Content uploaded by Min Basadur
Author content
All content in this area was uploaded by Min Basadur on Jul 31, 2020
Content may be subject to copyright.
MIN BASADUR
MARK A. RUNCO
LUIS A. VEGA
Understanding How Creative Thinking
Skills, Attitudes and Behaviors Work
Together: A Causal ProcessModel
ABSTRAG
Managers
(N
=
112) from a large international consumer goods
manufacturer participated in a field experiment in which they
learned and applied the Simplex process of creative thinking
to solve real management problems. The interrelationships
among six attitudinal and behavioral skill variables learned
during the training were measured to improve understanding
of how these variables contribute to the process. Predicted
relationships were tested and a best-fit causal model was
developed. Behavioral skill in generating quantity of options
was the most important variable overall: it was directly associ-
ated with behavioral skill in both generating quality options
and evaluating options. The key attitudinal skill and the sec-
ond most important variable overall was the preference for
avoiding premature evaluation of options (deferral of judg-
ment). The other attitude measured, the preference for active
divergence, played only an indirect role in the process.
INTRODUGION
Attempts to categorize the study of creativity (e.g. Murdock
&
Puccio, 1993) frequently emphasize the four "Ps": product,
person, press (environment) and process. Most research
focuses on one category. O'Quin and Besemer (1989) and
Jackson and Messick (1964), for example, focused on under-
standing and assessing the product of creative efforts. Mean-
while, one aspect of the "person" approach has been
identification:
the development of cognitive and personality
tests capable of identifying more or less creative people.
Dunnette (1976), Gough (1976), Roe (1976) and Torrance
(1974) provide comprehensive reviews of this identification
movement. Guilford's work (1968) is among the best-known
77
Volume
34
Number
2
Second Quarter 2000
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
in the cognitive realm and MacKinnon's (1962, 1977) in the
personality realm. Kirton (1976) and Myers (1962) addressed
the relationship between personality and creative behavior, and
Guilford (1968) addressed the cognitive aptitudes and abili-
ties associated with various kinds of (potentially creative) think-
ing. The study of environmental "presses" has been pursued
by Amabile and Gryskiewicz (1989), Andrews and Farris
(1972), and Baker, Winkofsky, Langmeyer and Sweeney
(1976), among others.
The Process The focus on the fourth "P" is apparent in research that
Approach models creativity as a process. For example, Basadur (1979,
1982, 1992) portrayed individual, team and organizational cre-
ativity as a continuous, dynamic, circular three-phase process
of finding good problems, solving them and implementing
good solutions (see Figure 1).
FIGURE 1.
Organizational Creativity as a Process
\\
\
\
\
\
\\\
)----------
/
/
/
/
/
/
I
I
/
Solution
Implementation
Activity
78
Environment
Problem
Finding
Activity
Problem
Solving
Activity
Journal of Creative Behavior
Problem finding
means continuously and deliberately dis-
covering and formulating new and useful problems to be solved.
In organizations, this includes generating new products or ser-
vices by anticipating new customer needs, by discovering ways
to improve existing products, services, procedures and pro-
cesses, or by identifying opportunities to improve the satisfac-
tion and well-being of organizational members and pertinent
groups outside the organization. Problem finding also includes
defining and conceptualizing such new problems and oppor-
tunities (regarding them as "fuzzy situations") accurately and
creatively to clearly visualize the big picture and to identify
more specific challenges and insights and relate them to one
another.
Problem solving
means developing new, useful,
imaginative solutions to these problems.
Solution implement-
ing
means successfully impiementing such new solutions.
Each implemented solution leads to new, useful problems to
be discovered - hence the circular process. Research shows
that effective organizations do what ittakes to mainstream such
a process (make it an everyday habit among its members) for
continuous innovation and for intrinsic motivation (Basadur,
1993,1997). Research also shows that skills in such a process
can be deliberately developed (Basadur, 1979, 1994).
A creativity process that incorporates divergent and con-
vergent thinking within multiple phases or steps is a "com-
plete" process. Basadur, Graen and Green (1982) identified a
two-step mini-process called
ideation-evaluation
in which
divergent and convergent thinking occur sequentially (see Fig-
ure 2). Osborn (1963), Parnes, Noller, and Biondi (1977), and
FIGURE 2.
Ideation-Evaluation: A Sequential Two-Step Creative Thinking
Mini-process.
STEP 1
DIVERGE STEP2
CONVERGE
Options
Poinls of View
Possibilities
Facts
Opinions
Items
Ideas
Things
Criteria
Problems
Solutions
Actions
~~~
oj<:I:>\'N
--{<c-
SJ
Quality
Judgmental
Disciplined
Intellect
Adult
Quantity
lmaginati
v
e
Fre;,
Gut
Child
79
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
Isaksen and Treffinger (1985) provided linear three-, five- and
six-step models of a complete process. Basadur (1974, 1979,
1981, 1982, 1983, 1992) used real-world, organizational appli-
cation experience and field research to extend the basic
Osborn-Parries three- and five-step models in two ways. The
process can be represented as eight steps within the three
phases of Figure 1. After the problem-finding phase has been
divided into two stages - problem
generating
and problem
formulating -
the process can also be represented as a circu-
lar, continuous, four-stage process of generating, formulating,
solving
and
implementing.
Each of the eight steps consists
of the two-step mini-process (see Figure 3); the entire process
is called Simplex (Basadur, 1981).
Ideation-evaluation
occurs within each of the eight steps
of the Simplex process.
Ideation,
or active divergence, is the
generation of options without evaluation (deferring judgment).
Evaluation,
or active convergence, is the application of judg-
ment to the generated options to select the most significant
options. Separating ideation from evaluation is a vital aspect
of this two-step process. This mini-process must be executed
FIGURE 3.
The Simplex Creative Process as a Whole.
