Solving Creative Problems in Groups:
A Comparison of Techniques for Use in Advertising Classrooms
by
Nancy Mitchell
and
Mary Ann Danielson
For presentation at the AEJMC convention in Anaheim, CA
Aug. 9-13, 1996 ABSTRACT
Solving Creative Problems in Groups:
A Comparison of Techniques for Use in Advertising Classrooms
One underinvestigated aspect of advertising education is the
consideration of methods by which groups effectively solve creative problems.
Brainstorming has been popularly used by teams in classes to generate ideas;
yet, little research exists that compares types of small group structures for
problem solving. Additionally, creativity is not the central focus of most small
group research concerning problem solving. To bridge this gap, this paper
compares structured group (Nominal Group Technique), unstructured group, and
individual efforts at addressing a creative project as measured by quantity of
ideas generated, equal opportunity for participation, group satisfaction,
confidence in decision, and efficiency and effectiveness of the group. The
researchers conclude that the Nominal Group Technique, while most effective for
creative projects, will be enhanced when combined with the
elaboration/interaction of unstructured group activities. (137) Creativity.
Once scorned as being less valuable than rational, logical thinking, creativity
and creativity training are taking over business and industry. While some
futurists are predicting a creative revolution to supercede the agricultural,
industrial, and informational revolutions of the past (LaBarre, 1994a), most
individuals just acknowledge the importance of creativity, especially to today's
businesses. According to Day (1994), business' fascination with creativity and
initiative borders on obsessive; in fact, "unshackling our workers to allow
their creative juices to flow has been the ultimate objective of just about
everything from quality circles to reengineering" (p. 7). Because the success of
advertising ultimately depends on the output from a team involved in the
creative process, it is vital to teach students effective methods for group
problem solving for their future use in agencies and other corporate settings.
While corporate creativity traces its roots to the 1950s, it has
truly blossomed in the 1980s and 1990s. Recent surveys indicate that 25-30
percent of all organizations offer some form of creativity training (Solomon,
1990, p. 66; Voss, 1991, p. 37); according to Solomon (1990), that is an
increase of 540 percent from four years ago.
Concurrent with the increasing emphasis on creativity training is the
rise in the number and variety of creativity programs offered. Numerous authors
have written on the topic of creativity, several centers for creativity have
been created (e.g., Center for Creative Leadership and the Du Pont Center for
Creativity and Innovation), software has been designed, and programs, such as
the Creative Whack Pack, Pocket Innovator, and Six Thinking Hats, packaged for
corporate use. Regardless of which type of program is selected or utilized, the
main goal of creativity training is to learn new ways of interacting with the
"information-rich environment so as to structure it in such a way as to do
something with it" (Shephard, in Solomon, 1990, p. 66). Identification and
application of such efforts could significantly enhance the advertising process.
In promoting creativity and creative problem solving, most trainers
will admit that there is no longer "one best way" of approaching problems. What
is really needed is a
"tool chest" of ideas so that individuals "can apply the proper tool
at the proper time" (Shephard, in Solomon, 1990, p. 66) and in the proper form.
While most creativity research focuses on specific "tools" or techniques, very
few consider the (dis)advantages of individual versus group structures (or form)
for creative problem solving.
Group research has historically followed two paths: individualistic
and group-oriented approaches to small group research (Steiner, 1974). Adopting
either the sociologically-based group-oriented or the "Gestalt" psychological
approach, small group results would seem to indicate that small groups are not
mere collections of individuals (Lewin, 1951) and, therefore, differ from
individuals in how they solve problems.
Unfortunately, the creativity literature does not generally account
for these differences. While most training is conducted in groups or teams, the
types (structured versus unstructured, small versus large, mixed versus same
gendered teams) of small group structures are not considered. Conversely, most
small group research focuses on decision-making and problem-solving exercises
where creativity is not the central focus. To bridge the gap between creativity
training research and small group decision making, this paper compares small
group and individual efforts at addressing a creative project. Specifically,
this paper will explore the relationship between creativity and small group
research, propose research questions and hypotheses, and conclude with results
and recommendations for the incorporation of small group structures in
creativity training and practice.
