i ab i l ity C o e f f i c ients - 12 items

A l ph a = . 7637

co
-...../
 APPENDIX V I I I

Re l i abi l i ty Analys i s

Column 4 (E)

I tem Corre l at i ons With Rest o f S c a l e and cronbach Alpha

Non- Ipsative S c al e , n=1 4 9

S cal e Scale Corrected

Mean Vari ance I t em- Squared Alpha
If I tem I f It em Total Mul t i p l e If Item
I tem Deleted Del eted Corre l ation Corre l at i o n Deleted

57 27 . 6216 33 . 4885 . 4488 . 2487 . 8 162

58 27 . 5 6 08 30 . 9827 . 6390 . 4562 . 7998
60 27 . 5743 35. 593 1 . 1859 . 2350 . 8 377
61 27 . 2635 32 . 3995 . 64 17 . 4626 . 8028
62 27 . 5811 33 . 4832 . 4377 . 2806 . 8171
63 27 . 4 4 59 33 . 1467 . . 4487 . 3 6 59 . 8163
65 27 . 6149 32 . 8507 . 4 663 . 4992 . 8 149
66 27 . 5000 30. 8231 . 6842 . 5688 . 7962
67 27 . 4797 31. 54 3 8 . 6103 . 4 14 6 . 8 027
69 27 . 1892 33 . 77 3 5 . 3 665 . 2104 . 8232
70 27 . 4797 32 . 0744 . 4536 . 3448 . 8 170
72 26 . 9459 33 . 6025 . 4598 . 3 152 . 8155

Re l i ab i l ity Coe f f i c i ents - 12 I tems

Alpha = . 82 65

co
co

•
 APPENDIX I X

Correl at i on Matrix - Inter- Item Corre lations

Column 1 (X)

Non- Ips ative s c al e , n= 1 0 7

Item

Item 3 4 6 7 8 9 11 12 13 15 16 18

3 1. 0000
4 . 3634 1. 0000
6 . 1026 . 3943 1. 0000
7 . 17 7 4 . 3 067 . 5665 1. 0000
8 . 50 3 3 . 3769 . 0145 . 1665 1. 0000
9 . 32 1 1 . 02 2 5 -. 0357 . 09 0 2 . 3 8 12 1. 0000
11 . 2329 . 2336 . 17 9 9 . 19 6 0 . 2615 . 3581 1. 0000
12 . 16 3 2 -. 0063 . 2301 . 2551 -. 0345 . 13 7 5 . 2 165 1. 0000
13 . 4708 . 13 6 1 -. 0384 . 1197 . 3762 . 3338 . 3 589 . 13 0 1 1. 0000
15 . 3830 . 2191 . 0159 . 18 4 4 . 4 104 . 2741 . 2 3 54 . 03 4 2 . 3595 1 . 0000
16 . 14 2 6 . 05 0 3 . 0027 . 2 186 . 19 2 2 . 1392 . 07 0 6 -. 0170 . 2881 . 2705 1 . 0000
18 . 157 6 . 13 0 6 . 3 187 . 2434 . 2 3 53 . 2904 . 3613 . 2480 . 17 6 4 . 1644 . 14 7 3 1 . 0000

co
'-0
 APPENDI X X

Corre l at i on Matrix - Inter-I tem Co rre l ations

Column 2 (I)

Non- Ipsat ive S cal e , n= 1 0 7

Item 21 22 24 25 26 27 29 30 31 33 34 36

21 1 . 0000
22 . 3058 1 . 0000
24 . 10 3 3 . 1687 1. 0000
25 . 1647 . 3232 . 3 6 14 1. 0000
26 . 2695 . 2 6 18 . 28 7 3 . 2961 1. 0000
27 . 2964 . 2 554 . 27 8 3 . 5083 . 4990 1. 0000
29 . 12 3 8 . 0 64 5 . 17 0 1 . 2 192 . 3985 . 3 109 1. 0000
30 . 17 6 4 . 07 2 7 . 17 1 6 . 2299 . 3 198 . 2 603 . 3 056 1. 0000
31 . 1132 . 12 7 0 . 18 5 6 . 2638 . 2 189 . 3 590 . 17 7 7 . 2940 1. 0000
33 . 3770 . 2638 . 3538 . 47 10 . 3 14 6 . 4 14 4 . 2 184 . 2497 . 4370 1 . 0000
34 . 2 3 17 . 1613 . 2569 . 2907 . 3701 . 3895 . 2 557 . 3229 . 1901 . 3460 1 . 0000
36 . 2650 . 1400 . 12 65 . 2 14 4 . 2236 . 1951 . 1560 . 0884 . 1285 . 2538 . 2299 1 . 0000