STAGE 111
PROBLEM
SOLVING
STAGE II
PROBLEM
FORMULATING
80
____________ - 00 _
-------------
Journal of Creative Behavior
skilfully. For example, Basadur (1995) suggested that, in or-
der to achieve high-quality, innovative, creative results, an in-
dividual or a group requires not only the appropriate content
(i.e., the knowledge or the
what)
but a creative, innovative
process (the
how)
for working on that content, as well as suf-
ficient skills in using the process. He identified four such pro-
cess skills within the Simplex process:
• Active divergence - the ability to assertively generate
a variety of options.
• Active convergence - the ability to evaluate and choose
from among options and advance the process.
• Deferral of judgment - the ability to separate active
divergence from active convergence.
• Vertical deferral of judgment - the ability to avoid un-
consciously leapfrogging past steps, phases or stages
of the process.
Basadur incorporated these three components into a Qual-
ity Results Equation:
Quality Results
=
Content
+
Process
+
Process Skills
These four process skills permit successful application of
the ideation-evaluation mini-process within each step of the
Simplex process, and from step to step or stage to stage (see
Figure 3) throughout the process. Mastering the skills and the
process enables individuals or groups to move about the
process in different sequences as situations require. Synchro-
nizing these skills is especially vital for group members attempt-
ing to work creatively together.
Basadur and Finkbeiner (1985) identified specific attitudes
that
enhance
these process skills. They suggested that, un-
less the ideation-evaluation process is
accepted
attitudinally,
then the process will not likely occur. Thus, the process skills
have both attitudinal and behavioral components.
Basadur and Finkbeiner established a 14-item questionnaire
to measure two specific attitudes that make up acceptance
of ideation-evaluation: the preference for ideation (active
divergence) and the tendency to (not) evaluate prematurely
(preference for deferral of judgment). They suggested that
these two attitudes enhance and encourage the practice of the
two related behavioral skills. Encouraging active divergence
leads to generation of more options and deliberate develop-
ment of many points of view. Encouraging avoidance of pre-
mature convergence reduces the urge to prematurely judge or
81
-----------------~-------------.----.-.-----------.-- --
--~------
...
-
-----------
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
Skills, Behaviors And
Atti tudes Are
Needed to Make
The Process Work
82
analyze a fledgling thought. Basadur and Finkbeiner also sug-
gested that a low tendency toward premature convergence
would trigger a high preference for active divergence. That is,
the former, more passive attitude is a prerequisite trigger for
the latter, more active attitude. When people become skilled
in reducing premature convergence and increasing active di-
vergence, they create more, higher-quality options. These two
measures of the acceptance attitudes are used in the research
reported in this paper as explained later. Measures of the be-
havioral skill (practice of the ideation-evaluation process) are
also described later.
Basadur, Graen and Green (1982) and Basadur (1979,1994)
reported research that indicated that one's preference for and
skill in applying the ideation-evaluation mini-process might
differ in each of the three phases of the complete process. For
example, someone might prefer to defer judgment and actively
diverge in the solution-finding phase rather than in the prob-
lem-finding phase, or vice versa. Basadur suggested there
might be different optimal ratios of ideation and evaluation in
each of the phases for different fields of endeavor, and later
(1995) provided supporting evidence.
Basadur, Graen and Green (1982) reported a field experi-
ment which tested the effects of training the complete Sim-
plex process in an applied setting. They expected that the
training would improve five variables: (a) acceptance of the
ideation-evaluation thinking mini-process; (b) deliberate prac-
tice of the ideation-evaluation thinking mini-process; (c) prob-
lem-finding performance; (d) problem-solving performance;
and
(e)
solution implementation performance. The first of these
five variables is attitudinal and the second is behavioral.
Basadur et al suggested that these two attitudinal and behav-
ioral variables were necessary antecedents ofthe three perfor-
mance variables. They believed that, unless a positive change
in attitudes and behaviors occurred - motivating participants
to separate divergent and convergent thinking and to deliber-
ately apply divergent thinking - training would not improve
ideation performance. The expected training effects are mod-
eled in Figure 4.
This belief was based on Basadur's (1979, 1994) analysis
of previous laboratory and field experiments on the value of
providing creativity training. For example, Basadur found that
most of the research had tested the brainstorming technique.
Although brainstorming is a divergent thinking technique based
on the
use
of ideation-evaluation, none of the brainstorming
Journal of Creative Behavior
FIGURE 4.
Effects of Simplex Training on Attitudes, Behaviors and
Performance.
Training
in "A Complete
Process of
Creative
Problem Solving"
Organizational
Creative
Output
~
~
Practice of
.
the Ideatian-
Evaluation
Process
From Center for Research in Applied Creativity, Dr. Min Basadur, Copyright
©
1979, 1989, reprinted by permission.
research had attempted to measure to what extent the sub-
jects accepted the value of and employed any skills in ide-
ation-evaluation during the particular experiment (or more
permanently back in the real-world setting). To what extent
brainstorming performance correlated with the willingness to
accept ideation-evaluation and the skill in using it was never
tested. In other words, many earlier research studies provided
no more training than "giving brainstorming instructions"
(as if this were sufficient to effect sudden changes in brain-
storming attitudes and behaviors). For participants to say they
understand brainstorming rules is entirely different from
using the brainstorming rules skilfully, especially on real-world
problems and meaningful issues.
For example, several laboratory experiments indicated that
inhibitory influences make "training" (giving brainstorming
instructions) of groups less valuable than training of individu-
als (Bouchard, 1972; Bouchard
&
Hare, 1977; Dunnette,
Campbell
&
Jaastad, 1963; Shaw, 1971; Taylor, Berry
&
Block,
1958). However, none of these experiments measured inter-
mediate attitudinal or behavioral effects of such "training".
Not only is it unlikely that simple brainstorming instructions
qualify as sufficient training, but it is unlikely that many group
participants would truly accept and use those instructions.
They were more likely inhibited within the group and collec-
tively lacked sufficient attitudes and skills in the ideation-
83
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
Measuring the
Needed Skills,
Attitudes and
Behaviors
84
evaluation mechanism. These groups should be called
un-
trained, undertrained,
or
underdeveloped groups.
