Brenda Dervin
Department of Communication
205 Derby Hall
154 N. Oval Mall
Ohio State University
Columbus, Ohio 43210

614-292-3400 work		
614-442-0721 home		
614-292-2055 fax

This is the originally submitted version of the chapter published as
follows: Dervin, Brenda (1992).  From the mind's eye of the user: the
sense-making qualitative-quantitative methodology.  In Jack D. Glazier and
Ronald R. Powell, Qualitative Research in Information Management
(Libraries Unlimited, 1992, pp. 61-84.)

	The purpose of this chapter is to set forth the sense-making
methodological approach which has been developed since 1972 in a
programmatic research effort specifically focused on developing
alternative approaches to the study of human use of information
 and information systems.  In terms of research genres as identified in
the current literature, sense-making has been used to study the needs,
images, and satisfactions of users and potential users of
information/communication systems-- in short, what use rs want from
systems, what they get, and what they think about them. 1
	In the first section of this chapter, the underlying assumptions
and theoretic foundations of the methodology are set forth.  This is
followed by a description of the methodology and its derivative methods. 
A third section presents exemplars of how th e approach has been used in
several different studies of users and potential users of
information/communication systems.  In a final section, the approach is
positioned in the context of the current "debates" in the social sciences
generally and specific ally in terms of the qualitative versus
quantitative research distinction. 
	Since the focus of this book is on qualitative research, it is
important at this point to note that the approach being described in this
chapter is a qualitative approach.  But, it is also a quantitative
approach.  It can in some senses be said to stand between methodological
cracks in that it refuses to be placed at either one end or another of the
many polarities that form the current contests in the social sciences: for
example, qualitative vs quantitative, administrative vs critical,
theoretic vs app lied, structuralist vs individualist.  These matters will
be discussed in greater detail in the final section of this chapter. 

	The term sense-making has come to be used to refer to a theoretic
net, a set of assumptions and propositions, and a set of methods which
have been developed to study the making of sense that people do in their
everyday experiences.  Some people call it a theory, others a set of
methods, others a methodology, others a body of findings.  In the most
general sense, it is all of these. 
	It is, first, and foremost, a set of meta-theoretic assumptions
and propositions about the nature of information, the nature of human use
of information, and the nature of human communicating.  Some of the
assumptions of sense-making are properly seen as
 axiomatic, in that they are taken as given.  Others are derived
deductively.  Others are propositions that have received empirical
	The assumptions and propositions of sense-making taken together
provide methodological guidance -- for framing research questions, for
collecting data, and for charting analyses.  Derived from these are a set
of methods, particularly methods for intervi ewing humans about their
experiences.  Sometimes, therefore, one sees sense-making referred to as a
theory of conducting interviews about sense-making.  It is that as well. 
	In essence, then, the term sense-making refers to a coherent set
of theoretically-derived methods for studying human sense-making.  The
coherence is of the kind where it may be said that method is a residual of
theoretic effort -- it falls out from the c onceptual frame which defines
it. This is an important point because, in fact, all method is residual of
theoretic effort but the difficulty is that for most use of methods to
study human behavior, particularly human information-using behavior, the
cohere nce among assumption, proposition, and method is hidden.  Certain
things are assumed to be true and entire enterprises for studying human
use of information systems are derived as if there were no other
approaches possible.  The best way to develop this p oint is to extract at
a very high level of abstraction the assumptions and propositions which
guide most research on human use of information and information systems
and to contrast these with the assumptions and propositions which guide
	It is useful to start by delineating the core assumption on which
sense-making rests -- the assumption of discontinuity.2 This assumption
purposes that discontinuity is a fundamental aspect of reality.  It is
assumed that there are discontinuities in al l existence -- between
entities (living and otherwise), between times and between spaces.  It is
assumed that this discontinuity condition exists between reality and human
sensors, between human sensors and the mind, between the mind and tongue,
between t he tongue and message created, between message created and
channel, between human at time one and human at time two, between home one
at time one and human two at time one, between human and culture, between
human and institution, between institution and institution, between nation
and nation, and so on.  Discontinuity is an assumed "constant" of nature
generally and the human condition specifically. 