'-0
0
 APPENDIX XI

Corr e l a t i o n Matrix - Inter- I t em Corr e l ations

Column 3 (T)

Non- Ipsative S cal e , n= 1 0 7

I tem

Item 39 40 42 43 44 45 47 48 49 51 52 54

39 1 . 0000
40 . 16 5 3 1 . 0000
42 . 1156 . 5699 1 . 0000
43 . 15 4 3 . 52 3 9 . 7077 1 . 0000
44 . 2 078 . 0602 - . 0377 . 1188 1. 0000
45 . 07 3 2 . 2 659 . 08 2 1 . 2225 . 4 2 57 1. 0000
47 . 18 6 1 . 4379 . 2022 . 3586 . 3022 . 4 567 1. 0000
48 . 15 8 5 . 3876 . 3209 . 3787 -. 0305 . 1674 . 2811 1. 0000
49 . 0000 . 3497 . 2704 . 3087 . 0000 . 2855 . 1206 . 2 3 59 1. 0000
51 . 0674 . 2680 . 0882 . 17 7 5 . 3244 . 3 127 . 2 19 0 . 3 165 . 2224 1 . 0000
52 . 2 097 . 5482 . 64 12 . 5488 -. 0561 . 0528 . 17 9 4 . 3439 . 27 3 9 . 1686 1 . 0000
54 . 0750 . 3528 . 43 2 0 . 4436 -. 1386 -. 0647 . 0976 . 3 8 16 . 0556 . 2 3 97 . 4 161 1 . 0000

'°
 APPENDIX X I I

Corre l at i on Matrix - Inter- Item corre l at i ons

Column 4 (E)

Non- Ips ative Scal e , n= l 0 7

I t em

Item 57 58 60 61 62 63 65 66 67 69 70 72

57 1. 0000
58 . 38 3 1 1 . 0000
60 . 16 3 9 - . 0150 1. 0000
61 . 3021 . 2752 -. 02 10 1. 0000
62 . 3020 . 17 10 -. 1 19 7 . 2 57 3 1 . 0000
63 . 107 3 . 1113 -. 0686 . 3 277 . 2372 1. 0000
65 . 3025 . 2666 . 4 163 . 1613 . 0371 -. 0648 1. 0000
66 . 2573 . 2989 . 1514 . 3 018 . 2 1 14 . 14 0 4 . 4306 1. 0000
67 . 2529 . 6098 -. 0810 . 29 04 . 2465 . 1571 . 2 3 62 . 2894 1. 0000
69 . 2293 . 2 7 17 . 06 5 1 . 3503 . .2 0 5 3 . 1678 . 2 085 . 2348 . 1661 1 . 0000
70 . 17 5 4 . 4 197 . 00 18 . 2990 . 12 5 2 . 14 5 0 . 2311 . 2837 . 4281 . 4265 1 . 0000
72 . 3 59 1 . 3 10 6 . 0887 . 3422 . 0296 . 3340 . 3 559 . 2383 . 4135 . 3 078 . 2821 1 . 0000

'°
"'
 APPENDIX XI I I

Rel i ab i l i ty Analys i s

Column 1 (X)

I t em corre l a t i ons With Rest o f s c al e and cronbach Alpha

Non- Ips ative Scal e , n= l 0 7

S cal e Scal e corre cted

Mean vari ance I t em- Alpha
I f Item I f I t em Total I f I t em
I tem Del eted Deleted Corre l at i on Deleted