Unless
trainees significantly increase their acceptance of and skill in
ideation-evaluation, neither they nor their groups should be
expected to improve their creative performance compared to
untrained individuals or untrained or nominal groups. In other
words, the process modeled in Figure 4 strongly suggests that
training in creative problem solving must be of sufficient qual-
ity, impact, and duration to effect
real
improvements in ac-
ceptance of the validity of the skill in the mini-process of
ideation-evaluation and in the skill itself. This line of thinking
explains why so few new management techniques become
permanent and why many earn the ironic label of "flavor of
the month" (see Basadur
&
Robinson, 1993).
Basadur et al. (1982) systematically measured for the first
time the impact of creative problem-solving training on indi-
viduals both immediately after training and after their return
to work. They expected that the three performance variables
would be improved only if the antecedent variables - the ac-
ceptance and practice of the ideation-evaluation process - were
improved. These expectations are consistent with Kraut's
(1976) traditional industrial/organizational psychology train-
ing model: Training must go beyond
understanding
to change
attitudes
and to change
behaviors
in order to achieve supe-
rior
results.
Basadur et al stressed that essentially none of the
previous research in creativity training had addressed the in-
termediate steps in Kraut's model. Their research attempted
to measure and understand to what extent changes in accep-
tance of (attitude) and practice of (behavior) ideation-evalua-
tion might
actually
result from training and accompany
changes in performance (results). In previous research, this
link between training and changes in acceptance and practice
of the fundamental ideation-evaluation process had simply
been
assumed
to occur.
Basadur et al. (1982) extended the previous research in sev-
eral other ways. One way was to attempt to understand prob-
lem
finding
and solution
implementation
as well as
problem
solving.
(Virtually all previous organizational research had
focused only on
problem solving.)
Another way was to focus
primarily on effects and mechanisms concerning
individual
attitudes, behaviors, and performance in a
real-world
setting
(rather than in a laboratory setting). What little previous
research had occurred in relatively real-world settings had
been limited to group variables. It was also suggested that,
85
Journal of Creative Behavior
compared to simple brainstorming, a complete process such
as Simplex would prove more useful and more credible (and
less subject to the skepticism that often plagues creativity train-
ing) among participants from real-world business and other
organizations (Basadur, 1997; Basadur, Graen,
&
Scandura,
1986). Unlike earlier research in which participants were asked
merely to apply brainstorming rules without training, Basadur,
Wakabayashi, and Graen (1990) stressed the importance of
building skills through at least two days of hands-on practice
using real-world problems.
Thus, Figure 4 offers the starting point of a theoretical model
to explain how training increases organizational creativity and
innovation. This model postulates that, in order to achieve
meaningful increases in problem finding, problem solving, or
solution implementation performance and organizational re-
sults, the impact of training must be sufficient to increase ac-
ceptance and practice of the ideation-evaluation process. For
simplicity's sake, the model excludes various organizational,
group and individual work-related factors that also affect cre-
ativity. The results of Basadur et al's research supported the
model in Figure 4. Compared to a control group, the experi-
mental training group achieved significant increases in the
acceptance and practice of ideation-evaluation and significant
increases in the performance variables measured. Potential
interrelationships among the constructs (indicated in dashed
lines) were left for future research. The model is useful prima-
rily for identifying the key constructs that must be affected in
order for training to succeed. The purpose of this paper is to
investigate the relationships between the two antecedent vari-
ables of Figure
4
in the problem-solving phase of the Simplex
creative problem-solving process.
While Basadur et al. (1982) restricted their measures to di-
verging variables, Runco and Basadur (1993) extended this
work to include converging variables. They demonstrated that
Simplex training participants significantly improved their evalu-
ative skill, and that evaluative skill correlated positively with
their ideational skill. This result is consistent with Runco and
Vega's (1990) suggestion that individuals with high ideational
abilities have more opportunity to exercise evaluative skills. In
summary, participants given appropriate training have im-
proved their evaluative skills, ideational skills and ideational
attitudes, which are associated with ideational skills.
The main objective of this study is to advance the research
reported in Basadur et al. (1982), Runco and Basadur (1993),
Understanding How Creative Thinking Skills. Attitudes and Behaviors Work Together
and Basadur (1994) by field testing and modelling the interre-
lationships among the various attitudinal and skill variables
discussed above. The testing and modelling is restricted to the
problem-solving phase of the complete Simplex process.
HYPOTHESES
The model in Figure 5 predicts the expected relationships
among these variables. It is expected that ideational skill will
be associated with evaluative skill for the following reasons.
First, generating more, and better, criteria for evaluation should
lead to better evaluation. Second, the preference for avoiding
premature convergence should be associated with evaluative
skill, as the evaluation will be less hasty. Finally, ideational skill
includes skill in conducting the two-step, ideation-evaluation
mini-process (discussed under "Training" in the "Method" sec-
tion). This skill includes the ability to separate ideation from
evaluation. Just as practicing evaluation during the ideation
step produces poor options, so practicing ideation during the
evaluation step produces poor choices and decisions. Thus,
better evaluation should result from better ideation. A set of
hypotheses consistent with these predictions follows:
HI: improvements in the attitude of preference for (avoid-
ing) premature convergence will be associated with
improvements in (a) the attitude of preference for ac-
tive divergence (H1A); (b) ideational skill (quantity)
(H1B); (c) ideational skill (quality) (H1C); and (d) evalu-
ative skill (HID); .
H2: improvements in the attitude of preference for active
divergence willbe associated with improvements in ide-
ational skill (quantity) (H2);
H3: improvements in ideational skill (quantity) will be as-
sociated with improvements in ideational skill (quality)
(H3A) and evaluative skill (H3B);
H4: improvements in ideational skill (quality) will be asso-
ciated with improvements in evaluative skill.
METHOD
Before and after training, participants were measured on the
variables contained in the hypotheses. As a procedural check,
their gains after training were checked against previous re-
search, using multivariate analysis of variance. This was done
to ensure that a normal "take" had occurred consistent with
previous research and that a control group was unnecessary.