Creativity training
Creative problem solving is one way to tap group members' thinking.
Creativity requires that individuals look at problems from a number of different
perspectives,
think in broad categories, and generate a variety of solutions. In
general, the purpose of creative thinking "is to develop novel and unusual
ideas" (Keeney, 1993, p. 52). By showing students in any area of advertising
how to solve problems creatively, instructors are encouraging individual
expression and hoping for happier, more cohesive, and more productive students
now and effective advertisers after graduation.
Creative training methods fall along a spectrum. On one end of the
spectrum are behaviorists who ignore creativity and simply examine what worked
for others in the past. Success with a particular strategy is taken as a
predictor of success in using that strategy in the future. On the other end of
the spectrum are the emerging ideas of Kamin and Muszynski (Hequet, 1992).
Kamin utilizes fables to stimulate creativity, whereas Muszynski uses music
(drums) as a metaphor for his exercises. Most creativity programs fall in the
middle of the range and tend to focus on brainstorming. While brainstorming can
be done both individually and in groups, it is expected that more and more
companies will be relying on group brainstorming (Rogers, in Solomon, 1990, p.
70). In fact, given the current matrix (i.e., Gore-Tex Corporation) and team
approaches used in today's agencies and corporations, it would seem that small
groups are and will continue to be the forum for creative problem solving.
One of the few studies conducted involving creative group problem
solving focused on Carter's strategies at the 1978 Camp David Summit (Hare &
Naveh, 1985). President Carter met with Egyptian President Anwar el-Sadat and
Israeli Prime Minister Menachem Begin in 1978 in order to forge a Middle East
peace agreement. Carter used both socio-emotional and task creativity to bring
the two parties to agreement. His strategies for a successful resolution were
dependent on both parties talking to one another (in verbal and/or written
form).
While the Hare and Naveh (1985) study highlights the importance of
communication in political problem solving, communication is equally important
in
other types of problem solving. LaBarre (1994b) stresses the
importance of communication in successfully managing employees of the "creative
revolution." Unfortunately, neither Hare and Naveh (1985) or La Barre (1994b)
offer any explanation or empirical support for types of group structure that
promote communication. While it is true that creative individuals may need a
"tool chest" of creativity-generating options, they may find that certain group
structures are better than others at facilitating communication and creativity.
Small group problem solving techniques
Small group problem solving can be structured or unstructured;
however, small group research seems to indicate that "structured techniques are
needed to ensure a solution's quality, acceptance, and implementation" (Frankel,
1987, p. 545). The majority of the research focuses on structured problem
solving, and the most common forms include brainstorming, the Delphi technique,
consensus decision making, and the Nominal Group Technique (NGT).
As described by O'Neil and Jackson (1983), NGT is a structured
activity facilitating group-based decision making. It is a group in name only,
hence, the "nominal" group designation. Group interaction, when it is allowed
to occur, is strictly controlled by the leader. Verbal interaction is limited,
to the extent possible, to leader-individual member dialogue. Two of the key
elements of the NGT are depersonalization (i.e., separating ideas from
personalities) and allowing equal participation in the discussion process
(Lowry, 1991, p. 21).
Developed by Delbecq and Van de Ven (1971), NGT is considered to be
one of the best structured techniques (Bartunek and Murninghan, 1984, in
Frankel, 1987,
p. 545) utilized in problem solving and decision making research.
While based on the "accepting ideas without valuing them" premise of
brainstorming groups, NGT's
unique characteristics distinguish it from both brainstorming and
other problem-solving groups. In its original form, NGT was characterized by
(a) individual work preceding group discussion, (b) round-robin reporting to
communicate ideas among the group members, (c) a period of unstructured group
discussion [in its modified form, group discussion is limited to clarification,
elaboration, and merging of common ideas], (d) individual polling of members to
converge on a specific solution, and (e) all individual and group work is
face-to-face (Hegedus & Rasmussen, 1986, p. 546). NGT, in its original form, has
been criticized for assuming that the problem statement and solutions are
clearly understood,allowing only one solution, and not providing a mechanism for
developing synergistic solutions (Frankel, 1987).