3 40 . 0283 29 . 8 182 . 5285 . 7320

4 4 0 . 3 0 19 30 . 5937 . 3906 . 7471
6 4 0 . 57 5 5 31. 4 657 . 2992 . 7579
7 4 0 . 528 3 29. 4325 . 4351 . 7420
8 39 . 9906 29. 8571 . 4974 . 7350
9 39 . 9245 30. 5657 . 3821 . 7482
11 39 . 9151 30. 6689 . 4739 . 7388
12 4 0 . 2 64 2 32 . 5581 . 2252 . 7649
13 4 0 . 4906 29. 7380 . 4584 . 7389
15 4 0 . 3585 30. 4226 . 4361 . 7 4 19
16 40 . 3491 32 . 4 199 . 2540 . 7614
18 4 0 . 0283 31. 3992 . 4226 . 7445

A l pha = 0 . 7 6 2 5

'°
w
 APPENDI X XIV

Rel i abi l i ty Analys i s

Co lumn 2 (I)

Item Corre l at ions With Rest o f S c a l e and cronbach Alpha

Non- Ipsat ive Scal e , n= l 0 7

S cal e Scale corrected

Mean vari ance I t em- Alpha
If Item I f I tem Total I f I t em
I t em Deleted Del eted Corre l ation Deleted

21 37 . 3208 31. 4771 . 3850 . 7949

22 37 . 2358 31. 5343 . 3410 . 8005
24 37 . 5849 31. 3 1 18 . 3 9 19 . 7944
25 37 . 3774 30. 6943 . 54 4 9 . 7798
26 37 . 8 3 02 30 . 4090 . 5568 . 7783
27 37 . 62 2 6 30 . 4848 . 6177 . 7743
29 38 . 122 6 32 . 6991 . 3729 . 7949
30 37 . 6604 32 . 1312 . 3931 . 79 3 3
31 37 . 5094 31. 6999 . 3946 . 7935
33 37 . 2 17 0 29 . 1239 . 6094 . 7 7 17
34 37 . 3 0 19 30. 5747 . 4845 . 7849
36 38 . 7358 33 . 7010 . 3 190 . 7988

A l pha = 0 . 8027

-0
.i::...

•
 APPEND I X XV

Re l i ab i l i ty Analy s i s

Co lumn 3 (T)

I tem Corre l at i ons With Re st o f s c a l e and cronbach Alpha

Non- Ips ative s c al e , n= 1 0 7

Scale S cale corrected

Mean Vari ance I t em- Alpha
I f I t em I f I tem Total I f I t em
I tem Deleted Deleted Corre l ation Del eted

39 39 . 6132 3 5 . 9347 . 2228 . 7978

40 38 . 3 3 02 3 1 . 0233 . 6 6 14 . 7585
42 38. 0 18 9 3 2 . 0377 . 5586 . 7 688
43 38 . 0566 3 0 . 87 3 0 . 6616 . 7580
44 39. 3396 3 5 . 5978 . 18 8 7 . 8037
45 38 . 64 15 3 3 . 2 607 . 3 7 12 . 7874
47 39. 0566 32 . 4730 . 4671 . 7774
48 38. 59 4 3 3 2 . 3 7 67 . 4861 . 7755
49 38. 1509 3 4 . 1484 . 3480 . 7886
51 38. 8962 3 3 . 02 7 2 . 3936 . 7851
52 38. 3 58 5 31. 9845 . 54 3 0 . 7700
54 38 . 6038 3 3 . 59 3 9 . 3 67 2 . 7873

A l pha = 0 . 7950

'°
01
 APPEND I X XVI

Re l i ab i l i ty Analy s i s

Column 4 (E)

I t em Corre l ations With Rest o f Scal e and Cronbach Alpha

Non- Ipsat ive S c al e , n= 1 0 7

Scale Scale Corrected

Mean variance Item- Alpha
I f I t em I f I tem Total If I tem
Item Del e t e d Deleted Corre l ation Del eted

57 39 . 7642 29 . 7819 . 4725 . 7501

58 39 . 6226 28 . 9991 . 5284 . 7435
60 3 9 . 7 3 58 33 . 2 629 . 0885 . 7 907
61 3 9 . 2 54 7 29 . 6774 . 4890 . 7484
62 3 9 . 7 07 5 31. 504 1 . 2779 . 7703
63 39 . 4528 31. 5073 . . 2 5 57 . 7734
65 3 9 . 8 0 19 30. 1413 . 4299 . 7546
66 3 9 . 5660 29 . 8099 . 4796 . 7495
67 39 . 4434 29 . 6396 . 5129 . 7463
69 39 . 3302 29 . 4614 . 4 502 . 7521
70 39 . 6132 28 . 0299 . 4766 . 7491
72 3 9 . 1604 29 . 8883 . 5 19 3 . 74 65

Alpha = 0 . 7724

'°
0.

