Designing and Instructional Design

Author(s): Gordon Rowland

Source: Educational Technology Research and Development, Vol. 41, No. 1 (1993), pp. 79-91
Published by: Springer
Stable URL: http://www.jstor.org/stable/30218372
Accessed: 18-09-2016 04:21 UTC

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 support@jstor.org.

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

Springer is collaborating with JSTOR to digitize, preserve and extend access to Educational Technology
Research and Development

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 Designing and Instructional Design

Gordon Rowland

What do we know about the process of O Instructional design (ID) is frequently traced
designing instruction? We have a large bodyto roots in areas such as systems theory and
of literature and numerous prescriptive mod-communications (e.g., Reiser, 1987), and repre-
els, yet it is not clear that designers actuallysentations of the process frequently reflect
operate as the literature and models suggest.those roots. But just how accurate are such
Other design fields, such as architecture andrepresentations when compared with what
engineering, have similar concerns, but havedesigners actually do in practice? We do not
acted upon those concerns by systematically really know. The body of literature describing
investigating design processes. Considering and prescribing ID processes is based primar-
the results of such studies may prove beneficialily on experts' opinions and recollections
to instructional designers in terms ofpromot- rather than on systematic investigation. This
ing critical analysis ofprocesses and decisionsliterature, on the whole, shares a view of ID as
and identifying questions and hypotheses fora deterministic, essentially rational and logical
research. In this article, results from numer-process, a set of procedures to be followed.
ous studies of design are synthesized to pro-Lack of success, either with the process or the
vide a basis for considering instructionalproduct, isblamed on poor implementationby
design as a type of designing rather than anthe designer(s) or on the primitive state of an
isolated phenomenon. emergent science of instructional design.
An alternative explanation is that the view
itself is inadequate. ID may be better charac-
terized as a creative process, based on intuition
as well as rationality, involving divergent as
well as convergent processes (e.g., Banathy,
1987; Earl, 1987), or as a dialogue rather than
a process of optimization (Tripp, 1991). Again,
we do not know, because we have relied on
what the experts say they do, or say what
others should do, and perhaps have missed
what instructional designers themselves actu-
ally do.
Concern that descriptions of instructional
design in the literature are discrepant with
practice is growing (e.g., Gayeski, 1991; Lewis

EIR&D, vol. 41, No.1, pp. 79-91 ISSN 1042-1629. 79

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 80 ETR&D, Vol. 41, No. 1

& Bjorkquist,1. the purpose or goal of

1992; designing
Pirolli & Gree
appears to mirror similar
2. relationships to other processes situ
enced in other design fields ov
3. factors that influence the design process
years. The theoretical bases of
architectural4. the nature
and of the designengineering
process.

been challenged on the grounds

to account for the complexities a
THE PURPOSE OF DESIGNING
of practice (see, for example,
Studies of designers engaged
Design is a goal-directed process in which the goal
designing have been carried ou
1978; Cross, is to conceive and realize some
1982; new thing (Cross,
Eastman, 19
Schon, 1983; 1982). Design has changed as societyStauffer,
Ullman, and tech-
1987), and, nology
as a haveconsequence,
changed. For example, Jones ne
and models (1970)
of describes
the a series of shifts in how designprocess
design
is carried
Instructional designout. Design as a craft
is gavelikely
way to to
similar design-by-drawing
studies. The as thefew
scope of products
studies
carried out became larger
to date and more have
complex. A third proved
have found method forclear
(1) designing, systems engineering,
similaritie
and other became
fields prevalent when
of the individual
design, was no and
which ID longer able to managediffers
practice all components of the subst
the common view of how instruction is or process. Across these three methods, the
should be designed (Kerr, 1983; Nelson, 1988; means and ends of designing and the attri-
Rowland, 1992). This article addresses the for- butes of design problems may have changed,
mer concern and examines the more general but the purpose has remained constant. De-
characteristics of design that are likely to hold sign is engaged in for the purpose of changing

for all types of designing, including ID. A gen- an existing situation into a preferred one
eral definition of design is offered, then elabo- (Simon, 1981). Whether that change--that is,
rated in a series of descriptive statements. The the goal of designing-involves a new com-
purpose is to consider what systematic exam- puter system or an illustration, an automobile

ination of different types of designing has re- or a building, a change in the skill of an indi-
vealed. To the extent that ID is a subset of vidual or a plan of action, the designer focuses

design, this may offer a new and fresh perspec- his or her efforts toward achieving a particular

tive on ID, and perhaps the means to get past end for the case at hand. Designing, therefore,
an outdated view. is a type of planning and results in an organ-
ized plan for achieving a special purpose.

A DEFINITION OF DESIGN To design is to plan and organize, to order and

Design is a disciplined inquiry engaged in for the
purpose of creating some new thing of practical
utility. It involves exploring an ill-defined situa-
tion, finding--as well as solving--a problem(s),
and specifying ways to effect change. Design is
carried out in numerous fields and will vary
depending on the designer and on the type of
thing that is designed. Designing requires a bal-
ance of mreason and intuition, an impetus to act,
and an ability to reflect on actions taken.
This general definition attempts to capture
many attributes of design cited in a variety of
literature. Individual attributes are stated and
described below in four loose categories:
relate and to control. In short, it embraces all
means that oppose disorder and accident. (Josef
Albers, in Lauer, 1985, p. 239)
The new thing that results from designing has
practical utility (Holt, Radcliffe, & Schoorl,
1985). There may be similarities between
designing and other processes, such as paint-
ing or composing music, but designing results
in something new or a new combination of
elements that serves a practical purpose. The
result of designing might be a poster to inform
people of an upcoming event, a machine com-
ponent to perform a specific function, an in-
structional lesson to teach something, an
environment in which certain activities can be