	While arguments can be offered for the utility of invoking the
discontinuity assumption for many realms of study, our concern here is its
utility for the study of human use of information and information systems. 
Sense-making assumes that the discontin uity assumption is an important
one to invoke in the study of human information use for those occasions
when you want to know about behavior that is internally-controlled.  One
can propose many research questions which do not require the discontinuity
ass umption for they pertain to questions relating to information when
conceptualized as something that exists apart from human considerations. 
For example, if one sets standards of accuracy for answering questions at
the reference desk, then one can examin e how many questions are answered
accurately by comparing question-in and answer-out to some externally
defined standard.  Or, if one wants to know what elements of a current
information system's data base (collection, etc.) are used, one can
examine the
 number of times each element is somehow activated (checked out,
	However, many, perhaps most, of the questions that concern us in
information management, design, and practise do involve human actors.  How
can we design data bases so they will be maximally used?  What services
should we offer?  How satisfied are our us ers?  Why don't some potential
users use us?  How can we capitalize on the flexibilities that new
technologies allow rather than merely using them, as we are, to do what we
now do in greater quantities, from further distances, and at faster
	What follows below is a series of contrasts between the
assumptions of traditional research approaches and the assumptions of
sense-making, all of which are derived from the core discontinuity
premise. These contrasts will lay out a logic which explains why the
discontinuity assumption is seen as a required assumption for much
information research.
	It is also important to point out at this juncture that the
assumptions reviewed below pertain not just to research but also to
information management, design, and practice.  What sense-making has tried
to do in its development over the past 17 years is break with the
undergirding assumptions that guide our current system designs which, in
turn, mandate our research. 
	Fundamental to the specific application of sense-making to the
study of human use of information and information systems is the way in
which information is conceptualized.  Drawing from the discontinuity
assumption, information is conceptualized as tha t sense created at a
specific moment in time-space by one or more humans.  Information is not
seen as something that exists apart from human behaviorial activity. 
Because there is no direct observation of reality, all observations result
from an applica tion of energy from humans in one or more forms. 
	This is not to say that sense-making takes a radical
constructivist, sometimes called the post-modern, position, assuming that
there is no "order" out there or that there are no tools humans can use to
arrive at more comprehensive and more stable picture s of that reality. 
Rather, it assumes, first, that whatever order is out there is itself
potentially discontinuous from time to time, space to space.  And, it
assumes that whatever order is out there is not directly accessible by
human observers whose ob servations are constrained by time, space, and
species as well as personal capabilities.  Further, it assumes that humans
do not have available to them an external standard to which they can turn
for an assessment of their "truth" either in an absolute or
 even a relative sense. 
	The "standards" humans use for personal as well as collective
conduct are created in and are themselves constructed, created in
interaction. From a sense-making perspective, the using of a standard is
itself a constructing.  One human may wish to judge
 a moment of information use by a standard he/she calls accuracy while
another may wish to judge by expediency or familiarity or comfort. 
Further, the order that humans live within can not be seen as given.  It
is made and humans by continuing dialogue and sharing of personal
observations do arrive at always limited but more stable observations. 
	What is important about this is what one proposes to study based
on one or another set of assumptions about information.  If one assumes
that information has an existence apart from human constructing, one
focuses exclusively on transmission questions (e .g. How much information
did someone get?  Was the information they got accurate?  What can we do
to be sure people get more accurate information?) rather than on
construction questions (e.g. What strategy did that individual apply that
lead him/her to ca ll that information accurate?  What strategy did he/she
apply that lead to rejecting information that you or I might call
accurate? How can we design systems that allow people to apply the
criteria they want to their information searches?)
	The point here is not to propose that there are no situations
under which accuracy questions are ever relevant.  Rather, it is to
propose that if one assumes that information has existence apart from
human observations and has the capacity to provide com plete instruction
(i.e. continunuity) then one proceeds to study human information use only
in that sub-set of human conditions to which continuity applies. 