Because of these and other limitations, the Nominal Group Technique
(NGT) has been modified and supplemented by various authors. Hegedus and
Rasmussen (1986) utilized a modified NGT to study an evaluation problem. Thomas,
McDaniel, and Dooris (1989) combined NGT with decision analysis to analyze
strategic issues. Finally, Frankel (1987) combined NGT with multidimensional
scaling (MDS) to investigate solutions to ill-structured problems. While the
results were somewhat mixed, the research seems to conclude that nominal groups
can outperform other groups under certain circumstances. The exact circumstances
tend to depend on the nature and characteristics of the task. Hegedus and
Rasmussen (1986) concluded that the NGT may be useful but insufficient as a
decision procedure when groups were faced with complex, ill-structured,
multi-level decisions.
Some creativity training groups have utilized aspects of the NGT in
their programs. For example, a typical seminar on creative thinking conducted by
Mattimore Communications (a creativity consulting firm) requires participants to
begin the problem solving process a week or two before the group actually meets.
Participants generally receive a ten- to twenty-page briefing document that
explains a
specific problem. "This way," says Mattimore, "everyone has a chance
to work on the issue and we can hit the ground running" (Mattimore, in Voss,
1991, p. 38). Mattimore emphasizes the individual's work preceding the group
work aspect of the NGT; however, other aspects of the NGT are not followed. In
practice, creativity trainers utilize small group settings, borrow sporadically
from various brainstorming, consensus, and NGT techniques, but rarely, if ever,
base their choices on empirically tested relationships between small group
structure and problem solving techniques.
A comparison of techniques
Reviewing previous research that compares nominal groups to other
decision making styles, the following variables have been studied: quantity of
ideas generated, equal opportunity for participation, group satisfaction,
confidence in decision, efficiency and effectiveness of the group (Hegedus &
Rasmussen, 1986; O'Neil & Jackson, 1983). None of the previous research with
these variables studied creative problems. Therefore, this study examines the
relationship between creative problem solving and group decision making
techniques as measured by the outcome variables identified in the Hegedus and
Rasmussen (1986) and O'Neil and Jackson (1983) research.
Three common techniques identified by the decision making literature
include: nominal group technique (NGT), unstructured group activities (UGA) and
individual participation (IND). NGT groups are highly structured groups led by
a facilitator, who limits group interaction. UGA groups and individuals (IND)
function independent of any outside facilitation.
Drawing upon the findings of Hegedus and Rasmussen (1986) and O'Neil
and Jackson (1983), the following research questions and hypotheses are offered.
Volume. It is expected that NGT groups will generate the greatest
number of ideas
because the idea generation stage involves both the individual
creation of ideas and a round-robin style of sharing those ideas.
H1: The volume of ideas generated will be greatest for NGT groups.
Group process. NGT groups require active participation by all members.
Additionally, group members are allowed to seek clarification and elaboration.
Because of the forced participation nature of the NGT, it is expected that
participation of group members would be more equal, with members listening to
one another and seeking elaboration in NGT groups. The following three
hypotheses are proposed:
H2: The degree to which members listened to each other will be rated
higher by members of NGT groups than by members of UGA groups.
H3: The degree to which group members perceived participation was
equal will be rated higher by NGT group members than by UGA group
members.
H4: The degree to which group members asked one another to elaborate
will be rated higher by NGT group members than by UGA group members.