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 81

carried out, or a tool forCross (1982), reporting

carrying out on a
a number
partic- of
ular set of tasks. In each case,
studies the
of design, end
argues result
that design is quiteof
the design process isdifferent from science. While
a devised scientists focusor
product
artifact that is intended on
tothehave
problem, practical
on discovering theutility.
rule that is
operating, designers focus on the solution, on
achieving the desired result (Lawson, 1980).
THE RELATIONSHIP OF DESIGN The scientific method is employed to discover
TO OTHER PROCESSES the nature of what exists, while design meth-
ods are employed to invent things which do
Design is similar to some types of composing. not exist (Gregory, 1966). Science is essentially
Composing, either prose or music, is fre- analytic, and design is constructive (Gregory,
quently intended for a particular situation, set- 1966). And while the designer is constrained to
ting, audience, and medium. For example, a produce a practical product within a specific
composer of music may be called upon to time limit, scientists choose to suspend judg-
create an accompaniment for a scene in a mo- ment and decisions (Cross, 1982).
tion picture. He or she is constrained by the The need for artistry in designing is noted
theater setting, the intended audience, and the by numerous authors (e.g., Allen, 1988; Holt et
method of reproduction, and is required to al., 1985; Hubbell, in Marshall & Kifer, 1989).
support the writer's and director's goals for Frequently, design is characterized as a combi-
the scene. This may be quite different from the nation of artistry and technical skill. For exam-
activity of composing a concert piece with few ple, Jones (1970) comments that successful
externally imposed constraints. Once again, designers are able to combine reason with im-
the key is that the new creation is intended to agination and to be simultaneously creative
satisfy a specific practical purpose. and practical.
Design may be a science, or a combination of Others agree that a mix of rational and
science and art, or neither science nor art. Some creative processes are found in design (and, for
argue that a science of design is possible and that matter, in science and in art), but argue
represents an important goal. For example, that design is a separate area of knowledge,
Hubka and Eder (1987) call for "determining neither science nor art. Cross (1982) provides
and categorizing all regular phenomena of the a number of distinctions between science, hu-
systems to be designed, and of the design pro- manities, and design. These are summarized
cess" (p. 124). The assumption is that there are, in Table 1. For the purpose of this discussion,
in fact, regular phenomena that can be codified the "humanities" column may be taken to rep-
and predicted, and that prescriptions can thus resent art (as in "arts and humanities"). One
be made. could argue against some of these distinctions,

TABLE 1 - Distinctions between Science, Humanities, and Design

Science Humanities Design
Phenomenon of study Natural world Human experience Man-made world

Appropriate methods Controlled experiment Analogy, metaphor Modeling

of Study Classification Criticism Pattern formation
Analysis Evaluation Synthesis

Values Objectivity Subjectivity Practicality

Rationality Imagination Ingenuity
Neutrality Commitment Empathy
Concern for "truth" Concern for "justice" Concern for
"appropriateness"
Note: This table was constructed from distinctions made by Cross (1982).

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
82 ETR&D. Vol. 41. No. 1

but ings,
the overall but
point isthe goals a
a reasona
response
sign is a particular to the dif
discipline wi
areas of interest, methods,
buildings and
would v
lik
sign process procee
Design requires so
FACTORS couldINFLUENCE
THAT be carried out
THE DESIGN PROCESS
and provide a par
individual's use. Mo
Thedesign is a isdependenton
process member of a th
on what he orintended to serve
she designs. The ma
de
will be profit.
affected Thefactors,
by many social u
the designer'sconcern, and skill,
knowledge, design
social interaction.
ence; the design task; the workin F
and environment;
sires,and methods
values, a
and pre
ment. Hubka tomers
and Ederand(1987) des
how mu
these and otherfor the product
factors mu
affect such
quality of the tions of team
design, memb
the durati
ciency of the producers, manager
design process, and
designing. Ofwill also affect
particular the
importa
object(s) beingdesigning involves
designed and the p
the designer, such
i.e., as communicati
does he or she vi
lem as one well as
which any conflicts
requires design
regard to the et al., neither
latter, 1985). the pro
solution The
determine designer
that will f
designing
more importantclient. As Thomas
is how an
the indi
ceives the situation and his or he
the designer and the
respect to it expertise&
(Thomas to the pr
Carroll,
perspective isknowledge
primary in of goals
determ
tion.
problem solving The
will designer b
occur.
In terms of tionships
the betwee
design objects, H
Eder (1987) solutions.
found Thedesign
that the two "
edge to produce
substantially influenced a so
by what
the individual exist) and is
designed. likely
For examto
cess to designgoals" (p. 6). motor
an electric Clearlyc
and their interactio
pected to be different from that u
a poster or sign process
a building. Thisoccurs
appear
contradict Thomas and Carroll's
tention that the designer's persp
THE NATURE
primary importance. OF T
However, T
Carroll's argument focuses on th
the object, not necessarily
Designing the
involves ty
pr
For example, lem perceptions
the solving is not de
tha
must performtions in which
a practical task peop
and
ject currently to cross
exists to the gapthat
satisfy bet
trigger where
designing, they
while want
the wayto
t
proceeds maygap, they
differ must und
depending
the object is a problem
computer and find
system, a
a plan of 1987).
action. In The creation
particular, v
designing always
may involve
depend on theass
go
teria setgap,
by the i.e., understand
designer in resp
object's There
purpose. For are problem
example, a m
ing require designing.
plant and an art museum are