..
 APPENDIX XVI I

Rel i ab i l i ty Analy s i s

B i -p o l ar S c a l e X-T

I t em Correl ations W i th Re st of Scale and Cronbach Al � ha

Non - Ip s ative Sc.ale , n=1 4 9

Scale Scale corrected

Mean Vari an c e I tem- Alpha
I f I t em I f I t em Total I f I t em
I tem Del eted Deleted correlation Del eted

X3T3 9 46. 3960 3 5 . 2 4 08 . 19 8 6 . 6088

X4T4 0 48. 0738 3 3 . 7 17 5 . 2854 . 59 3 6
X 6T 4 2 48 . 5570 32 . 8160 . 2993 . 5903
X7T4 3 48 . 2349 3 0 . 8972 . 3836 . 5705
X8T4 4 46 . 5369 34 . 5611 . 2254 . 6045
X9T4 5 47 . 5570 32 . 2484 . 3370 . 5822
X 1 1T 4 7 46. 7047 3 3 . 52 03 . 3622 . 58 19
X12T48 47 . 6 1 07 3 3 . 5502 . 2559 . 59 9 2
X13T49 48 . 2 8 19 32 . 0011 . 3240 . 5847
X 1 5T5 1 47 . 3826 3 6 . 2 9 19 . 0791 . 63 12
X 1 6T 5 2 48 . 4295 3 3 . 4494 . 2861 . 5932
X18T54 47 . 6443 35 . 4875 . 1563 . 6165

Alpha = 0 . 6178

'-0
'-.!
 APPENDIX XVI I I

Rel i abi l i ty Analys i s

B i -p o l ar s c a l e I - E

I t em corre l at i ons W i t h Re st o f Scal e a n d cronbach Alpha

Non - I p s at ive s c al e , n= l 4 9

Scale scale . corre cted

Mean Vari ance I tem- Alpha
I f I tem I f I tem Total If I tem
I tem Del eted Deleted Corre l ation Deleted

I 2 1E57 42 . 9930 4 2 . 08 2 1 . 2964 . 7300

I 2 2E58 42 . 8099 3 9 . 1764 . 4803 . 7035
I24E60 4 3 . 2 394 43 . 4458 . 2794 . 7302
I2 5E61 4 3 . 3 662 40. 6376 . 5069 . 7030
I 2 6E62 43 . 5915 4 1 . 3213 . 3747 . 7 187
I27E63 43 . 2746 40 . 9093 . 4723 . 7 068
I2 9E65 4 3 . 58 4 5 42 . 9680 . 3 173 . 7256
I 3 0E6 6 43 . 3521 42 . 6553 . 3739 . 7 192
I 3 1E67 43 . 3099 4 3 . 3 07 6 . 2351 . 7296
I 3 3E69 43 . 2535 44 . 3324 . 2897 . 7355
I 34E70 42 . 9930 4 0 . 9007 . 3830 . 7 177
I 3 6E7 2 44 . 9507 4 1 . 0868 . 4519 . 7 092

A l pha = 0 . 7 3 6 6

'-0
co

" ...
 APPENDIX X I X

Re l i abi l i ty Analys i s

Bi -po l ar s c a l e X-T

I t em Corr e l at i ons With Rest o f S c a l e and Cronbach Alpha

Non- Ips ative S c al e , n= 1 0 7

S cal e Scale corre cted

Mean Vari ance I tem- Squared Alpha
I f Item I f I tem Total Mu l t i p l e If I tem
I tem Deleted Deleted corre l ation Corre l at i o n Deleted