Then the hypothesized relationships among the variables as
portrayed in Figure 5 were tested using a special causal analy-
sis model (Bollen, 1989).
86
--_
..
_---_. --_._---_._----_._-------
87
Journal of Creative Behavior
FIGURE 5.
Predicted Causal Model From Theory and Previous Research.
Participants
The participants
(f'f
=
112)were lower-through upper-middle
managers employed by a large international consumer goods
manufacturer. They represented a wide range of functions, in-
cluding finance, manufacturing, operations, employee relations
and distribution. Most participants had marketing and sales
responsibilities. The organization was keenly interested in en-
hancing its members' creative problem-solving process skills
to help them respond to a dynamic environment including
changing marketing channels, customer needs and new tech-
nology.
Training
Participants underwent 20 hours of intensive training in the
Simplex creative process over two days. This training was pri-
marily hands-on and experiential, with exercises and diverse
tasks that encouraged participants to discover for themselves
such concepts as the value of ideation-evaluation. Using their
chosen real-worldproblems, the participants practised ideation-
evaluation in completing all three phases of the process: find-
ing and defining their own problems, solving those problems,
and preparing to implement their own solutions (implementa-
tion would occur after the training).
Various tools and techniques of ideation and evaluation were
practiced in each of the eight steps and three phases of the
process (see Basadur, 1995). The participants frequently
--- ----------
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
Instrumentation and
Procedure
88
reviewed
their experiences, learnings and insights. One goal
of the training is to induce participants to become skilled in
synchronizing divergent and convergent thinking within and
among each of the eight steps and the three phases of the
process. Another goal is to encourage participants to value
concrete experience as well as abstract thinking, and to ap-
preciate how attitudinal processes enhance cognitive pro-
cesses in creative thinking and problem solving.
Two ideational attitudes, "preference for active divergence"
and "preference for (avoiding) premature convergence," were
measured using the Basadur 14-item Ideation-Evaluation Pref-
erence Scale (Basadur [;. Finkbeiner, 1985). Participants rated
their level of agreement or disagreement with each of the 14
statements on a five-point Likert scale before and after the train-
ing. The minimum/maximum range for "preference for active
divergence" was 6 to 30, and for preference for (avoiding) pre-
mature convergence, 8 to 40.
This study also employed measures of ideational skill and
evaluative skill. The two ideational skill measures included
quantity and quality (originality) of ideas generated to solve
real-world managerial problems. The measure of evaluative skill
assessed managers' ability to recognize creative and original
ideas. The measures were applied to responses to four open-
ended tasks (two tasks performed before training and two train-
ing). Participants were randomly assigned to two subgroups.
Each subgroup took two minutes to write down as many ideas
as possible to
solve
each of two work-related problems. All four
problems were selected in consultation with senior managers
to ensure that the problems were relevant to the participants.
The instructions for these problems were as follows: "Work
alone. Remember that these are not 'tests'. They do not re-
ceive grades nor are there' incorrect' answers. In addition, your
responses are completely confidential."
The tasks were ordered so that one subgroup received
two of the four tasks
before
training and the other subgroup
received the same two tasks
after
training, and vice versa. This
added precaution was taken to minimize the potential effects
of pre- and post-task differences, even though great care was
taken to ensure that all the problems were equally open-ended
and meaningful.
The four tasks were:
A: How might we install our vending equipment accounts
faster?
-- -_._----- ----------------------_._-- --------------------
-- ---- ------
89
Journal of Creative Behavior
B: How might we get into and out of our markets faster?
C: How might we make better decisions as a team?
D: How might we communicate better among ourselves?
Tasks A and B were considered to be equivalent technical
problems, and tasks C and D were considered equivalent or-
ganizational or behavioral problems. One subgroup received
tasks A and C before training and tasks Band D after training,
and the other subgroup did the reverse. Ensuring a minimum
of content overlap between tasks A and C and tasks Band
D avoided the possibility that similarity of content made the
after-training task easier than the before-training task.
The responses to these four problems were scored for quan-
tity and quality. Quantity scores were derived by counting the
number of ideas given. Quality scores were derived from the
number of original or unique ideas (ideas given by only one
participant). Quantity scores therefore represent ideational flu-
ency, and quality scores represent ideational originality
(Guilford, 1968; Runco
&
Albert, 1985; Torrance, 1974).
Evaluative skill was assessed by asking each participant to
rate his or her own ideas on a seven-level scale (1
=
entirely
unoriginal; 7
=
highly original). Rather than calculate correla-
tional scores between ratings and actual originality as in Runco
and Vega (1990), a simplified scoring technique was used. The
evaluative scores were determined by calculating the number
of original ideas that were accurately rated. A second score
reflected the number of unoriginal ideas correctly identified.
This evaluative measure was used previously with manag-
ers in Runco and Basadur (1993), and is analogous to that
used with satisfactory reliability with parents, teachers and
children in Runco (1991) and in Runco and Vega (1990). Be-
cause significant cognitive differences exist between intra-per-
sonal evaluations and interpersonal evaluations (Runco
&
Chand, 1994; Runco
&
Smith, 1992), this measure is impor-
tant. Because participants know so much about the "associa-
tive history" of their own ideas, intra-personal evaluations are
generally much easier to make than interpersonal evaluations.
Conversely, because individuals tend to judge their own ideas
either extremely strictly or extremely leniently, they may find
it easier to remain objective about someone else's ideas.
Rnalysis
Two procedural checks were made. The first was to ensure
that the responses to the two task sets (AC and BD) were not
significantly different. A simple t-test analysis compared task
set means both before and after the training.
----------- -----
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
The second procedural check was to ensure that before- and
after-training effects were similar to previous training research
results in several controlled field experiments, ensuring that a
"take" had been achieved and removing the necessity for a
control group. These before- and after-training effects were
analyzed through multivariate analysis of variance (MANOVA).
The causal modelling analysis was performed using Lisrel 8
(Joreskog
&
Sorbom, 1993).