	Directly derived from the discussion above is another sense-making
assumption:  that human use of information and information systems needs
to be studied from the perspective of the actor, not from the perspective
of the observer.  Almost all our curren t research applies an observer
perspective.  We ask users questions which start from our worlds, not
theirs:  What of the things we can do would you like us to do?  What of
the things we now offer do you use?  Do you like us?  Which of the things
we do do
 you like?  Are we convenient for you?  How much of what we have is good
for you?  Would you use this service if we provided it?  Are your skills
sufficient to use us?  And, so on. 
	While some of these questions are more "user-oriented" than
others, all start with a system microscope.  They are predicated on the
idea that the system is the essential order and the person/user bends to
it rather than the other way around.  When one pr esents users with a long
list of services and has them check off which ones they want, one has
constructed a world for users.  The extrapolation from the data to
practice appears straightforward but the examples of the failures of
information systems base d on such input are legion.  The difficulty is
that the data tell us nothing about humans and what is real to them and
don't show us how people manage to get utility out of systems which
systems don't even predict, or how what looks like a failure from th e
system's perspective is actually a success when seen from the human's eye. 
The data don't help us understand why a service people said they wanted
goes unused, or why as communication technologies spread we appear to be
creating a more demarcated worl d of communication haves vs have-nots. 
The data don't tell us where we might move our system if we are to really
serve people on their terms. 

	Sense-making focuses on behavior.  As such it assumes that the
important things that can be learned about human use of information and
information systems must be conceptualized as behaviors: the step-takings
that human beings undertake to construct sens e of their worlds.  These
step-takings, or communicatings, involve both internal behaviors (e.g. 
comparings, categorizings, likings, dislikings, polarizings,
stereotypings, etc.) and external behaviors (e.g. shoutings, ignorings,
agreeings, disagreeings,
 attendings, listenings, etc.). 
	While almost all social science-guided research now professes a
belief in the power of process views of human behavior in fact little
research implements such a view.  Usually, the focus is on states and
entities rather than process and behaviorial strat egies and tactics.  For
example, the typical study will ask who uses an information system and
formulate an answer in terms of who the user is, what the user has access
to, how connected the user is to other users, what skills the user has,
and so on.  Th e typical study does not ask what constructed views lead a
person to reach out to an information system.  In fact, the constructed
view is assumed as a constant -- a state of information need.  The
qualities of this state of need are not explored because they are actually
not assumed to exist.  In a monolithic view of information use as
transmission, the state of need is necessarily assumed to also exist
	The use of state assumptions has numerous consequences for the
conduct of research.  Typically, for example, information research
attempts to predict and explain human use of information and systems based
on across time-space formulations rather than tim e-space bound
formulations.  We focus, for example, on levels of analysis (e.g.
interpersonal information exchanges vs mass media information exchanges)
as if they ought to explain differences in behavior.  Or, we assume
topical contexts as we define them
 ought to explain the difference (e.g. health information vs political
information vs science information).  Or, we focus on across time-space
characteristics of the person (e.g. demography, personality, skills,
resources) as if they ought to explain beh avior.  In fact, all of these
have explained very little. 
	This has had enormous consequences even for what we think is
possible in understanding human information use.  It has been frequently
challenged, for example, that individual behavior vis-a-vis information is
too chaotic to expect much from systematic st udy.  Proof here is the
frequently low variances accounted for in attempts to predict anything
more complicated than habitual patterns of channel use.  A result of this
assumed chaos is that we find voices calling for two radically different
kinds of retr eat -- one is the retreat to qualitative and highly
contextualized understandings of individuals; the other is the retreat
from individual to structural understandings. 
	It is not the purpose of this chapter to suggest that qualitative
approaches or structural understandings are not useful.  Rather, what is
being suggested is that these responses when framed as ways of handling
the chaos of individuality are conseque nces of the application of state
assumptions.  If human information use can be best understood using
process assumptions then attempts to do so using state or entity
assumptions will yield limited results.  These limited results will
suggest that individu al behavior is at worst chaotic and at best
capricious and recalcitrant.  In fact, though, it may be quite systematic
if studied from a process perspective.