X03T3 9 4 4 . 5 5 14 58 . 5893 . 4236 . 3476 . 7010

X04T4 0 4 6 . 112 1 53 . 5722 . 5541 . 4350 . 67 9 2
X 0 6T4 2 46 . 6916 58 . 5927 . 3 2 05 . 64 8 2 . 7139
X07T4 3 4 6 . 6 07 5 53 . 6935 . 52 55 . 59 3 4 . 6834
X0 8T4 4 44 . 7850 57 . 7 9 3 0 . 3618 . 44 12 . 7080
X09T4 5 4 5 . 4 112 60. 5463 . 27 63 . 4344 . 7 188
X l lT 4 7 4 4 . 9 9 07 56 . 2735 . 5088 . 3854 . 6890
X 1 2T4 8 4 5 . 8 3 18 6 1 . 5563 . 2311 . 19 2 8 . 7242
X13T4 9 46. 4360 59 . 7 050 . 2721 . 1869 . 7206
Xl5T51 4 5 . 5888 60 . 2821 . 3 000 . 3 14 5 . 7 1 57
X1 6T5 2 46. 1215 60 . 3908 . 3 095 . 27 8 5 . 7145
X 1 8 T5 4 4 5 . 5888 62 . 3 953 . 2 161 . 2415 . 7251

Alpha = . 7261

'-0
'-0
 APPEND I X XX

Rel i abi l i ty Analys i s

Bi-po l ar s c a l e I - E

I t em Correl ati ons Wi th Rest o f Scal e and Cronbach Alpha

Non- Ipsative scal e , n= 1 0 7

Scale scale Corrected

Mean Variance I t em- squared Alpha
If I tem I f I tem Total Mul t i p l e If Item
I tem Deleted Deleted Corr e l at i on Corre l at i on Del eted

I 2 1E57 4 1 . 5047 5 3 . 08 2 5 . 3384 . 18 8 5 . 7 559

I 22E58 4 1 . 5607 52 . 3 052 . 3927 . 2877 . 7488
I24E60 4 1 . 7944 57 . 4 4 7 9 . 2073 . 1083 . 7665
I 2 5E 6 1 42 . 0561 52 . 9025 . 4843 . 2976 . 7391
I 2 6E 6 2 42 . 0841 52 . 5117 . 4 14 6 . 3 151 . 74 59
I27E63 4 2 . 12 15 53 . 3530 . 4 105 . 3335 . 7 4 64
I 2 9E65 4 2 . 2 617 56 . 9 2 1 0 . 2796 . 13 9 1 . 7592
I 3 0E 6 6 42 . 0374 55 . 3571 . 3331 . 1847 . 7545
I3 1E67 4 2 . 02 8 0 53 . 5558 . 4506 . 3005 . 7428
I33E69 4 1 . 8224 49 . 9587 . 5259 . 3445 . 7 3 18
I 3 4E7 0 4 1 . 6636 5 0 . 3 19 7 . 44 56 . 2748 . 7425
I3 6E72 4 3 . 52 3 4 52 . 1386 . 5174 . 3 4 14 . 7 3 52

Alpha = . 7638

__,
0
0

..
..
 MANAGEMENT OF INNOVATION AND NEW TECHNOLOGY
WORKING PAPER SERIES

1. R.G. Cooper and E.J. Kleinschmidt, "How the New Product Impacts on Success and Failure
in the Chemical Industry", February, 1 992.

2. R.G. Cooper and E.J. Kleinschmidt, "Major New Products: What Distinguishes the Winners
in the Chemical Industry", February, 1 992.

3. J. Miltenburg, " On the Equivalence of IlT and MRP as Technologies for Reducing Wastes
in Manufacturing, March, 1 992.

4. J.B. Kim, I . Krinsky and J. Lee, "Valuation of Initial Public Offerings: Evidence from
Korea", February, 1 992.

5. M . B asadur and S . Robinson, "The New Creative Thinking Skills Needed for Total Quality
Management to Become Fact, Not Just Philosophy" , April, 1 992.

6. S . Edgett and S . Parkinson, "The Development of New Services Distinguishing Between

Success and Failure", April, 1 992.