RESULTS
Both procedural checks were successful. There were no sig-
nificant differences (p S .05) between mean task set scores on
any of the ideational or evaluational skill measures, either pre-
test or post-test (see Table 1). As demonstrated in the multi-
variate and univariate analysis results in Table 2, the training
produced the intended results for all but evaluation of non-
original solution ideas. The quantity of ideas produced in-
creased significantly from a mean of 9.1 to a mean of 13.4
ideas (p S .001). There was also a significant increase in the
TABLE 1.
Comparison of Responses to the Two Different Task Sets.
Pre-test Means
Task Set
BD AC
t
Ideational Skill (Quantity) 9.2 9.2 0.1 (n.s.)
Ideational Skill (Quality) 3.5 3.5
a
(n.s.)
Evaluational Skill (Original Ideas) 1.2 0.8
1.6
(n.s.)
Evaluational Skill
1.8
2.1 0.2 (n.s.)
(Non-Original Ideas )
Post-test Means
Task Set
BD AC t
Ideational Skill (Quantity) 13.6 14.3 0.4 (n.s.)
Ideational Skill (Quality) 5.3 5.9 0.6 (n.s.)
Evaluational Skill (Original Ideas) 2.6 2.3 0.4 (n.s.)
Evaluational Skill 2.1 1.9 0.4 (n.s.)
(Non-Original Ideas)
Note: n.s. means that the difference was not significant at
p
<
.05.
90
-----------.---------
..
_-
Journal of Creative Behavior
TABLE 2.
Multivariate Analysis of Variance for the Pre- and Post-Training
Test Scores.
MULTIVARIATE: Wilks' Criterion:
F(6,78)
= 21.75,
P
<
.001
UNIVARIATE:
PRE POST
t-Value F-Value D.F.
Rc
Ideation Skill M9.1 13.4*** 6.7 45.1 (1,83) 0.57
(Quantity) (SD) (4.2) (6.5)
Ideation Skill M3.1 5.4*** 5.6 31.8 (1,83) 0.48
(Quality) (SD) (3.1) (4.1)
Evaluation Skill M1.0 2.5*** 4.7 22.4 (1,83) 0.40
(Originality) (SD) (1.3) (3.1)
Evaluation Skill M2.1 2.0
n.s.
-0.5 0.3 (1,83) -0.04
(Non-originality) (SD) (1.8) (1.9)
Preference for
Premature M 3.1 2.4*** -9.4 87.6 (1,83) -0.79
Converqence!" (SD) (0.5) (0.7)
Preference for
Active M 3.8 4.0** 2.9 8.3 (1,83) 0.25
Divergence (SD) (0.6) (0.7)
Note:
**
p
<
.01
***
P
<
.001
n.s.
p>
.05
(I)
Special Note: In this table, a lower score on the attitude of preference for
premature convergence indicates a stronger attitude of
preference for (avoiding) premature convergence.
quality of ideas produced from a mean of 3.1 to a mean of 5.4
original ideas (p S.001). This increase may partly reflect the
participants' improved evaluative skills. After training, they
recognized significantly more accurately which of their ideas
were original (an increase to a mean of 2.5 original ideas
accurately evaluated from a mean of 1.0). It is somewhat sur-
prising that participants failed to improve their recognition of
unoriginal ideas, but, practically speaking, this may not be
a critical point. Even with a drop in this index, the production
of original ideas improved. Both ideational attitude means
increased significantly after training. (Note that the preference
for premature convergence attitude change score is negative
in Table 2 as expected and must be reversed to represent the
preference for
avoiding
premature convergence attitude.)
The results of the causal model using Lisrel 8 are shown in
Figure 6. The fit indices show that the data fit the hypothesized
91
Understanding How Creative Thinking Skills. Attitudes and Behaviors Work Together
model of Figure 5 quite well, although two of the hypothesized
relationships were not supported and the direction of two oth-
ers was reversed. The fit of the model was evaluated using dif-
ferent criteria (e.g., Bollen,
1989):
the chi-square goodness of
fit statistic (in which a non-significant chi-square indicates ex-
cellent model fit) and the chi-square/degrees offreedom ratio
(generally, values below
2.0
are viewed as acceptable); the
Goodness of Fit Index (GFJ); and the Adjusted Goodness of
Fit Index (AGFI) (values over
.90
are viewed as indicating
good fit) (Joreskog
&
Sorborn, 1989).
The AGFI adjusts for
the degrees of freedom of a model relative to the number of
variables; as a result, the AGFI rewards simpler models with
fewer parameters (Bollen,
1989).
These indices for the model
were X2
=
22.37
(p
=
.498),
X2/df
(22.37/23
=
.98),
GFI
=
.975,
and AGFI
=
.870,
all indicating excellent fit.
Figure 6 presents the standardized solution to the model.
The values in each path are the structural coefficients and can
be interpreted in the same manner as standardized weights in
FIGURE 6.
Results of Testing the Predicted Model.
... ~qr
~,q~!f'~1'Y.~••
_>
.10 (H2)
r
.32"
(H1A)
... ';?T.
S.ICjC'!F,!fi~'!r•••
>
.07 (H1C)
Evaluation
Skill
(Original
Ideas)
Note:
*p
<
.05,
h
P
<
.01
Chi-square with 23 degrees of freedom 22.37
(p
=
.498)
Goodness of fit index
=
.975
Adjusted goodness of fit index
=
.870
Root Mean Square Residual
=
1.16
92
----------._-------------_
..
__
._----
Journal of Creative Behavior
multiple regression. As can be seen in Figure 6, Hypothesis
H2, which stated that increased preference for active divergence
would be associated with increased ideational skill (quantity),
was not supported (i.e., .10,
p
>
.05). Nor was Hypothesis HIe,
which stated that increased preference for avoiding premature
convergence would be associated with increased ideational skill
(quality/originality), (i.e., .07, p
>
.05).
Also, Hypothesis H3B, associating increase in ideational skill
(quantity) with increases in evaluative skill, was supported
directly only in terms of the evaluation of non-original ideas.
The corresponding association with evaluation of original ideas
turned out to be an indirect one via pathway H3A, then fol-
lowed by new pathway H4A (explained below).