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 83

nature of the problem and

and one can
innever
the say relationship
with certainty which
solution
between the problem and is best.solution.
the One can only hope Design
for a satis-
factory
problems are ill-defined solution Reitman,
(e.g., that meets most or all of the
1965;
Suchman, 1987; Thomas & Carroll,
requirements. Neither the 1979). In
initial conditions
nor the
general, the problem and the mostmeans
appropriate and
to efficient
solve process
it
are unclear and must be to obtain a satisfactory
found by the solution are entirely
designer.
And when designing is clear. Also, the complexity
finished, there of the is
problem is
still
not the
some uncertainty as to the key distinction.
adequacy ofA mathematics
the solu- prob-
lem canare
tion. To illustrate this, here be very complex,
two but the initial condi-
problems:
1. An architect is asked tions ofto design
the problem, a singleasolution,
new and a
building. She has an idea limitedof
number
howof paths
to to that solution are
proceed
but cannot be certain that this will lead to an generally agreed upon.
effective design. She seeks to gain a rich under- Because design problems are ill-defined, all
standing of how the building is to be used, who information is never available to the designer;
will use it, what the owners' priorities are, and design problems are not susceptible toexhaus-
so on. She attempts to create a design that tive analysis (Cross, 1982). This is, perhaps, the
satisfies these requirements, but she is never main reason that designers tend to be solution-
entirely sure that all requirements were iden- focused rather than problem-focused (Law-
tified, which variations in design components son, 1980). They use conjecture to arrive at
best accommodate the constraints of the situa- solutions that contain the problem within
tion, or how stable requirements will be over manageable bounds.
time. The process and criteria were not clear at
the beginning, and the adequacy of her solu- The essential step is to recognize that nobody,
tion is not entirely clear at the end. least of all the chief designer, has, at the start, the
knowledge to say how the design will turn out,
2. A mathematics student is presented
or even what the problem really is-how it will
with a problem. He searches the problem state-
seem when, eventually, everyone's intuitions be-
ment for the variables that are involved, and come informed by the experience of having de-
for which values are given and which must be signed it. At the start, one's intuition is likely to
found. That is, he tries to understand what the be wrong, informed by what is, but not by what
is conjured into existence. (Jones, 1979, pp. 33-34)
problem is. He identifies the problem as being
a certain type and obtains an appropriate for-
mula(s), either from memory or from some It is important to recall that the ill-defined
external source. He applies the formula(s) and nature of a design problem is not necessarily a
derives the solution. The instructor reviews condition contained within the problem itself.
the student's solution and marks it correct, i.e., As stated earlier, the designer must hold the
a single correct solution is known to the in- perspective that designing is required. For ex-
structor. It is likely also that the instructor ample, it is possible for the building problem
knows some limited number of appropriate above to be treated as well-defined. A blue-
and efficient paths to that solution and can print for the building could be generated by a
assess the student's problem-solving process. computer using a set of formal rules or algo-
rithms. A single "best" solution would be ob-
Both examples involve problem solving, tained-"best" to the extent of the program's
but only the first can be called designing. The capability. This method would strike Jones and
mathematics problem is well-defined. It has a others as something other than designing. The
single correct solution that the instructor or reverse could occur with the mathematics
anyone else with the appropriate knowledge problem. The student could seek a new and
can obtain, given the problem statement. The elegant proof for the solution, i.e., he or she
initial conditions, and appropriate and effi- could treat the problem as ill-defined, chang-
cient paths to the solution can be identified up ing it to one which requires designing.
front. This is not the case with the building This points out that design problems must
problem. A nearly infinite number of different be found as well as solved (Lawson, 1980), and
solutions to this same problem are possible, since individuals interpret and understand

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 84 ETR&D, Vol. 41, No. 1

problems Darke (reported

differently (ain Lera, 1983) had similar
condition that is u
avoidable for results. She found that no
ill-defined attempts were made
problems), it is acc
rate to say that each
by her individual
subjects (architects) to analyze in detail solves
different problemtherather
requirements and their
thaninterrelationships
just generating
different solutionprior
to to an the
imagined solution.
same She argues problem.
that
requirements can only be worked out in the
For us, design is a type of problem-solving
context of a particular solution.
which the problem-solver views his/her proble
In the systems
or acts as though there isengineering
some view, problem
ill-definedness
understanding and problem solving
the goals, initial conditions, or are car-
allowable tran
formations. (Thomas ried out & Carroll,
sequentially, and preconceptions1979,
are p. 5; ital
added)
avoided. The solution concept is sought only
In designing, problem understanding and prob- after all necessary data have been obtained.
lem solving may be simultaneous or sequential This is not the case in "exploratory" design. In
processes. Systems engineering models (or, exploratory design, preconditions are sought
more generally, systematic methods) typically and subsequently challenged. Rather than at-
call for complete understanding of the prob- tempting to withhold judgment, the designer
lem prior to solution efforts (e.g., Pahl & Beitz, sees preconceptions as powerful, brings them
1984). A series of formal stages or steps are into the open and subjects them to analysis,
engaged in, one after the other, and a descrip- evaluation, and criticism (Robinson, 1986).
tion of the problem and a definition of goals Both views have strengths, and both have
are completed at the end of the "problem def- inherent limitations. Obtaining a "complete"
inition phase." Robinson (1986), Allen (1988), problem definition prior to solution activities
and others argue that this severely restricts the may save the designer from wasting time
designer's ability to understand the problem. working out what are eventually found to be
They feel that, with regard to design problems, poor solutions. But it may also "result in mis-
understanding is developed through efforts to directed efforts carried out with great compe-
solve the problem. The two processes are inter- tence" (Holt et al., 1985, p. 109). The designer
dependent and simultaneous or cyclical, and could perceive that the problem is adequately
goals are gradually uncovered in the context understood when that is not the case. On the

of solution attempts. As Lawson (1980) argues, other hand, an exploratory view, or what Holt
the problem and the solution emerge together; et al. refer to as "soft-systems analysis," may
one does not follow logically from the other, so result in a distorted definition of the task based

the process is thus dynamic and unpredict-
able.
Those taking the latter view-what Robin-
son (1986) refers to as "exploratory" design--
feel that dividing understanding and solving
on the differing world views of decision mak-
ers. The designer is faced with the additional
task of discovering the "real need" within this
context of personal perspectives and distorted
information.