7. A.R. Montazemi and K.M. Gupta, "Planning and Development of Information Systems
Towards Strategic Advantage of a Firm", April, 1 992.

8. A.R. Montazemi, "Reducing the Complexity o f MIS Innovation Through Hypermedia and
Expert Systems" , May, 1 992.

9. M . B asadur and Bruce Paton, "Creativity Boosts Profits i n Recessionary Times - Broadening
the P laying Field" , June, 1 992.

10. Robert G . Cooper and Elko Kleinschmidt, " Stage-Gate Systems for Product Innovation:
Rationale and Results", June, 1 992.

11. S .A.W. Drew, "The Strategic Management oflnnovation in the Financial Services Industry :
An Empirical Study" , July, 1 992.

12. M . Shehata and M.E. Ibrahim, "The Impact of Tax Policies o n Firms' R & D Spending
Behavior: The Case of R & D Tax Credit", July, 1 992.
13. Willi H . Wiesner, "Development Interview Technology: Implications for Innovative
Organizations", July, 1 992.

14. Isik U. Zeytinoglu, "Technological Innovation and the Creation of a New Type of
Employment: Telework", August, 1 992.

15. John W . Medcof, "An Integrated Model for Teaching the Management of Innovation in the
Introduction to Organizational Behaviour Course", October, 1 992.

1 6. Min Basadur, "The Why-What's Stopping Analysis: A New Methodology for Formulating
Ill-Structured Problems", October, 1 992.

1 7. Stephen A.W. Drew, "Strategy, Innovation and Organizational Learning an Integrative

Framework, Case Histories and Directions for Research", November, 1 992.

1 8. Stephen A.W. Drew, "Innovation and Strategy in Financial Services", November, 1 992.

1 9. Scott Edgett, "New Product Development Practices for Retail Financial Services",
November, 1 992.

20. Robert G. Cooper and Elko J. Kleinschmidt, "New Product Winners and Losers: The
Relative Importance of Success Factors - Perception vs. Reality", November, 1 992.

21. Robert G. Cooper and Elko J. Kleinschmidt, " A New Product Success Factors Model : An �
Empirical Validation", November, 1 992.

22. Robert G. Cooper & Elko J. Kleinschmidt, " Stage Gate Systems: A Game Plan for New
Product Success", November, 1 992.

23. Min Basadur, "Optimal Ideation-Evaluation Ratios", March, 1 993.

24. Christopher K. Bart, "Gagging on Chaos", March, 1 993 .

25. Yufei Yuan, "The Role of lnformation Technology in Business Innovation", July, 1 993 .

26. Isik Urla Zeytinoglu, "Innovation in Employment: A Telework Experiment in Ontario ", July,
1 993 .

27. John Miltenburg and David Sparling, "Managing and Reducing Total Cycle Time: Models
and Analysis", August, 1 993. •

28. R.G. Cooper, C.J. Easingwood, S . Edgett, E.J. Kleinschmidt and C. Storey, "What
Distinguishes the Top Performers in Financial Services", September, 1 993.

29. B.E. Lynn, "Innovation and Accounting Research", September, 1 993 .

30. Min Basadur and Peter Hausdorf, "Measuring Additional Divergent Thinking Attitudes
Related to Creative Problem Solving and Innovation Management", November, 1 993.

31. R.G. Cooper and E.J. Kleinschmidt, "Determinants of Time Efficiency in Product
Development", December, 1 993 .

32. Christopher K. Bart, "Back to the Future: Timeless Lessons for Organizational Success",
February, 1 994.

33. Ken R . Deal and Scott J . Edgett, "Determining Success Criteria for New Financial Products;
A Comparative Analysis of CART, Logit and Discriminant Analysis", February, 1 995.

34. Christopher K. Bart and Mark C. Baetz, "Does Mission Matter?" , February, 1 995.

35. Christopher K. Bart, "Controlling New Products : A Contingency Approach" , February,

1 995.

36. Christopher K. Bart, "Is Fortune Magazine Right? An Investigation into the Application
of Deutschman's 1 6 High-Tech Management Practices", February, 1 995.

