The Delphi Technique: A Case Study

Author(s): Frederick R. Cyphert and Walter L. Gant

Source: The Phi Delta Kappan, Vol. 52, No. 5 (Jan., 1971), pp. 272-273
Published by: Phi Delta Kappa International
Stable URL: http://www.jstor.org/stable/20372869 .
Accessed: 28/06/2014 19:05

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Phi Delta Kappa International is collaborating with JSTOR to digitize, preserve and extend access to The Phi
Delta Kappan.

http://www.jstor.org

This content downloaded from 193.142.30.55 on Sat, 28 Jun 2014 19:05:05 PM

All use subject to JSTOR Terms and Conditions
 The A Case
Delphi Technique: Study
By Frederick R. Cyphert and Walter L. Gant

Most schools of education and compromise, is often derived under tele was the motivation behind the
most universities operate
on the undue influence of certain psycho use, here described, of the Delphi
the apparent assumption that logical factors, such as
specious persua Technique by the School of Education
persons inside the organization control sion by the group member with the at the University of Virginia.
its destiny. While no one can deny the greatest supposed authority or even
influence of students, faculty, merely the loudest voice, an unwilling
and
Implementing the Survey
administration on the organization's ness to abandon publicly expressed
welfare and mission, opinions, and the bandwagon effect of
it is equally falla Sample selection involved a rather
cious to ignore the impact of persons majority opinion. In contrast, with the arbitrary definition of the "power
outside the organization. Perhaps one Delphi Technique an attempt is made structure" of the Commonwealth of
reason why schools of education have to overcome these factors by not Virginia in relation to the School of
not received the support they desire bringing the participants together in Education. Six major categories of
on either a moral or a fiscal level is one place and by not reporting individ persons were included: 1) the faculty
that they have not accurately assessed ual opinions.1 This eliminates commit of the School of Education and select
the judgments made about them by tee activity and replaces it with a ed student leaders from the graduate
others. If schools of education are to carefully designed program of sequen and undergraduate populations; and 2)
increase their effectiveness, they need tial interrogations (with question deans, the president's cabinet, and the
to give fresh attention to the clarifica naires) interspersed with information elected members of the university fac
tion and ordering of their goals. and opinion feedback. ulty senate. Off-campus elements in
For example,
a study of short-term
corporated into the sample included:
The Delphi Technique predictions was conducted with 20 3) educators, i.e., elementary and sec

graduate business students at the Uni ondary school teachers and administra
Traditionally, the method for versity of California in Los Angeles tors who held elective office in state
achieving consensus is a round-table during the fall of 1965. They were wide professional organizations and
discussion among individuals who ar asked to forecast gross national prod the deans of the state's major schools
rive at a group position. There are a uct, defense expenditures, and 14 oth of education; 4) organizational leaders,
number of objections to this proce er business indices. A general consen such as the officers of the Virginia
dure. The final position, usually a sus was reached after a series of four School Boards Association, the Vir
questionnaires. Forecasts were later ginia PTA, the State Council of Higher
checked with the actual happenings. Education, and the State Board of
tions of the world." A conceptual system is Although consensus varied greatly Education; 5) persons of influence in
one's mental structure comprised of: (a) from what occurred in certain the education
actually political circles, e.g.,
one's differentiated and interrelated con
indices, predictions were better on 13 committees of the Virginia House and
cepts of the environment, and (b) one's
rules and principles for defining concepts. of the 16 items than the predictions of Senate, U.S. senators and representa
Concepts and rules function to accomplish a 20 graduate students who were not tives, the governor, etc.; and 6) leading
range of goals and help the individual adapt
to the environment. Conceptual system is
involved in the process.2 newspaper editors and persons dealing
defined therefore as a cognitive structure The Institute of Government and with education in such groups as the
to evolve into a
which has the potential Public Affairs at the University of Virginia AFL-CIO, NAACP, and cham
more and more complex set of "rules" and
This construct has California in Los Angeles used the ber of commerce. It was decided to
interrelated concepts.
been shown to be remarkably relevant to Delphi Technique in an attempt to
studies of future perspective. See, specifical some useful on
An Investigation into the generate perspectives
ly, W. T. Weaver,
Between Level and changes in American education. The
Relationship Conceptual FREDERICK R. CYPHERT (1149,
Forecasting Future Events, doctoral disser authors of the study felt that the is
tation, Syracuse University, 1969; and J. S. University of Virginia Chapter)
Waldron, "An Investigation into the Rela procedure was regarded by almost all dean, School of Education, University
tionship Among Conceptual Level, Time of the participants as potentially use of Virginia. WALTER L. GANT
Delay of Information Feedback, and Perfor ful in educational planning at all lev
mance in the Delphi
(1059, University of Virginia Chapter)
Process, unpublished
els.3 Scientific assessment of the is director, Department of Adjustive
doctoral dissertation, Syracuse University,
1970. needs, desires, and opinions of clien Services, Norfolk City Schools.