into two phases results in conflicts being re-
solved prematurely, just so that a clear defini-
tion (in the worst case, any clear definition) can
be achieved. Studying designer's problem-
solving behavior, Hykin (reported in Lera,
1983) found evidence that supports this posi-
tion.
... exploration of alternatives led to clearer un-
derstandingof the problem, and... many import-
ant subproblems were not recognised or
understood, until several solutions had been at-
tempted or pursued to an advanced stage. There-
fore, properties and relationships could not be
established at the beginning of the process, nor
could the strategy be preselected and controlled.
(p. 136)
A basic task of designing is to convert informa-
tion in the form of requirements into information
in the form of specifications (Hubka & Eder,
1987). When a need for some new product is
felt, the designer's job is to identify what the
new product must do and to create something
that will satisfy those requirements. The de-
signer may not produce the product. More
typically, he or she completes a set of specifi-
cations for the product and passes the specifi-
cations on to someone else.
In order to make the transformation from
requirements to specifications, the designer
needs to have learned a language or system of
codes. To the information obtained from the

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 85

result.
situation, the designer adds an As "ordering
Rinderle (1986) points
prin- out, seeking
ciple" through which "the
andabstract patterns
then preserving these typesof
of relation-
user requirements are turned into
ships may lead concrete
to costly, overly large, and
patterns of an actual object" (Cross,
poorly integrated 1982,
designs. The p.
parts rather
224). This language maythan
bethe the
wholebasis for ex-
are optimized.
pression of "designerly ways of knowing"
Alternatively, many-to-one or one-to-many
(Cross, 1982). relationships among forms and functions--re-
Design specifications can be made that meet lationships that may not be found by the sys-
requirements separately or as a whole. Designers tematic methods described above--may be
often employ systematic methods, i.e., they characteristic of better designs. Rather than
follow a series of general steps or stages, such defining all problems prior to attempting to
as problem definition, analysis, design, devel- solve any of them, the designer may await the
opment, and evaluation. These methods typi- emergence of subproblems during prelimi-
cally involve solving problems by breaking nary solution attempts, and, by focusing on
them down into subproblems which can be subproblems as they occur, may find a more
understood and solved separately and then elegant solution to the whole. Again, the pro-
recombined. That is, rather than solving the cess implied is much more dynamic. Cycles of
total problem at once, the designer solves a set problem solving are derived dynamically
of related subproblems one at a time. He or she during the design process, vary in duration
balances resources and organizes the design and extent, and address subproblems when
process according to relationships between and in whatever form they present themselves
the subproblems (Churchman, 1968), and a (Thomas & Carroll, 1979). Neither the sub-
series of problem-solving cycles is implied. problems nor the means to address them are felt
to be completely specifiable at the beginning.
The overall attack on a design problem [is] often It is important to note that these differences
organized into relatively smaller and simpler in processes and consequences concern the use
"cycles": confrontations of portions of the total or non-use of systematic methods, not of a
problem.... Each cycle addresses a specific sub- "systems approach" per se. While methods
problem or set of subproblems constituent to the may prove limiting, thinking of problems and
overall design problem. (Thomas & Carroll, 1979,
pp. 9-10) solutions as elements of systems may be im-
portant to generating elegant and effective de-
signs. Still, as Kerr (1983) points out, designing
ultimately involves personal choices based on
Given the limits of short-term memory, this a sense of what is right. A systems approach is
is a natural and understandable behavior, as it not itself a mechanism for making these deci-
allows the designer to concentrate on one man- sions and may provide only a framework in
ageable task at a time (Newell & Simon, 1972). which the decisions can be made (Nelson,
However, the nature of subproblems and the 1988).
ways that relationships among them are re- The design process is a learning process. By
tained in the solution can vary significantly. As engaging in design, the designer discovers
stated above, systematic methods typically what he or she knows and does not know
call for (1) a period of analysis in which all about a problem and its solution. Filling that
subproblems are defined, say, for example, by gap is a learning process. In a sense, each
hierarchical decomposition of the problem, action taken generates an answer to a question
followed by (2) separate problem-solving epi- and enables the next question to be posed
sodes that address the individual sub- (Jones, 1979). Design can thus be thought to
problems, then finally by (3) recombination of as a knowledge-building cycle in which
occur
solutions to the subproblems into the total
the designer makes hypotheses (predictions
solution. From this approach, isomorphic or
relating the anticipated outcomes of each ac-
one-to-one relationships between forms (spec-
tion with features of the design product),
ifications for solution components) and func-
challenges them, and develops arguments to
tions (requirements of problem components)
support them (Robinson, 1986).