Satisfaction. One of the claimed advantages of NGT is high group
satisfaction with the process (O'Neil & Jackson, 1983). It is assumed that
because all individuals participate, satisfaction with individual contributions
and satisfaction with the group process will both be high. However, NGT does not
allow for social interaction between the members. Small group scholars (Brilhart
& Galanes, 1992; Forsyth, 1990) agree that interaction between group members is
an essential component that characterizes small groups. Interaction, according
to Bales' equilibrium model (1965), has both task and socio-emotional
components; the group spends the majority of its time attempting to balance the
task and socio-emotional needs of the group. Given that UGA groups are allowed
to interact and potentially develop both social and task dimensions of decision
making, it is expected that they could also develop high group satisfaction.
Because there are no clear indications of the effect of group
structure on satisfaction, the following two questions are proposed:
RQ 1: Will there be a significant difference in levels of individual
"satisfaction with the group decision" between NGT group members, UGA group
members, and individuals (INDs)?
RQ 2: Will there be a significant difference between NGT group
members, UGA group members, and INDs in the degree to which they are
satisfied with
the process by which the decision was reached?
Confidence. The sharing of ideas between group members should heighten
the degree of confidence perceived by individuals working in groups. However,
interaction between group members may highlight deficiencies in group thinking,
thereby reducing one's confidence in the group's decision. Individuals, on the
other hand, have no opportunity for reinforcement of ideas and theoretically
should be less confident of their decision. Some individuals who may prefer to
work alone and have a high internal locus of control should be more confident in
their decision. It is expected that differences may occur between the three
decision making structures; however, based on previous literature, there is no
clear indication as to the nature of those differences. Therefore, the following
two research questions are posed:
RQ3: Will there be a significant difference in level of "confidence
in the decision" between NGT group members, UGA group members, and
INDs?
RQ4: Will there be a significant difference in level of
"confidence about their individual participation in the project" between NGT
group members,
UGA group members, and INDs?
Efficiency. Unstructured groups, which are able to develop both task
and social dimensions, may find that social elements detract from their task
performance. UGA members may find their experiences more satisfying but less
efficient. Structured
groups (NGT) moderated by a facilitator and individuals (IND), who are
not distracted by other group members, should rate their performance as
efficient.
H5: NGT group members and INDs will rate themselves significantly
higher than UGA group members on efficiency.
Methods
Participants
Data were collected from 46 undergraduate students enrolled in
courses in two midwestern universities. Participants were divided into one of
three groups, NGT group (N=16), UGA group (N=20), or individuals (N=10). NGT
and UGA groups consisted of four members. While O'Neil and Jackson (1983) regard
a group of 8-10 persons as optimal for the NGT process, a summary of research
findings (Hegedus & Rasmussen, 1986) indicates as many positive findings for NGT
or unstructured (UGA) groups containing three or four members as groups composed
of seven or eight members. Size of the groups should not affect the research
outcomes. The participants included both men (N=21) and women (N=21). NGT groups
consisted of one all female group, one all male group, and two mixed gender
groups. UGA groups consisted of one all female group and four mixed gender
groups. Individuals were both male (N=4) and female (N=3).
The task
The task involved solving a creative problem. Specifically, the
participants were asked to generate possible themes/topics for a new magazine
targeting college and university students. Their objective was to generate a
list of potential themes and then select the one theme that they believed would
be best for the important initial issue of the magazine, keeping in mind the
need to attract potential advertisers. (Examples of the participants'
information sheets are included in Appendix A.)
Procedures
NGT Group. Each of the four NGT groups met with a researcher who
informed the members of the four-step procedure. After familiarizing the group
with the task, the researcher instructed members of the group to work
individually, for ten minutes, to generate potential themes (step one). They
were informed that after the ten-minute period they would be required to share
their themes with the rest of the group. In step
two, the individual group members shared their ideas, in a round-robin
style, until all ideas were exhausted. Ideas were clarified, elaborated, and
merged in step three. No evaluation of ideas was allowed. At the end of step
three, the group had a composite list of all suggestions generated. In the final
step, group members were polled. Based on a polling method developed by Cook
(1981), individuals listed their top seven choices from the composite list.