All other associations in the hypotheses were supported.
Although the association for Hypothesis HI D (preference for
avoiding premature convergence with evaluative skill) was
positive for non-original ideas (+.21), it was negative for origi-
nal ideas (-.33). Similarly, while the association for Hypoth-
esis H4 (ideational skill (quality) with evaluative skill) was
positive for evaluating original ideas (.74), it was negative for
evaluating non-original ideas (-.31). Accordingly in Figure 6,
Hypothesis H 1 D has been split into two components: a revised
Hypothesis H1D (-.33) and a new Hypothesis H1E (.21); and
Hypothesis H4 has been split into new Hypothesis H4A (.74)
and, H4B (-.31). Paths H3A and H4A were shown to be very
strong, associating ideational skill (quantity) with ideational
skill (quality/originality) (.80) and ideational skill (quality/origi-
nality) with evaluative skill (original ideas) (.74).
It is interesting that the active divergence attitude was asso-
ciated only with the (avoiding) premature convergence atti-
tude (HIA, .32), and not with any skills, contrary to the
prediction of H2. Thus, although this research supported
Basadur and Finkbeiner's (1985) suggestion that the ideational
attitude of preference for (avoiding) premature convergence
is necessary for the ideational attitude of preference for active
divergence, the usefulness of including the divergence attitude
in the overall causal model relationships is questionable. To
test this suggestion, an alternative model omitting this path
(H lA) was created. Figure 7 displays the results. The fit is not
as good (p
=
.075) as the model in Figure 6, in which that path
(H1A) was included (p
=
.498).
Several alternative models proved poorer than that of
Figure 6. For example, when the pathway H3B is dropped
from Figure 6, the model is rejected (p
=
.035). The results of
93
--------
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
FIGURE 7.
Results of Testing Alternative Model #1.
••• ,!?!.S!~'!!!!c:~r:r•••
>
.10
Ideational
Skill
(Qualityl .56'
Originality) \
\~
\0
_31·\"....
. \%
\~
..::1
NOT SIGNIFICANT
...........•.......
,
.07
Note:
"p
<
.05, **
P
<
.01
Chi-square with 24 degrees of freedom 34.55
(p
=
.075)
Goodness of fit index
=
.962
Adjusted goodness of fit index
=
.809
Root Mean Square Residual
=
1.16
testing this second alternative example are shown in Figure 8.
No other alternative model was found to test as well as the
model of Figure 6.
DISCUSSION
This study suggests how training in a complete creativity pro-
cess increases a manager's ideation and evaluation skills. The
training increases the acceptance of (preference for) avoiding
premature evaluation (convergence) of new solution ideas. This
probably increases the acceptance of (preference for) active
divergence - the free-wheeling generation of options without
judging their quality or analyzing their relevance. Precisely why
the preference for active divergence increases is unclear, but
increasing the preference for avoiding premature convergence
seems to encourage skill in ideational fluency (quantity of
solution ideas generated). It also appears to enhance the skill
of evaluation in accurately recognizing non-original solution
ideas. Meanwhile, increasing skill in ideational fluency (quan-
tity of solution ideas generated) apparently translates directly
94
Journal of Creative Behavior
FIGURE 8.
Results of Testing Alternative Model #2.
t
.32"
Note:
*p
<
.05
Chi-square with 24 degrees of freedom 37.94
(p
=
.035)
Goodness of fit index
=
.959
Adjusted goodness of fit index
=
.796
Root Mean Square Residual
=
1.215
into increased skills in both ideation (quality [originality
1
of
solution ideas) and evaluation (recognizing non-original solu-
tion ideas more accurately). Increases in evaluation skill
apparently emerge both directly from the attitude of prefer-
ring to avoid premature convergence and indirectly from
greater ideational skill in generating more solution ideas.
Greater ideational skill in generating higher-quality, more
original solution ideas appears to be directly associated with
greater evaluation skill in recognizing original, high-quality
ideas. This skill is also apparently required to offset the nega-
tive, but smaller, effects of a higher preference for avoiding
premature convergence on the skill of evaluating high-quality,
original ideas. Similarly, although increased ideational skill in
generating high-quality, original ideas has a negative effect on
the skill of evaluating lower-quality, non-original ideas, this is
apparently offset by the greater impact of higher ideational
skill in fluency (quantity) of ideas generated.
95
Understanding How Creative Thinking Skills. Attitudes and Behaviors Work Together
Thus, the key appears to be ideational skill in generating a
quantity of ideas. This skill is directly and strongly related to
ideational skill in quality of ideas generated (.80) and to evalu-
ation skill in recognizing lower-quality ideas (.56). In turn, the
skill of ideation (quality/originality) is strongly associated with
evaluation skill in recognizing more original, higher quality
ideas (.74). This key set of mechanisms tends to offset the
milder negative impacts on evaluative skills of the attitude of
avoiding premature convergence for recognizing higher-qual-
ity, more original ideas
(-.33)
and of the ideation skill (qual-
ity/originality) for recognizing lower-quality, less original ideas
(-.31).
Overall, with respect to evaluation skills, it appears that
the positive impacts of increasing ideational skills and attitudes
substantially outweigh the negative impacts.
One might speculate that the mechanism by which the pref-
erence for avoiding premature convergence may interfere with
evaluative skill in recognizing original high-quality ideas may
be a reluctance to judge ideas too quickly. The individual or
group might continue to consider options even as they are run-
ning out of time for making a sound judgment. Similarly, in-
creased skill in creating high-quality, original ideas may
interfere with identifying non-original, low-quality ideas: indi-
viduals who can improve ideas might be reluctant to drop a
low-quality idea in the hope that they can somehow salvage it.
The training emphasizes the positive approach to idea gen-
eration: suspending judgment in order to improve ideas. This
may best explain the lack of improvement after training in rec-
ognizing non-original ideas. For managers and others, improv-
ing skill in recognizing a few, original high-quality ideas might
be a far higher priority in solving problems than improving skill
in recognizing non-original, and thus less useful ideas, that they
are unlikely to use.