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 86 ETR&D, Vol. 41, No. 1

The design process is

"black box" and "glass box"one of
conceptions of the devisin
riencing a process ofMore
mind in psychology. rapid learning
recently, a concept
thing that does not yet exist by ex
of the designer as a self-organizing system has
interdependencies of problem and
become popular (as has a similar conception
new and the old. (Jones, 1979, p. 31
of mind, or rather, of what it means to know
Designing involves technical skills an
and learn-see Bednar, Cunningham, Duffy, &
ity, rational and intuitive though
Perry, 1991). Design expertise is thought to lie
(Hubka & Eder, 1987; Nadin & No
not only in knowledge and skill, but in the
Lawson, 1980). While various auth
designer's ability to reflect on his or her own
size different processes, some balan
actions. Rather than being a magician or com-
nique and creativity
puter, the designer must seems
be a self-organizing tobe ne
example, technical skill and ratio
system capable of controlling both rational
processes are and
required
creative processes, knowing whenanalyze
to to t
and to identify
requirements, whil
apply each and varying strategies and tactics
is important in coming
as the situation demands. up with i
new product. Several authors
Design is carried out as a reflective conversation
hav
the combination or balance needed.
with the materials of the situation (Schon, 1983).
The designer as a self-organizing system must
The truly creative scientist needs something of reflect on his or her actions. She or he takes
the artist's divergent thought to see new possibil-
certain actions, which are reflected back by the
ities while for his part the artist needs to be able
to apply the single-minded perseverance of the situation. The designer then assesses the con-
scientist to develop his ideas. What makes design sequences and implications of those actions
such a challenging task psychologically is the and decides on further actions. Each set of
very even balance of these two sets of mental
actions is an experiment that changes the situ-
skills that are needed to produce creative work.
(Lawson, 1980, p.116)
ation, frequently in ways that are un-
anticipated. The designer allows herself or
In designing, this difficulty [in solving problems] himself to be surprised and responds with
appears as the separation of the rational from the redirected actions.
intuitive, the practical from the creative. But the By "reflection-in-action," a designer there-
briefest study of how the most successful artists,
fore spins out a web of moves, each based on
engineers, etc. work and think suggests that they
"backtalk" from the situation. This conversa-
have one thing in common: they have found
ways of avoiding this split, of combining reason tion begins at the start of the design process,
with imagination, of being both creative and prac- when the designer develops an understanding
tical, of knowing when it's rational to be irra- of initial conditions, and continues through to
tional and when it's rational to work by
the proposed solution. It guides the designer
experience. To reconcile what seems to be oppo-
sites, to resolve contradictions, is the essence of as he or she reframes the problem, represent-
design. (Jones, 1979, p. 33) ing it via a chain of inferences, and as he or she
hypothesizes and tests solution strategies. It
While studies of designers' thought pro- helps the designer relate the current situation
cesses have been performed (e.g., Allen, 1988; to experience, i.e., to see the problem at hand
Lawson, 1980; Thomas & Carroll, 1979), the as similar to another problem encountered
specific combination or combinations of skills previously. Importantly, reflection-in-action
required for designing are not clear and at this does not bring the designer to fixed under-
point are matters of speculation. Characteriz- standing, but rather serves to lead inquiry for-
ing the process as a whole, Jones (1970) ward. Designing as reflection-in-action is thus
illustrates three possibilities through a series an intelligent activity, and one in which know-
of metaphors which match the historical ing how and knowing what (what Anderson,
trends described earlier. The field has moved 1976, refers to as procedural and declarative
from a concept of the designer as a magician, knowledge, respectively) cannot.be separated.
where the process depends on creativity, to a Allen (1988) makes a distinction that relates
concept of the designer as a computer, where to Jones' self-organizing system and to Schon's
logic is paramount. These are not unlike the "reflection-in-action." She speaks of situated

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 87

designing: "a view of activity which

does not reflect recognizes
practice. Both in prose and in
that the unexpected things in
graphic the
models, path
design are
is portrayed not
in ways
only obstacles to be overcome, but
that studies of design also
processes oppor-
(Le, studies of
tunities for new views onpractice)
the suggest are idealized. and
problem, For example,
can
produce new elements for the
the typical designer
view to one
in engineering remains use
in
in forming the next action" (p. is12).
which design described as a systematic and
A "situated action" is an action
rational in response
decision-making process that occurs
to the current situation being encountered
as a linear sequence of steps or phases (e.g.,
(Suchman, 1987). Its function is not,
Luzadder, 1986). however,
In contrast, some architects
refer to design
only to respond to a stimulus. as a creative process
It serves to shapewhich
the situation for subsequent
occurs as a decisions, and
series of cycles and depends it
more
does so in ways that are on
not entirely
opportunities predict-
than plans (Hubble, in Mar-
able. Thus, the environment (situation)
shall & Kifer, is dy-
1989). When the processes en-
namic, not static, and each
gagedaction is formed
in by these engineers in
and architects
terms of the effects of previous
when designing actions. This
are studied, these is
predomi-
what distinguishes a situated action
nantly "rational" or "creative"from
views essen-a
plan. A plan is fashioned tially
prior to,
disappear. rather
Different objects arethan
created,
during, a series of actions.
different Those who
goals and criteria rely
are set, and there-
fore different decisions
heavily on plans assume that the path that are made, but the un-
needs to be taken is predictable.
derlying processes arePlans
similar. As may
discussed
therefore be more consistent with
above, how a "rational"
well- or ill-defined the problem is,
view of designing, one that sees
and in particular theproblems
designer's perception as
of
well-defined, while thehow
concept of
well- or ill-defined situated
the problem is, may be
designing may be more consistent with
most important. Idealized a "cre-
views given in the
ative" view, one that sees problems
literature may be held, but atas ill-de-
this point it is not
fined. (Note that "plan" is influence
clear what usedthey here in
have on how
reference to the means of designing
designingproceeds. rather than
the ends. A design is itself a plan for change,
but the design process does not necessarily fol-
low a plan.) ID AS A TYPE OF DESIGNING

To some who appear to take a predomi-

nantly rational view (e.g., Martin, 1984), a re-
liance on creativity exists for controlling
rational processes. This might be taken as a
concession that not all aspects of the process
can be predetermined. Regardless of the view
taken, some level of situated designing, and of
reflection-in-action, is apparently necessary
for designers. In a sense, reflection-in-action
may describe the process of controlling situ-
ated actions (or, perhaps, of following a plan),
and the mind engaged in both is a self-organ-
izing system.
Design models reflect idealized views. The def-
inition and descriptive statements above were
based primarily on studies of the design pro-
cess. It is important to emphasize that these
studies represent only a small portion of de-
sign literature, and at times they contradict
what the main body of literature purports. As
was hypothesized earlier tobe the current state
regarding ID, much of the literature on design
Designing instances of instruction, or more
generally, planning and preparing to instruct,
can be considered a subset of designing, and
the defining characteristics described above
for all types of design appear to hold true for
ID. Instructional design is directed toward the
practical purpose of learning, i.e., the designer
seeks to create new instructional materials or
systems in which students learn. To do this, he
or she attempts to develop an understanding
of the conditions and the desired outcomes of
instruction, and to use this understanding in
specifying methods (Reigeluth, 1983).
As in other fields, instructional-design pro-
cesses vary, depending on what is being de-
signed. For example, different tools and
techniques may be employed, depending on
whether a new system or a performance im-
provement is involved or whether a single
piece of instructional material or an entire cur-
riculum is to be created.