Group members were asked to divide 25 points among the choices and were told
that they could distribute the points in whatever manner they wanted; however,
every listing had to have at least one point. The theme with the most points
would be selected as the group's theme for the first issue of the magazine. The
researcher compiled the four individual lists, tallied the results, and
announced the selection to the group. Given the "group's" choice, the
individuals were asked to complete a one-page questionnaire (See Appendix B).
The time limit for the exercise was forty minutes.
UGA Group. Five UGA groups were formed and familiarized with the
task. Groups were told they had forty minutes to discuss and select the
magazine's theme. Once the group agreed that they understood the task and goal,
they were allowed to proceed with the decision making process in any manner they
chose. Two of the groups worked for forty minutes without a break; the other
three groups met for four ten-minute periods over the course of five days. At
the end of the forty-minute work period, the group members were asked to submit
their lists of suggestions and the one
selected theme. The final step of the group activity was the
completion of a group questionnaire (See Appendix B).
Individuals (IND). Like the NGT and UGA groups, individuals were
introduced to the task by the researcher, who then directed them to work
independently on the task. They were told they had a maximum of forty minutes
to complete the task. Eight individuals accomplished the task in a single period
of up to forty minutes; two others
worked on the task for ten minutes a day for four days. When the
individuals submitted their choices for themes, they were asked to complete a
questionnaire that was similar to the questionnaire completed by the NGT and UGA
groups. The wording on the individuals' questionnaires was modified to reflect
the individual's, not group's, work on the project (See Appendix C).
Questionnaires
The questionnaires completed by participants in the research project
were based on those used by Hegedus and Rasmussen (1986). The ten-item
questionnaire was divided into demographic and evaluative information. The first
two items deal with demographic information such as type of group (NGT, UGA, and
IND) and gender (female, male). The remaining eight items were designed to
measure the following factors: interaction processes such as listening to one
another, equality of participation, and elaboration of ideas; satisfaction with
the group's decision and with the individual's contribution to the group;
confidence in one's own thinking and in the group's decision; and efficiency of
the decision-making process as perceived by the participant. All eight items
were seven point Likert-type scales asking the individual to rate the degree to
which they believed something occurred (1=not at all and 7=a great deal). All
participants were given the opportunity to add additional comments or
observations about their participation in an open-ended question at the end of
the survey.
Individuals received questionnaires that were similar to those
completed by the groups. The wording, however, was modified on items three,
four, and five to reflect that the participants worked independently, instead of
in groups. Rather than omit these three items, the range of options was changed
to reflect eight options, with "0" reflecting a "not applicable" response. It
was expected that all individuals would circle "0" for items three, four, and
five.
Analysis
The research questions and hypotheses were tested using t-tests and
analysis of variance (ANOVA) to assess whether significant differences occurred
between the decision making structures. Post hoc analysis, using the
Student-Newman-Keuls and the Scheff tests, was conducted on all statistically
significant differences. Insufficient detail of analysis on item ten required
the additional use of the Duncan test. The reliability of the measures was
assessed using Cronbach's alpha coefficient. Additionally, a frequency count of
all items generated was conducted. Finally, responses to the open-ended question
were analyzed for emergent themes concerning the group and individual decision
making process.
Results
While 46 participants volunteered for the study, data were collected
from 42 participants [NGT (N=16), UGA (N=19), IND (N=7) ]. Three individuals and
one group member missed one or more sessions and were unable to complete the
questionnaire; therefore, their information was excluded for purposes of
analysis. Volume. In support of hypothesis 1, the number of unduplicated ideas
generated was greatest for the NGT group (IND, x=14; UGA, x=24.6; NGT, x=26.75).
It should be noted that the number of unduplicated ideas in one of the NGT
groups may not have accurately reflected the work of the group. If this outlier
was omitted, the mean of the
NGT groups would be considerably higher (x=33.33).