It also appears that increases in preference for avoiding pre-
mature convergence do not directly impact skill in ideational
quality, but rather
indirectly
through increasing skill in ide-
ational quantity. It is well known that attitudes do not always
translate directly into behaviors. Thus, the attitude of avoiding
premature convergence may have been related directly to ide-
ational quantity skill but only indirectly to ideational quality
skill. Osborn's prediction
(1963)
that "quantity breeds qual-
ity" (a primary rule for the brainstorming process) comes im-
mediately to mind as an example. This would explain the
indirect effect in our final causal model of the avoidance of
premature convergence attitude on ideational quality skill. The
96
CONCLUSIONS AND
FUTURE RESEARCH
97
Journal of Creative Behavior
intermediate step of increasing ideational quantity skill in or-
der to increase ideational quality skill fits the causal model
and lends credibility to Osborn's belief. It is also consistent
with Mednick's (1962) theory about remote associations be-
ing found only after obvious responses, and with laboratory
and field research supporting this belief as described earlier in
this paper.
Furthermore, perhaps the increased preference for active
divergence, which is apparently triggered by the increased pref-
erence for avoiding premature convergence,
indirectly
en-
hances the increased ideational skills in quantity/fluency and/
or quality/originality. This raises the possibility of a moderat-
ing relationship that could be explored in further research.
It appears that during the problem-solving phase of a com-
plete process of creativity; an attitude of preference for active
divergence is triggered by an attitude of preference for avoid-
ing premature evaluation of options as predicted by Basadur
and Finkbeiner (1985). However, the latter attitude - the
acceptance of the value of deferring judgment - is the more
powerful attitude as it is the only one significantly associated
with any of the divergent and convergent thinking skill vari-
ables. The former attitude - the preference for active diver-
gence - seems to merely accompany the development of
divergent and convergent behavioral skills. It also appears
that evaluative skills for non-original and original ideas may
be different skills that are related in different ways to ideational
attitudes and skills. This raises the possibility that various kinds
of evaluation tasks may work best under a different set of atti-
tudes and behaviors.
The causa! model of Figure 6 should now be tested as a
predictive model with a larger base size. Because of sample
size limitations in this research, the models are offered as pre-
liminary and tentative until this work is replicated with a larger
sample. Additional measures of personality traits and differ-
ent cognitive aptitudes and abilities associated with creativity
could be added to test for moderating effects on the model
and on training.
There are additional opportunities for extending this re-
search. For example, replication with different samples and
different tasks and perhaps additional measures would be one
major avenue. Another path would be to investigate the other
phases or stages of a complete process of creativity and inno-
vation such as the Simplex process of Figure 3.
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
REFERENCES
AMABILE, T. M.
s
GRYSKIEWICZ, N. D.
(1989).
The creative environment
scales: Work environment inventory.
Creativity Research Journal, 2,
231-253.
ANDREWS, F. M.
&
FARRIS, G.F.
(1972).
Time pressure and performance
of scientists and engineers: A five year panel study.
Organizational
Behavior and Human Performance,
8, 185-200.
BAKER, N. R., WINKOFSKY, E., LANGMEYER, L.,
s
SWEENEY, D. J.
(1976).
Idea generation:
A
procrustean bed of variables, hypotheses and
implications.
Cincinnati: College of Business Administration, University
of Cincinnati.
BASADUR, M. S.
(1974,
July-September). Think or sink.
The Deliberate
Methods Change Bulletin,
Procter & Gamble Management Systems
Division, Cincinnati, Ohio, 2.
BASADUR,
tvi.
S.
(1979).
Training in creative problem solving: Effects on
deferredjudgment and problem finding and solving in an industrial
research organization.
Unpublished doctoral dissertation. University of
Cincinnati, Cincinnati, Ohio.
BASADUR. M. S.
(1981).
Training in creative problem solving and measuring
improvement.
Engineering Digest,
27 (3), 59-61.
BASADUR, M. S.
(1982).
Research in creative problem solving training in
business and industry.
Proceedings of Creativity Week
4. Greensboro,
NC: Center for Creative Leadership.
BASADUR, M. S.
(1983).
Employee involvement creative problem-solving
workshop. In
Ford Education and Training Catalog,
Ford Education
and Personnel Research Department, Dearborn, MI, September, p.
115.
BASADUR, M. S.
(1992).
Managing Creativity: A Japanese model.
Academy
of Management Executive,
6
(2), 29-42.
BASADUR, M. S.
(1993).
impacts and outcomes of creativity in organizational
settings. In S. G. Isaksen,
r:
C. Murdock, R. L. Firestein, and D. J.
Treffinger (Eds.),
Nurturing and developing creativity: The emergence
of
a
discipline
(pp.
278-313).
Norwood, NJ: Ablex.
BASADUR, M. S.
(1994).
Managing the creative process in organizations. In
M. A. Runco (Ed.),
Problem finding, problem solving, and creativity.
Norwood. NJ: Ablex.
BASADUR,
t«.
S.
(1995).
The power of innovation.
London, UK: Pitman
Professional Publishing.
BASADUR, M. S.
(1997).
Organizational development interventions for
enhancing creativity in the workplace.
Journal of Creative Behavior,
37,59-72.
BASADUR, M. S.
&
FINKBEINER,
c.
T.
(1985).
Measuring preference for
ideation in creative problem solving training.
Journal of Applied
Behavioral Science,
27,37-49.
BASADUR, M. S., GRAEN, G. B., & GREEN, S. G.
(1982).
Training in creative
problem solving: Effects on ideation and problem finding in an applied
research organization.
Organizational Behavior and Human
Performance,
30, 41-70.
BASADUR, M. S., GRAEN, G.
B.,
& SCANDURA, T. A.
(1986).
Training effects
on attitudes toward divergent thinking among manufacturing engineers.
Journal of Applied PSYChology,
7/,612-617.
98
_
..
_-_._-_.
-
---
....