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 88 ETR&D. Vol. 41. No. 1

Social interaction is
rational, and systematic, important
and the designer can
tional design. be likened
In to a order
technician or perhaps to
to an determ
ments and create effective methods, the engineer. This view underlies many system-
designer must work with the clients and spon- atic models (e.g., Dick & Carey, 1990) and can
sors of projects, subject-matter experts, pro- be taken to represent an underlying belief in
ducers and actors, teachers and learners. the prescriptive power of a science of instruc-
Instructional problems can be seen as well- tional design.
defined or ill-defined. That is, the designer
may interpret the initial information regarding Instructional design is [a] linking science-a
conditions and outcomes as accurate and com- body of knowledge that prescribes instructional
actions to optimize desired instructional out-
plete, and may see a simple path to an effective comes, such as achievement and affect.
instructional method. In contrast, another de-
(Reigeluth, 1983, p. 5)
signer might see the same situation as poorly
defined by that information and have consid- Instructional design can be defined as the science
erably less faith in solution ideas. ofcreating detailed specificationsfor the development,
Effort to understand an instructional prob- evaluation, and maintenance of situations whichfacil-

lem (or opportunity for learning) may precede itate the learning of both large and small units of
subject matter. (Richey, 1986, p. 9)
consideration of methods, or methods may be
considered simultaneously. In the former case,
Although in the minority, other individuals
the designer first seeks an understanding of describe instructional design as a creative pro-
the problem (determines conditions and out-
cess in which designing is driven by the recog-
comes) then proceeds to solve it (selects meth-
nition of opportunities and is carried out in
ods). In the latter case, ideas for instructional
iterative cycles. The designer interprets needs
methods may help the designer understand and identifies potential strategies in the con-
the problem. That is, if method X seems to be
text of the specific situation at hand. Standard
a good match, then the situation is more likely
rules or procedures are not employed, as they
to contain a problem of type X. In either case, are felt to be based on a reduction or oversim-
methods can be matched to conditions and
plification of factors affecting the instructional
outcomes in an isomorphic fashion or with system. The process is intuitive, creative, or
many-to-one or one-to-many relationships.
artistic, and emphasizes early attempts at so-
Lastly, instructional design clearly involves
lution rather than complete understanding
rational and creative or intuitive thought pro-
prior to solution attempts.
cesses. Less clear is how, when, and for what
purposes either type of process is or should be Design is a creative, disciplined, and decision-
oriented inquiry that aims to: (a) formulate and
emphasized. Also unclear is the extent to
clarify ideas and images of alternative desired
which instructional design represents a "re- states of a system; (b) prepare descriptions, rep-
flective conservation."
resentations or 'models' of the system; and (c)
devise a plan for the development and im-
TWO VIEWS OF INSTRUCTIONAL DESIGN plementation of the selected (most promising)
model. (Banathy, 1987, p. 89)

While authors agree that instructional design Thinking up a design for a course or lesson is an
involves a combination of rational and creativeintuitive, creative and logical process. Since it is
thought processes, they tend to accentuate one a creative process it will not run smoothly from
or the other extreme. Some individuals take a beginning to end.... An experienced designer
soon learns to "sense" when she or he is on the
"rational" view and describe instructional de-
right or wrong track. This is intuition at work.
sign as a technical process in which designing Richness of ideas, ingenuity in seeking a solution
is driven by known rules, principles, and pro- to the problem (the choice of optimum design),
cedures. The designer operates in a step-by- and originality come from the designer's creativ-
ity. The disciplined weighing, testing and selec-
step manner, extracting some standard types
tion or rejection of ideas is based on goal-directed
of information from the situation, clearly
logical thinking. Intuition, creativity, and logical
defining goals and objectives, and deriving thinking are at work in a designer's think tank.
an "optimal" design. The process is logical, (Earl, 1987, p. 32)

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 89

to believe
All four authors quoted above would likelythat conditions and outcomes do
concur that design is not simple and
notthat it determine methods, and con-
entirely
requires high-level cognitive processes. The
tinued to question the adequacy of solu-
nature of those processes, however, is in even
tions ques-
after development.
tion. Reigeluth and Richey place emphasis on
rationality, on the importance of* following
Expert instructional designers appeared to
rules and procedures. They feel that design
delay working out the details of solutions
principles accurately predict futurepending
phenom-a more complete understanding of
ena-in this case, learning-and therefore canbut generated solution possi-
the problem,
be used to prescribe instructional events.
bilities veryIn
early in the process. These so-
this sense, design is a science when lution
an under-
ideas seemed to constrain the pro-
standing of principles is sought, and isand
cess closer
to serve as a joint context in which
to engineering when those principles are understanding
problem ap- occurred. That is,
plied. The design practitioner wouldboth
function
problem and solution were matched
primarily as an "engineer." to integrated problem-solution patterns in
Banathy and Earl place more or equal em-
memory, some retrieved via specific case
phasis on creativity, suggesting the impor-
experiences.
tance of artistry and the subjective. Individuals
* Rather than
taking this view argue that prescriptions are responding yes or no to
whether the
not useful. Congruent with perspectives onproblem was instructional in
nature,
design in other fields (e.g., Jones, 1970; expert designers explored, and ap-
Schon,
1983), they feel that the phenomena peared to consider within the scope of "in-
with
structional
which designers work are so complex, involve design," a variety of problem
so many variables, and are so uncertain thatpossibilities. Many different
and solution
causal
the designer must treat each design as factors relating to the performance
a unique
of product
case, not a recurring event. The design individuals and the organization were
and the design process are bound toconsidered,
context. and a range of instructional
and non-instructional interventions were
The "artist" or creative instructional designer
sees rules and procedures as havingspecified.
limited
application. He or she practices a craft or art in
* "Scientific principles" of instructional
solving problems.
design may have served as heuristics for
deriving a solution or for evaluating pre-
SOME PRELIMINARY FINDINGS viously imagined solution ideas. It was rare
for a designer to make a clear prescription
These discrepancies in the literature between of method from a small set of known fac-
views of what ID is beg the question, How do tors. More common was a "rule of thumb"