	The difficulty we have is lack of examples which allow us to
envision the possibility.  Sense-making assumes there is something
systematic about individual behavior when the individual is
reconceptualized not as an entity but as an entity-behaving at a m oment
in time-space.  It is assumed that the individual constructs ideas of
these moments, that these constructings are themselves strategies, that
these constructings are sometimes repetitions of ideas used in the past
and sometimes newly created in term s of how the individual defines the
new situation.  It is further assumed that the individual will implement
his/her pictures using behavioral tactics which are responsive to the
individual's ideas of the situation.  Some of these tactics will again be
repetitions of behaviors of the past given the rule-based characteristics
of much of human behavior.  What tactic is used has consequences for the
kind of idea created; and, the kind of idea created has implication for
tactic used. 
	This formulation leads to a proposition which states that
individual use of information and information systems is responsive to
situational conditions as defined by that individual.  In essence, the
individual defines and attempts to bridges discontin uities or gaps.  It
is this focus on gap-defining and gap-bridging which is seen as offering a
way of introducing order to conceptualizations of individual behavior.  It
is not the individual entity that is seen as ordered but rather the
gap-defining and gap-bridging that is ordered. 
	What is proposed here is the idea that the "essence" of the
communicating moment is best addressed by focusing on how the actor in the
moment defined that moment and attempted to bridge that moment when
conceptualized in gap terms.  It is assumed that th e "gap" idea gets to
the essence of the communicating moment both in terms of describing and
explaining that moment as seen by the actor and in terms of predicting the
behavior of that actor in that moment. 
	At a specific moment in time-space, therefore, an individual who
defines self as facing a gap of a particular kind may use communicating
tactics of a particular kind.  In a different moment facing a different
gap he/she may use a different tactic.  He/sh e may, in fact, be very
rigid but the rigidity may be of the kind which says "given this gap, then
this tactic."  Or, he/she may be very flexible...or perhaps entirely
capricious.  The point, though, is that by focusing on the gap defining
and bridging we
 allow to emerge for examination human flexibilities and rigidities and
allow the possibility that both are amenable to systematic analysis. 
	Sense-making, thus, sets forth the gap idea as a theoretic
assumption and as a guiding frame for method -- methods of framing
questions, methods of interviewing and methods of analysis.  It is
proposed that focusing on the gap idea moves research toward a new kind of
generalizability -- a generalizability at a more abstract, more
fundamental, and more powerful level applicable across situations but at
the same time more pertinent and more relevant to specific moments in
	In proposing that by assuming across time-space constancy we have
missed time-space bound constancy, I am not suggesting that there are no
across time-space rigidities or patterns in human use of information and
systems.  Sense-making assumes that human s are to varying extents under
varying conditions responsive to external constraints on their behavior. 
Given the law of least effort, for example, we would expect humans to
repeat applications of past strategies and tactics to new moments in
time-space if these new moments are themselves seen as repetitions of the
past.  However, since much of human life is inherently unpredictable, much
of human behavior involves creating new responses.  But, external
conditions such as economic class, income, and educ ation are illustrative
of the kind of structural constraints which delimit the creating of new
responses.  To the extent that these external conditions are perceived as
operating, we would expect to find constancies across time-space in human
behavior.  I t would be expected, therefore, to find constancies in use of
channels (e.g. how much a person uses a library or even reads a particular
class of book) more than constancies in use of information (e.g. what a
person does with what he/she reads). 
	An important aspect of this formulation is the idea that
communicating behaviors are the link between individual and
structure/institution/culture.  While current contests in the social
sciences seem to pit the individual against structure and conceptual ize
the structure as an across time-space entity that persists despite
individuals, in fact they are part of a whole.  It is a whole we have
often missed in part because just as we have assumed that information
sharing could be conceptualized as transmis sion or transfer we have
assumed that the relationship between structure and person could be
conceptualized as transmission or transfer.  Anthropological terms such as
"acculturation" arise from such a formulation.
	  Sense-making does assume that the individual is situated at
cultural/historical moments in time-space and that
culture/history/institution define much of the world within which the
individual lives.  But, sense-making also assumes that the individual' s
relationship to these moments and the structures that define them is
always a matter of self-construction, no matter how non-individualistic
the person or his/her time-space may seem.  Structure is energized by,
maintained, reified, changed, and create d by individual acts of
communicating.  Because we have sought only across time-space
understandings, we have missed so much of the whole range of human
existence that involves struggling with, breaking with, coming to terms
with, and changing whatever st ructure the human finds oneself in. 