Group process. Contrary to expectations, NGT group members did not
rate listening to each other higher than UGA group members. In fact, UGA group
members' ratings were significantly higher than NGT group members (NGT, x=5.56;
UGA, x=6.32,
p< .01). While NGT and UGA group members did not differ significantly
in terms of participation, UGA group members rated themselves significantly
higher on seeking elaboration than did NGT group members (NGT, x=3.06; UGA,
x=4.89, p< .01). According to the criteria of listening, equal participation,
and elaboration, UGA group members appear to be more involved in the group
process.
Satisfaction. Findings indicate significant differences concerning the
levels of individual satisfaction with the group decision. Post hoc analysis
using the Student-Newman-Keuls revealed significant differences between NGT and
UGA group members (F [2, 41] = 4.05, p < .05), with individuals in UGA groups
rating satisfaction with the group decision highest. (See Table 1 for summary of
means.)
UGA group members also rated satisfaction with the decision making
process highest. Post hoc analysis detected significant differences between UGA
group members and both NGT group members and INDs (F [2, 41] = 10.57, p < .01).
NGT and IND ratings of satisfaction on this variable were approximately the
same, as shown in Table 1.
___________________
Insert Table 1 about here
___________________
Confidence. No significant differences emerged for either research
question concerning confidence in the decision or confidence about the
individual participation in the project. Consistent across both questions, UGA
members rated confidence
higher, but not significantly higher, than NGT group members or IND
participants. (See Table 1 for means.) While ratings were not statistically
significant, responses to the open-ended question on the questionnaire reveal
individual (IND) participants' concern with confidence. Individuals commented
that they would have preferred to "have had input other than my own" and were
"interested in hearing others [ideas] to make sure I wasn't on the wrong track."
Efficiency. The hypothesis concerning perceived efficiency revealed a
significant finding, but not in the predicted direction. Members of UGA groups
perceived their performances significantly higher in efficiency than NGT group
members or INDs
(F [2, 41] = 3.15, p < .05). Post hoc analysis using the Duncan test
indicated UGA group members were significantly different from individuals in
their ratings of efficiency. This result seems to support the concept that small
groups are not merely a collection of individuals; rather, groups are capable of
creating more than "the sum of its parts."
Additional responses. In one of the NGT groups, members commented on
the need to sell their ideas better and their concern about the clarification
step. Two members of one of the UGA groups provided comments demonstrating a
concern about limited contributions from a shy member. While a number of
individuals (IND) did supply comments, their comments tended to center around
one theme; they wanted to rely on input other than their own.
The reliability of the measures was assessed using Cronbach's alpha
coefficient. A reliability score of .83 was achieved.
Discussion
The results of this research add to the mixed findings, noted by
Frankel (1987), concerning the use of the Nominal Group Technique (NGT). Use of
the NGT on a creative problem-solving exercise resulted in a larger quantity of
ideas generated, but
generally reflected less evidence of group process and lower
satisfaction, confidence, and efficiency. Ultimately, the findings reinforce the
importance of the task as it relates to the various outcomes. Creative tasks,
according to our findings, are not analogous to ill-structured, multi-level
tasks. Therefore, our findings suggest that the NGT technique is useful, but not
optimal for creative decision making.
The individual generation and group sharing of ideas are facilitated
by the NGT technique; however, the technique falls short in the later stages
where individuals express preferences for greater interaction. Individuals, in
their follow-up comments, repeatedly expressed the desire for the sharing of
ideas, as did members of one NGT group.
Observations of UGA groups detected more "piggy backing" or
"springboarding" of ideas between group members. NGT group members and
individuals (IND) were not permitted this option. Given their open-ended
comments, this lack of open interaction and participation may explain lower
confidence, satisfaction, and perceived efficiency.
It appears that groups attempting creative problem solving would
benefit from a mixed NGT-UGA design. The initial steps of NGT, to include
individual idea creation and forced participation in a group setting, will
contribute to a higher number of generated ideas and will initiate group
interaction. Once ideas are generated and shared, ideas need to be elaborated or
clarified. It is clear that the UGA structure facilitates elaboration of ideas.