ID processes actually occur? Kerr (1983), Nel- being used to select a type of solution or to
son (1988), and Rowland (1992) have asked evaluate the quality of a particular idea. In
this question and have found that, as suggested doing this selecting and evaluating,
above, the thought processes engaged in by "global" as well as "local" criteria were
designers in the act of designing instruction applied (i.e., a wide range of systemic fac-
are well predicted by neither the "rational" nor tors were considered).
"creative" views in the literature. Results,
however, do correspond to those from studies
* Expert processes were better characterized

of design processes in other fields. as situated actions taken in response to mo-

Some key results reported by Rowland ment-to-moment conditions than as prede-

(1992) include the following:

* Expert instructional designers appeared to
interpret and treat problems as ill-defined.
Designers in the study took a skeptical pos-
ture toward "given" information, seemed
termined steps. While a general plan was
evident (at least in retrospect), decisions on
how to proceed were made on an ad hoc
basis with respect to the goal of a solved
problem rather than with respect to a for-
mal plan for how to solve it.

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
90 ETR&D., Vol. 41, No. 1

These results match studies

examined of thought
the design pro-
proc
cesses in other fields, but contradict
engaged views
in designing i
instr
Nelson,
the literature on ID, 1988;
especially Rowland,
those 19
represent
tives
ing a purely rational to the deterministic,
perspective. Even what
designers have
considered to appeared (e.g., Banat
be a relatively simpl
1990). A as
problem was interpreted great deal moreThe
ill-defined. ef
notion of gaining tions is needed,
"complete" and a under
problem much
ing of to
standing before trying what ID processes i
solve-something
implied in most IDWe should notcontradicted
models-was expect that
by the almost translate
immediate directly from
consideration oth
of so
lutions. The we
yes/no limit our
decision of view
entry to into
whatID
processes, our own
also typical lens,
of ID i.e., todid
models, what
no
directed, closed
occur. Use of instructional-design perspec
principles
was not evident, and adherence
knowledge to a forma
between desig
plan (e.g., a the
sequence field
of of
steps instructional
to be taken) wa
not observed. important. 0
Of course, these results are from a single
study which used a single design problem an
Gordon Rowland is Assistant Profess
a small group of designers. Other studies tha
Corporate Communication at Ithac
test questions raised by this work are needed
For example, the utility of current ID model
could be tested in ways similar to Higgins and
Igoe's (1989) test of media selection models.
REFERENCES
Perhaps current ID models do not adequately
reflect or support design
Akin, processes,
O. (1978). How dobut do
archite
serve important pedagogical, communication
Latombe (Ed.), Artificial intelligence
and/or management functions.
ognition in computer-aided design
North-Holland Publishing Compan
Allen, C. (1988). Situated designin
master's thesis, Carnegie-Mellon Un
SUMMARY
burgh, PA.
Anderson, J. R. (1976). Language, memory, and
In this article, an thought.
attempt Hillsdale, NJ: Lawrencehas
ErIbaum. been made to
synthesize design Banathy, B. H. (1987). Instructional systems design.from
characteristics a num-
In R. M. Gagne(Ed.). Instructional technology: Foun-
ber of fields. The synthesis is offered as a start-
dations (pp.85-112). Hillsdale, NJ: Lawrence
ing point, a perspective
Erlbaum. from which to
re-examine views our ofSystems
Banathy, B. H. (1991). ID. design Significant
ofeducation: A over
lap between ID and other
journey to create the future.types of designing
Englewood Cliffs, NJ:
seems to exist, and Educational
the Technology Publications.
attributes described
above can be seen to hold D.,for
Bednar, A. K., Cunningham, Duffy, T. M., instructional
& Perry,
J. D. (1991). Theory into practice: How do we link?
designing. Preliminary research suggests thi
In G. J. Anglin (Ed.), Instructional technology: Past,
to be the case andpresent,
indicates that
and future (pp. 88-101). Englewood, CO: important
differences exist between
Libraries Unlimited. traditional views o
ID and actual Carroll, J. M. (1990).
practice. An overview of minimalist much
However, work
remains. instruction. Proceedings: 23rd Hawaii International
Conference on Systems Science (HICSS-23), January
Instructional technologists
2-6, 1990. IEEE Computer Society. have begun to
examine more carefully what
Churchman, C. W. (1968). The systems approach. other
New design
fields have York: Dell.
learned. Tripp (1991) explore
views of design asCross, N. (1982). Designerly ways of knowing. De-
optimization (Simon, 1981
sign Studies, 3(4), 221-227.
and design as dialogue (Schon, 1987), then
Dick, W., & Carey, L (1990). The systematic design of
relates those views to ID. Streibel
instruction (3rd ed.). Glenview, IL: Scott, Foresman.
(1991) con
siders situated actions in
Earl, T. (1987). The relation
art and craft of course design. New to the design
and use of York: Nichols.
instructional systems. Others hav

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms
 DESIGNING AND INSTRUCTIONAL DESIGN 91