37. Christopher K. Bart, "The Impact of Mission on Firm Innovativeness", February, 1 995 .

38. John W. Medcof, "Transnational Technology Networks", April, 1 995.

39. R.G. Cooper and E.J. Kleinschmidt, "Benchmarking the Critical Success Factors of Firms'
New Product Development Programs", April, 1 995.

40. John W . Medcof, "Trends in Selected High Technology Industries", July, 1 99 5 .

41. Robert C . Cooper & E.J. Kleinschmidt, "Benchmarking Firms' New Product Performance
& Practices", September, 1 995.

42. Min Basadur and Darryl Kirkland, "Training Effects on the Divergent Thinking Attitudes
of South American Managers" , November, 1 995.

43. Min Basadur, "Organizational Development Interventions for Enhancing Creativity in the
Workplace", November, 1 995.

44. Min Basadur, "Training Managerial Evaluative and Ideational Skills in Creative Problem
Solving: A Causal Model", December, 1 995.
45. Min Basadur, Pam Pringle and Simon Taggar, "Improving the Reliability of Three New
Scales Which Measure Three New Divergent Thinking Attitudes Related to Organizational
Creativity", December, 1 995.
•

46. N. P. Archer and F. Ghasemzadeh, "Proj ect Portfolio Selection Techniques : A Review and
a Suggested Integrated Approach", February, 1 996.

47. Elko J. Kleinschmidt, "Successful new product development in Australia: An empirical

analysis", February, 1 996.

48. Christopher K. Bart, "Industrial Firms & the Power of Mission," April, 1 996.

49. N. P. Archer and F. Ghasemzadeh, "Project Portfolio Selection Management through

Decision Support: A System Prototype," April, 1 996.

50. John W. Medcof, "Challenges in Collaboration Management in Overseas Technology Units,"

April, 1 996.

51. Susan L . Kichuk and Willi H. Wiesner, "Personality and Team Performance: Implications
for Selecting Successful Product Design Teams," May, 1 996.

52. Susan L. Kichuk and Willi H. Wiesner, "Selection Measures for a Team Environment: The
Relationships among the Wonderlic Personnel Test, The Neo-FFI, and the Teamwork KSA
Test, " May, 1 996. '

53. Susan L. Kichuk and Willi H . Wiesner, "Personality, Performance, Satisfaction, and Potential
Longevity in Product Design Teams," June, 1 996.

54. John W. Medcof, "Learning, Positioning and Alliance Partner Selection," June, 1 996.

55. Scott J. Edgett, "The New Product Development Process fo r Commercial Financial
Services," July, 1 996.

56. Christopher K. Bart, "Sex, Lies & Mission Statements," September, 1 996.

57. Stuart Mestelman and Mohamed Shehata, "The Impact of Research and Development
Subsidies on the Employment of Research and Development Inputs," November, 1 996.

58. Mark C. Baetz and Christopher K . Bart, "Developing Mission Statements Which Work,"
�
November, 1 996.

59. Fereidoun Ghasemzadeh, Norm Archer and Paul Iyogun, "A Zero-One Model for Project
Portfolio Selection and Scheduling," December, 1 996.
60. R. G. Cooper, S . J. Edgett, E. J. Kleinschmidt, "Portfolio Management in New Product
Development: Lessons from Leading Firms," February 1 997.

61. R. G. Cooper, S . J. Edgett, E. J. Kleinschmidt, "Portfolio Management in New Product

Development: Lessons from Leading Firms -- Part II," February 1 997.

62. C. K. Bart, "A Comparison of Mission Statements & Their Rationales in Innovative and
Non-Innovative Firms," February 1 997.

63 . R. Bassett, N. P. Archer and W. G. Truscott, "Data Webs: An Evaluation of an Innovative

Information Management Tool that Integrates Databases with the World Wide Web," April
1 997.

64. S. Taggar, "Intelligence, Personality, Creativity and Behaviour: The Antecedents of Superior
Team Performance," April 1 997.