Additionally, the UGA structure allows for a more active form of interaction
between members and members' ideas. A combination of the NGT and UGA structures
should increase members' satisfaction with both the product and the process by
which the product was derived and increase perceptions of group efficiency. It
would appear that the advantages of both the NGT and UGA structures are enhanced
when combined.
These findings have implications for instructors of advertising and
anyone employing small group decision making structures. It is speculated that
the NGT technique, while advantageous for creative projects, will be enhanced
when combined with UGA elaboration/interaction. Creative trainers should provide
some initial structure and facilitation for their groups, but after idea sharing
(step two), trainers should minimize their role in the group. Adaptation of this
procedure is illustrated by building upon the Mattimore seminar discussed
earlier. This study provides empirical support for Mattimore's initial
individual preparation. It would appear that his participants really do "hit the
ground running." Once the group is convened and shares their prepared
materials, Mattimore would do better to minimize his role in the group. If the
group is allowed some latitude in how they proceed to solve the problem, results
of this study suggest that group members will be more effective (e.g., generate
more possibilities/solutions) and more satisfied with the results.
While these research findings are suggestive and offer some
implications for corporate applications, they are not conclusive.
Generalizability of findings is potentially limited by the small sample size,
the differences in time allotments, and our particular use of the Nominal Group
Technique. Ideally, future research should include a larger number of groups and
individuals; although in this study, significant results were achieved with as
few as four groups.
Future researchers also need to consider the amount of time allocated
and the distribution of that time. While all groups and individuals completed
the task in forty minutes (or less), Hegedus and Rasmussen (1986) warn that
forcing groups to finish within a limited time frame may produce results tht
confound the effects of the decision procedure. Of greater concern, however, is
the possibility that the distribution of time may bias the findings (i.e., favor
one decision making structure over another). Most group research is conducted in
one setting with a stated amount of time. These "block"
or "marathon" sessions may unnecessarily favor the NGT technique or
even individual efforts. The NGT was designed to be completed in one sitting.
Our results also indicate that individuals were not willing to work on this
creative task for the entire forty minutes; only two of the seven individuals
used the entire forty-minute period.
The group structure hindered by the use of a single session is the
unstructured group (UGA). Since interaction and development of both task and
social dimensions are important to group functioning, unstructured groups need a
different form of time allocation. Unstructured groups need time to allow
themselves to develop as groups. Therefore, it is suggested that UGA groups be
given the same amount of time, but distribute it over a number of days. This
distribution of time over a number of days may prompt negative effects such as
absences and gaps in data collection and a biasing of the research findings.
Finally, it is important to recognize that the Nominal Group
Technique has many modifications. Our particular use of the technique may have
influenced our results. If a less stringent version of the technique were used
(i.e., greater interaction between group members were allowed), the NGT group
members may have scored higher on a number of the dimensions.
Despite these limitations, it is important to remember that it is a
misconception that creativity if a gift that people either possess or they do
not (Hogarth, 1980). Rather, given an opportunity to develop and expand their
creative possibilities, students may respond to some creative problem-solving
techniques better than others. "It is one of the most important responsibilities
that managers [and teachers of the creative process] can have - to make that
process happen well" (Kao, in LaBarre, 1994a, p. 19). Advertising instructors
would be well served to consider the structure (NGT + UGA) by which they attempt
to foster creativity.
References
Brilhart, J. K., & Galanes, G. J. (1992). Effective group
discussion. Dubuque, IA: William C. Brown.
Cook, C. W. (1980). Nominal group methods enrich classroom
learning. Exchange: The Organizational Behavior Teaching Journal, 5, 33-36.
Day, C. R. (1994, May 2). Creativity is rarely complexity.
Industry Week, p. 7.
Delbecq, A. L., & Van de Van, A. H. (1971). A group process model
for problem
identification and program planning. Journal of Applied Behavioral
Science, 7,
466-492.