Eastman, C. M. (1972). On thelished analysis

doctoral dissertation.
of Virginia Polytechnic
intuitive
design processes. In G. T. Moore (Ed.),
Institute and State Emerging
University, Blacksburg, VA.
methods in environmental Newell,
design A., & and
Simon, planning
H. A. (1972). Human problem
(pp.
solving. Englewood
21-37). Cambridge, MA: M.I.T. Press. Cliffs, NJ: Prentice-Hall.
Pahl, G., & Beitz,
Foz, A. T. K. (1973). Observations on W. (1984). Engineering
designer design. Ber-
behav-
ior. DMG-DRS Journal: Design lin: Springer-Verlag.
Research (Originally
and published
Meth- in Ger-
ods, 7(4), 320-323. man, 1977)
Pirolli, P. L.,
Gayeski, D. M. (1991). Software & Greeno, for
tools J. G. (1988). The problem
empower-
ing instructional developers.space of instructional design. InImprove-
Performance J. Psotka, L. D.
ment Quarterly, 4(2), 21-36. Massey, & S. A. Mutter (Eds.), Inteldhligent tutoring sys-
Gregory, S. A. (1966). Designtemsand(pp. 181-201).
theHillsdale,
design NJ: Lawrence Erlbaum.
method.
In S. A. Gregory (Ed.), The Reigeluth,
design C. M. method.
(Ed.). (1983). Instructional
London: design
Butterworths. theories and models: An overview of their current sta-
Hayes, J. R. (1987). The complete problem
tus. Hillsdale, solver.
NJ: Lawrence Erlbaum. Hills-
dale, NJ: Lawrence Erlbaum.
Reiser,(Originally published
R. A. (1987). Instructional technology: Ahistory.
in 1981 by The Franklin Institute, Philadelphia)
In R M. Gagn6 (Ed.), Instructional technology: Foun-
Higgins, N., & Igoe, A. (1989). An
dations (pp. 11-48). analysis
Hillsdale,NJ: of
LawrenceErlbaum.
intuitive and model-directed media-selection Reitman, R. R. (1965). Cognition and thought: An in-
decisions. Educational Technology Research & Devel- formation processing approach. New York: Wiley.
opment, 37(4), 55-64. Richey, R (1986). The theoretical and conceptual basis
Holt, J. E., Radcliffe, D. E, & Schoorl, D. (1985). of instructional design. New York: Nichols.
Design or problem solving: A critical choice for theRinderle, J. R. (1986). Function, form, fabrication
engineeringprofession. Design Studies, 6(2), 107-110. relations and decomposition strategies in design.
Hubka, V., & Eder, W. E. (1987). Ascientific approach Proceedings ASME Computers in Engineering Confer-
to engineering design. Design Studies, 8(3), 123-137. ence, July 1986, Chicago.
Jones, J. C. (1970). Design methods: Seeds of humanRobinson, J. W. (1986). Design as exploration. Design
futures. London: Wiley-Interscience. Studies, 7(2), 67-78.
Jones, J. C. (1979). Designing designing. Design Stud-Rowland, G. (1992). What do instructional designers
ies, 1(1), 31-35. actually do? An initial investigation of expert prac-
Kerr, S. T. (1983). Inside the black box: Making de- tice. Performance Improvement Quarterly,5(2),65-86.
sign decisions for instruction. British Journal ofEd-Schon, D. A. (1983). The reflective practitioner: How
ucational Technology, 14(1), 45-58. professionals think and act. New York: Basic Books.
Lauer, D. A. (1985). Design basics (2nd ed.). NewSchon, D. A. (1987). Educating the reflective practi-
York: Holt, Rinehart and Winston,. tioner: Toward a new design for teaching and learning
Lawson, B. (1980). How designers think. Westfield, NJ: in the professions. San Francisco: Jossey-Bass.
Eastview Editions. Simon, H. A. (1981). The sciences of the artificial (2nd
Lera, S. (1983). Synopses of some recent published ed.). Cambridge, MA: MIT Press.
studies of the design process and designer behavi-Streibel, M. J. (1991). Instructional plans and situated
our. Design Studies, 4(2), 133-140. learning: The challenge of Suchman's theory of
Lewis, T., & Bjorkquist, D. C. (1992). Needs assess- situated action for instructional designers and in-
ment: A critical reappraisel. Performance Improve- structional systems. In G. J. Anglin (Ed.), Instruc-
ment Quarterly, 5(4), 33-54. tional technology: Past, present, and future (pp.
Luzadder, W. J. (1986). Fundamentals of engineering 117-132). Englewood, CO: Libraries Unlimited.
drawing (9th ed.). Englewood Cliffs, NJ: Prentice-Suchman, L. A. (1987). Plans and situated actions: The
Hall. problem of human-machine communication. New
Marshall, P. (Producer), & Kifer, V. (Producer/Direc- York: Cambridge University Press.
tor). (1989). The art and vision of James Hubbell Thomas, J. C., & Carroll, J. M. (1979). The psycholog-
[Video]. San Diego: WKPBS-TV. ical study of design. Design Studies, 1(1), 5-11.
Martin, B. L. (1984). Internalizing instructional de- Tripp, S. D. (1991, February). Two theories ofdesign and
sign. Educational Technology, 24(5), 13-18. instructional design. Paper presented at the Annual
Nadin, M., & Novak, M. (1987). MIND: A design Meeting of AECT, Orlando, FL.
machine. In P. J. W. ten Hagen & T Tomiyama Ullman, D. G., Stauffer, L. A., & Dietterich, T. G.
(Eds.), Intelligent CAD systems I(pp. 146-171). New (1987, February). Preliminary results ofan experimen-
York: Springer-Verlag. tal study of the mechanical design process. Paper pre-
Nelson, W. A. (1988). Selection and utilization of prob- sented at the NSF Workshop on Design Theory and
lem information by instructional designers. Unpub- Methodology, Oakland, CA.

This content downloaded from 142.66.3.42 on Sun, 18 Sep 2016 04:21:33 UTC
All use subject to http://about.jstor.org/terms