	In essence, we have done better at developing understandings of
human rigidities than of human creativities.  One reason for proposing the
use of discontinuity assumptions in the study of human use of information
and information systems is that it is in the realm of information behavior
that we ought to find humans at their most creative, least constrained by
external forces since so much of individual information use is private. 

	In the most general sense, the methodological approach that is
called sense-making is an approach to studying the constructing that
humans do to make sense of their experiences.  For our purposes in this
chapter, the experiences we want to study are expe riences relevant to
information and communication system design, management, and practice. 
	There is no direct way we can point to standard genres in the
literature and say that this is the focus of sense-making.  It is true
that the approach has been used to construct studies of aspects of
experience which systems (and the researchers hired by
 these systems) call information need, satisfaction, or image studies. 
What this means is that sense-making has been used to study human
sense-making in situations where humans: reached out for something they
called information, used something they saw a s a potential source and
judged whether it helped or not, or created an idea about an institution
based on experience with it. 
	Sense-making is seen, thus, as a generalizable methodology
developed for the study of all situations which involve communicating.  It
is implemented in all studies with a simple operational metaphor, derived
directly from the discontinuity or gap idea.  This metaphor is pictured in
Figure 1.  While it can be applied to entities other than individual human
entities (e.g. collectivities), it will for purposes here be applied to
individual behavior. 

	Assume a human being taking steps through experiences: each
moment, a new step. The step may be a repetition of past behavior but it
is always theoretically a new step because it occurs in a new moment in
time-space.  Assume a moment of discontinuity in
 which step-taking turns from free-flowing journey to stop.  Focus on the
the individual at this moment of discontinuity, this stop which does not
permit the individual, in his/her own perception, to move forward without
constructing a new/changed sense. 
  Determine how the individual interprets and bridges this moment -- what
strategy he/she used to define the situation which was the gap; how he/she
conceptualized the discontinuity as gap and the bridge across it; how
he/she moved tactically to bridge th e gap; how he/she proceeded with the
journey after crossing the bridge. 
	This metaphor constructs the "sense-making triangle" of
situation-gap-help/use, pictured in Figure 2.  As an individual moves
through an experience, each moment is potentially a


sense-making moment.  The essence of that sense-making moment is assumed
to be addressed by focusing on how the actor defined and dealt with the
situation, the gap, the bridge, and the continuation of the journey after
crossing the bridge. 
	The metaphor is a highly abstract one seen as applying at all
levels (e.g. intrapersonal; interpersonal, small group, organizational,
mass, telecomm, data base, societal) of information use and seeking and in
all contexts (e.g. health, political, scient ific, instructional).  It
directs attention to the steps the actor takes as defined on his/her own
terms to address the gaps he/she faces as defined on his/her own terms. 
It is not intended to suggest that all situation-facing is linear or
purposive.  W hen seen from the actor's perspective, time can be
constructed in a variety of ways, linear, cyclically and otherwise.  And,
situation-facing may or not be goal-oriented in the usual sense. 
Sense-making assumes there is always present a mandate to cope w ith self
even if the purpose in the situation is defined by the individual as to
forget self or lose control of self.  Nor does the method enforce any idea
of correct divisions of situations into moments.  Given the discontinuity
assumption, the moments a re seen as vehicles for examining gap-defining
and gap-bridging not representations of reality as it is.
	As was noted above, the sense-making approach is applied
throughout the research process -- in framing questions, in collecting
data, and in analyzing data.  All of these are part and parcel of the same
holistic process.  The place where sense-making met hods show most clearly
is in collecting data so this will be the focus of this section.  The
approaches impact on the framing of research questions as was illustrated
in section I; the impact on the analysis of data will be illustrated in
section III belo w.