~-----------
._----------------_.---------------
Journal of Creative Behavior
BASADUR, M. S.
s
ROBINSON, S. J. (1993). The new creative thinking
skills needed for total quality management to become fact not just
philosophy.
American Behavioral Scientist,
37, 121-138.
BASADUR,
r:
S., WAKABAYASHI,
t«,
t:
GRAEN, G. B. (1990). Individual
problem solving styles and attitudes toward divergent thinking before
and after training.
Creativity Research Journal,
3, 22-32.
BOLLEN, K. A. (1989).
Structural equations with latent variables.
New
York: Wiley.
BOUCHARD, T. J. (1972). Training, motivation, and personality as
determinants of the effectiveness of brainstorming groups and
individuals.
Journal of Applied Psychology,
56, 324-331.
BOUCHARD, T. J.
£,
HARE, M. (1977). Size, performance and potential in
brainstorming groups.
Journal of Applied Psychology,
54, 51-55.
DUNNETTE, M. D. (1976). Aptitudes, abilities, and skills. In M. D. Dunnette
(Ed.),
Handbook of industrial and organizational psychology
(pp.
571-607). Chicago: Rand Mcnally.
DUNNETTE, M. D., CAMPBELL, J.,
£,
JAASTAD, K. (1963). The effects of
group participation on brainstorming effectiveness for two industrial
samples.
Journal of Applied Psychology,
47,30-37.
GOUGH, H. (1976). Personality and personality assessment. In M. D.
Dunnette (Ed.),
Handbook of industrial and organizational
psychology
(pp. 571-607). Chicago: Rand Mcnally.
GUILFORD, J. P. (1968).
Creativity, intelligence and their educational
implications.
San Diego: Knapp.
ISAKSEN, S. G.
£,
TREFFINGER, D. J. (1985).
Creative problem solving:
The basic course.
Buffalo, NY: Bearly.
JACKSON, P.W.
£,
MESSICK, S. (1964).
The person, the product and the
response: Conceptual problems in the assessment of creativity.
Princeton, ('!J: Educational Testing Service.
JORESKOG, K. G.
£,
SORBOM, D. (1993).
New features in LfSREL
8:
User's
reference guide.
Chicago: Scientific Software International.
JORESKOG, K. G.
s
SORBOM, D. (1989).
LfSREL
7
user's reference guide.
Stockholm, Sweden: Scientific Software.
KIRTON, M. (1976). Adaptors and innovators: A description and measure.
Journal of Applied Psychology,
61,622-629.
KRAUT, A.!. (1976). Developing managerial skills via modeling techniques:
Some positive research findings - a symposium.
Personnel Psychology,
29,325-328.
MAcKINNON, D. W. (1977). Foreword. In S. Parnes, A. Noller,
£,
A. Biondi
(Eds.),
Guide to creative action
(p. xiii). New York: Charles Scribner's
Sons.
MAcKINNON, D. W. (1962). The nature and nurture of the creative talent.
American Psychologist,
17,484-495.
MEDNICK, S. A. (1962). The associative basis of the creative process.
Psychological Review,
69, 220-232.
MURDOCK, M. C. ePUCCIO, G. J. (1993). A contextual organizer for
conducting creativity research. In S. G. Isaksen, M. C. Murdock, R. L.
Firestein, and D. J. Treffinger (Eds.),
Nurturing and developing cre-
ativity: The emergence of
a
discipline
(pp. 249-280). Norwood, NJ:
Ablex.
99
--------------------------------------------------
Understanding How Creative Thinking Skills, Attitudes and Behaviors Work Together
MYERS, I. B. (1962).
Myers-Briggs Type Indicator manual.
Princeton, NJ:
Educational Testing Service.
O'QUIN, K.
s
BESEMER, S. P. (1989). The development, reliability and
validity of the Revised Creative Product Semantic Scale.
Creativity
Research Journal,
2,
268-279.
OSBORN, A. F. (1963).
Applied imagination.
New York: Scribners.
PARNES, S. J., NOLLER, R. B.,
s
BIONDI, A. M. (1977).
Guide to creative
action.
New York: Scribner's.
ROE, A. (1976). Psychological approaches to creativity in science. In A.
Rothenberg and C. R. Hausman (Eds.),
The creativity question.
Durham, NC: Duke University.
RUNCO, M. A. (1991). The evaluative, valuative and divergent thinking
abilities of children.
Journal
of
Creative Behavior,
25, 311-319.
RUNCO, M. A. [,. ALBERT, R. S. (1985). The reliability and validity of
ideational originality in the divergent thinking of academically gifted and
nongifted children.
Educational and Psychological Measurement, 45,
483-501.
RUNCO, M. A. [,. BASADUR, M. S. (1993). Assessing ideational and evaluative
skills and creative styles and attitudes.
Creativity and Innovation
Management,
2 (3), 166-173.
RUNCO, M. A.
t:
CHAND, I. (1994). Problem finding, evaluative thinking,
and creativity. In M. A. Runco (Ed.),
Problem finding, problem solving,
and creativity
(pp. 40-76). Norwood, NJ: Ablex.
RUNCO, M. A. [,. SfViITH, W. R. (1992). Interpersonal and intrapersonal
evaluations of creative ideas.
Personality and Individual Differences,
13,295-302.
RUNCO, M.
,A,. [,.
VEGA, L. (1990). Evaluating the creativity of children's
ideas.
Journal
of
Social
Behavior and Personality,
5, 439-452.
SHAW, M. E. (1971).
Group dynamics: The psychology of small group
behavior.
New York: McGraw-HilI.
TAYLOR, D. W., BERRY, P.
c.,
e.
BLOCK, C. H. (1958). Does group
participation when using brainstorming facilitate or inhibit creative
thinking?
Administrative Science Quarterly,
3,23-47.
TORRANCE, E. P. (1974).
The Torrance Tests
of
Creative Thinking.
Bensenville: Personnel Press.
Min Basadur, Michael G. DeGroote School of Business, McMaster University.
Mark A. Runco, California State University at Fullerton. Luis A. Vega,
California State University at Bakersfield
100