Forsyth, D. R. (1990). Group dynamics. Pacific Grove, CA:
Brooks/Cole Publishing.
Frankel, S. (1987). NGT + MDS: An adaptation of the nominal group
technique for ill-structured problems. The Journal of Applied Behavioral
Science, 23, 543-551.
Hare, A. P., & Naveh, D. (1985). Creative problem solving: Camp
David Summit, 1978. Small Group Behavior, 16, 123-138.
Hegedus, D. M., & Rasmussen, R. V. (1986). Task effectiveness and
interaction process of a modified nominal group technique in solving an
evaluation problem. Journal of Management, 12, 545-560.
Hequet, M. (1992). Creativity training gets creative. Training, 29,
41-46.
Hogarth, R. M. (1980). Judgement and Choice. New York: Wiley.
Keeney, R. L. (1993). Creativity in MS/OR: Value-focused
thinking-creativity directed toward decision making. Interfaces, 23, 62-67.
LaBarre, P. (1994a, May 16). The creative revolution. Industry
Week, pp. 12-19.
LaBarre, P. (1994b, May 16). The 'fourth-wave' worker: A five-step
plan for managing the employees of the "creative revolution." Industry Week, pp.
19-20.
Lewin, K. (1951). Field theory in social science. New York: Harper.
Lowry, P. E. (1991). The assessment center: Reducing interassessor
influence.
Public Personnel Managment, 20, 19-26.
O'Neil, M. J., & Jackson, L. (1983). Nominal group technique: A
process for initiating curriculum development in higher education. Studies in
Higher Education, 8,
129-138.
Solomon, C. M. (1990). What an idea: Creativity training. Personnel
Journal, 69,
64-71.
Steiner, I. D. (1974). Whatever happened to the group in social
psychology? Journal of Experimental Social Psychology, 10, 94-108.
Thomas, J. B., McDaniel, R. R., & Dooris, M. J. (1989). Strategic
issue analysis:
NGT + decision analysis for resolving strategic issues. The Journal
of Applied
Behavioral Science, 25, 189-200.
Voss, B. (1991, July). What's the big idea? Sales and Marketing
Management, 143, 36-41.
Table 1
Mean Ratings of Decision Making Structures
Decision Making Structure
Variables NGT UGA IND
Satisfaction 5.44* 6.37* 5.95
Satisfaction / process 4.75** 6.21** 4.71**
Confidence / group decision 5.56 5.95 5.57
Confidence / ind. participation 5.63 6.11 5.71
Efficiency 5.50 6.11* 5.14*
_____________________________________________________________
Note. Responses ranged from 1 (not at all) to 7 (a great deal). Post
hoc analysis using Student-Newman-Keuls, Scheffe, and Duncan tests denote
significance between groups. Significance denoted by asterisks: (*) = p < .05,
(**) = p < .01.
H1: The volume of ideas generated will be greatest for NGT groups.
GROUP PROCESS
H2: The degree to which members listened to each other will be rated
higher by members of NGT groups than by members of UGA groups.
H3: The degree to which group members perceived participation was
equal will be rated higher by NGT group members than by UGA group
members.
H4: The degree to which group members asked one another to elaborate
will be rated higher by NGT group members than by UGA group members.
SATISFACTION
RQ 1: Will there be a significant difference in levels of individual
"satisfaction with the group decision" between NGT group members, UGA group
members, and individuals (INDs)?
RQ 2: Will there be a significant difference between NGT group
members, UGA group members, and INDs in the degree to which they are
satisfied with the process by which the decision was reached?
CONFIDENCE
RQ3: Will there be a significant difference in level of "confidence
in the decision" between NGT group members, UGA group members, and INDs?
RQ4: Will there be a significant difference in level of "confidence
about their individual participation in the project" between NGT group
members, UGA group members, and INDs?
EFFICIENCY
H5: NGT group members and INDs will rate themselves significantly
higher than UGA group members on efficiency.
H1: The volume of ideas generated will be greatest for NGT groups.
|