272 PHIDELTA KAPPAN

This content downloaded from 193.142.30.55 on Sat, 28 Jun 2014 19:05:05 PM

All use subject to JSTOR Terms and Conditions
 seek perspective by including a seventh generally been used to predict what ^Marvin Adelson, "Planning Education
for the Future: Comments on a Pilot
group of selected teacher educators of will happen rather than to seek agree
Study," American Behavioral Scientist,
national reputation from across the ment concerning what should happen. March, 1967, pp. 1-31.
nation. The initial sample included a Fourth, consensus in this study was
total of 421 persons. No person was defined as the mode of the distribu
informed of the classification which tion of ratings on each goal, whereas > NEWYORK CITY'SEDUCATION
was the basis for his inclusion or of the other studies defined consensus as the AL CONSTRUCTIONFUND will fi
groups and other individuals involved interquartile range. However, several nance a new $20 million high school
in the survey. significant generalizations concerning on Wall Street by leasing the building's
Respondents were asked to suggest the Delphi Technique grew out of this air rights.
prime targets on which the School of experience: Under Fund sponsorship, a real
Education should concentrate its ener 1. Prospective participants must be estate developer builds a dual-occupan
gies and resources in the next decade. made to feel that their response is cy skyscraper. The bottom floors of
Questionnaire I was returned by 68% valid so that they will take part. the structure contain a school; the top
of the sample and the more than 750 2. The variation in agreement with floors are given over to apartments or
individual suggestions were reduced to the consensus rating on all the goals by office space. While the school and the
61 generic statements. individuals ranged from less than 20% rights to the air space above it are
Questionnaire it contained a ran to agreement with 100% of the con owned by the Fund, the tower is
dom listing of the 61 items with a sensus ratings. owned by the developer. The Fund
five-point grid for rating each on a 3. A bogus item which read, "Em finances the school with a bond issue,
priority continuum. Questionnaire III phasizing the production of doctoral leases its air rights to the developer,
differed from its predecessor only in graduates who can
improve the pro and uses the resulting rental income to
that it reported both the group con grams in schools of medicine, law, pay off the bonds.
census and the respondent's prior rat nursing, and engineering," was initially The New York City Board of Edu
ing for each item. Respondents were rated below average. However, when cation rents the school facilities for a
asked to re-rate all items in light of the the feedback was distorted to reflect a dollar a year until the bonds are
additional information concerning high ranking, the participants then retired, at which time title to the
group feeling. For all items where the rated the item considerably above av facilities is transferred to the city.
participant wished to remain outside erage, although it was not among the
the concensus, he was asked to state 10 highest-ranked targets. The hypoth > OVERTURNING TWO PROVI
his primary reason for so doing. esis that the technique can be used to SIONS of the California Education
Ratings on Questionnaire III were mold opinion as well as to collect it Code and a school regulation forbid
computed to determine any changes was
supported. ding on-campus distribution of un
which had occurred in the distribution 4. When respondents disagreed approved student publications, three
of priorities. Questionnaire IV was with the consensus rating of a goal, federal judges ruled that: "Student
then constructed, containing as before they tended to attribute the consensus communications cannot be prohibited
the mode for each item, the respon to a group of which they were not a because the school officials disagree
dents' prior ratings, and, in addition, a member. with what is being said or because they
list of the major "dissenting opinions." 5. Virtually all (99%) of the re think students should only be exposed
Final ratings were based upon the spondents' changes in oc to ideas which they approve."
opinion
respondents'
own values and a knowl curred on Questionnaire III which As defendants in the case, Campbell
edge of majority and minority views. informed them of the first "consen Union High School District officials
Final ranking of items was tabulated sus" reached by the group. With hind had argued that the restrictions were
on the basis of 262 returns (or 62% of sight, one can seriously question the justified by the students' immaturity
the original population). Of these, 194 need for going beyond Questionnaire and that published student criticism of
had changed ratings on Questionnaire III. The university family made the the faculty and administration under
IV. greatest change in ratings throughout mined their authority. However, the
the study, while the off-campus educa judges dismissed both arguments, say
Conclusions tors modified original ratings least.
Regarding the Technique ing that "the mere invocation of 'im
In conclusion, besides giving the maturity' does not serve to validate
There were several differences in satisfaction of planning the future general restrictions on students'
the way this study applied the Delphi with the assistance of data, this survey rights," and added that "the problems
Technique in comparison with its use made the influential persons in the presented by criticism (of the staff) do
in preceding studies. First, the tech commonwealth aware of the school's not justify the total prohibition of its
nique has usually been used with existence and gave them a vested course."
groups of 50 or fewer respondents interest in its future accomplishments. The ruling was a victory for two
rather than with the 400 involved in Campbell high school students who
the present survey. Second, most parti ?Olaf Helmer, Analysis of the Future: sought to distribute newspapers deal
cipants in prior studies have felt a The Delphi Method. Santa Monica, Calif.:
ing with "student activities, affairs,
Rand Corporation, 1967, p. 7.
greater degree of expertise in the field and opinions" and were told by the
2John Pfeiffer, New Look at Education.
being surveyed. Third, and perhaps Poughkeepsie, N.Y.: Odyssey Press, 1968,
school principal that school officials
most significant, the technique has pp. 152-57. first had to approve them.

January, 1971 273

This content downloaded from 193.142.30.55 on Sat, 28 Jun 2014 19:05:05 PM

All use subject to JSTOR Terms and Conditions