65. R. Deaves and I. Krinsky, "New Tools for Investment Decision-Making: Real Options
Analysis," May 1 997.

66. J. W. Medcof (ed.), "Trends and Events in Selected High Technology Industries," May,
1 997. (On the WEB only)

67. C. K. Bart, "Product Innovation Charters: A State-of-the-Art Review," May, 1 997.

68. John W. Medcof, "Strategic Contingencies and Power i n Networks of Internationally

Dispersed R&D Facilities", August, 1 997.

69. John W. Medcof, "Research Intensity and the Identification ofHigh Technology Industries,"
September, 1 997.

70. Christopher K. Bart and John C. Tabone, "Mission Statements in the Not-for-profit Health
Care Sector: A State of the Art Review," September, 1 997.

71. Elko J. Kleinschmidt, "In-house and Partnership New Product Development in Austria: An
Empirical Analysis on Outcome and Explanatory Factors," October, 1 997.

72. Robert G. Cooper, Scott J. Edgett and Elko J. Kleinschmidt, "R&D Portfolio Management
Best Practices: Methods Used & Performance Results Achieved," January, 1 998.

73. Christopher K. Bart and Simon Taggar, "A Model of the Impact of Mission Rationale,
Content, Process and Alignment on Firm Performance," March, 1 99 8 .

74. Christopher K. Bart, John Parkinson and Simon Taggar, "The Implementation of Strategy:
Behavioural vs Budgetary Approaches and the Effect of Participation," March, 1 998.
75. John W. Medcof, "The Resource Based View and the New Competitive Landscape:
1
Characterizing Positions of Dynamic Capability," May, 1 998.

76. F. Ghasemzadeh and N. P. Archer, "Proj ect Portfolio Selection Through Decision Support,"
June, 1 998.

77. Y. Yuan, N. Archer, and R. Bassett, "The Impact of Electronic Commerce Innovations on
Marketing Management," June, 1 998.

78. Kenneth S . Chan, James Chowhan, Stuart Mestelman, Mohamed Shehata, "Value
Orientations and Income and Displacement Effects," July 1 998.

79. Min Basadur, Laurent Lapierre, "Predicting Creative Problem Solving Behaviors within
Teams," September, 1 998.

80. Min Basadur, "Simplex: Modelling the Phases and Stages ofthe Innovation Process in Open­
System Organizations, October, 1 998.

81. Ken Deal, Ben Long and Bryan Scott, "New Pricing Product Design for Competitive
Advantage, November, 1 998.

82. Min Basadur, Mark A. Runco and Luis A. Vega, "Understanding How Creative Thinking
Skills, Attitudes and Behaviors Work Together in Real World Managerial Problem Solving,"
November, 1 998. )

innova/papers.irc

..

lllllH�I��lllll'if!�Hiii�II
3 9005 0245 5434 1

-� ��-- ������
 � -----------i

l .r., --1 ., Or
t i 1n J f\. � !

1-\ 0
4S
_ Wlo�7
£f\ O . 33
 �. ' .

. " . ., �- , -

.'
.

"' . '��l;
D. i

The activities of the Management of

Innovation and New Technology Research :-.
• ' ' } ''":·1.

Centre are generously supported by:

i .
• The DeGroote family
• DuPont Canada Inc
• Nortel
• Royal Bank
• The Society of Management
Accountants of Ontario

MCMASTER
• U N I V E R S I T Y•

MICHAEL G . D•GROOTE

SCHOOL OF BUSINESS

I N F O R M AT I O N . . .

For information about the Management of l ·. ' \\

>: '
Innovation and New Technology Research � . , l

Centre or innovation research activities at the

Michael G. DeGroote School of Business:

Director; Dr. C. K. Bart, Ph. D., C.A.

MINT - Research Centre
Michael G. DeGroote School of Business
McMaster University, 1 280 Main St. W.
Hamilton ON Canada LBS 4M4 '

- . f l -; � .

, (... .

Phone: 905-525-9 1 40, Ext. 24 1 05 [ ' ·- .. · '.l

Fax: 905-52 1 -8 995

email: [email protected]
- '. . ..
www: http://mint. mctnaster.ca/ i

/ , ·� - - F;
*Management of Innovation and New Technology Research Centre
(MINT) is an official mark of the
Michael G. DeGroote School of Business,
McMaster University.
� '
.
I-,

/ ;

(
•