	Sense-making provides a theory of how to conduct interviews with
respondents.  It is seen as applicable both to formalized and extended
interviews in research studies as well as the less formalized and often
briefer interviews by which the institution in tersects with its
users/potential users in order to provide service.  Both of these
interview situations are seen as requiring implementation based on gap
assumptions.  Our focus in this section will be on research interviews.
	Showing the variety of sense-making interviewing methods requires
that we start with the core method, the one which is most clearly
theoretically derived and most isomorphic with the sense-making
assumptions.  This is called the micro-moment time-line in terview.  An
example of such an interview is presented in Figure 3.  In the
micro-moment interview, the respondent is 

asked to reconstruct a situation in terms of what happened in the
situation (the time-line steps).  The respondent is then asked to describe
each step in detail.  The core focus of the description is directed to the
sense-making triangle, circling the mic ro-moment in terms of how the
actor saw the situation, the gap, and the help he/she wanted -- where he
she wanted to land after crossing the bridge.  What additional elements
are examined and what elements are stressed depend on research purposes. 
	In studies of information needs, emphasis has been placed on
understanding how the individual saw self as stopped, what questions or
confusions he/she defined, what strategies he/she preferred for arriving
at answers, what success he/she had in arriving at answers, how he/she was
helped by answers (i.e. how he/she put the answers to use), and what
barriers he/she saw standing in the way to arriving at answers.  The
situations or experiences which have been the foci of interviews directed
at needs asses sment have been ones pertinent to the service mandate of
the institution.  In a study for a public library, for example, the
situations involved all everyday need experiences -- situations in which
the actor needed an answer to one or more questions, for
 example.  In a study for a blood center, the situations involved each
donor's last experience with donating.  In a study of cancer patients, the
situations involved each patient's most difficult treatment session.  In a
study of users of television prog ramming guides, the situations involved
each viewer's last use of such a guide. 3
	In contrast, studies of satisfactions with an institution have
focused on descriptions of the actors last (in other studies, worst, best,
most memorable, and so on) use of the institution from the time when the
possibility of use came to mind.  Again, time line interviews focus at
each step on the sense-making triangle.  A satisfaction study focuses more
emphasis, however, on what helps actors sought and what ones they found;
what barriers they saw standing in the way; what they saw as causing
barriers ; what barriers they got around; and the ways the institution or
its representatives helped in this process.  Again, situations which are
the focus of the interview are selected to fit research purposes.  In
studies for public libraries, for example, situ ations have focused on the
user's last use, worst use, and most remembered use.  In a study for an
archdiocese, respondents were asked to recollect situations in which they
were helped by church, and situations in which they were hindered by
	As a third example, studies of images of institutions have
examined the most recent situation in which the institution was in the
actor's consciousness via contact, conversation, media, or simply memory. 
Again, the sense-making triangle is used to focus
 attention with special emphasis placed here on what ideas the actor saw
self as having as a result of this most recent situation of consciousness
and what impacts these ideas had on his/her behavior (i.e. how she/he made
them useful).5
	In each of these research genres, sense-making is focused on
different elements depending on the research purpose.  For each of them,
the respondent is asked to reconstruct different situations, again
depending on the purpose.  What is common to all, ho wever, is that the
respondent is focused on real experiences and that while these experiences
can be infinitely variable across respondents they are given systematic
order because recollections are guided in terms of the sense-making
metaphor and focused on gap-defining and gap-bridging. 
	A number of derivative interviewing methods have been developed to
suit different purposes.  One of these is the abbreviated time-line
interview in which respondents are asked to focus on only one (e.g.. the
worst, the best, the most important, the most troublesome) step or
question or help or barrier.  This interviewing method is useful
particularly in research situations involving routine or habitual behavior
where the depth and detail of the micro-moment interview would be
	Another approach -- called the help chain -- focuses in particular
on how the respondent constructs the connection between
information/system/structure and self.  Here the respondent is asked to
chain his/her answers to the query:  How did [the library, the book, the
data base, the article, etc.] help you?  The chaining involves repeated
queries to the respondent of the type "And, how did that help you?" until
the respondent feels the statement of help has been made in the most
personal and life-relevant
	Another approach is message/q-ing in which the intent is to
utilize respondent sense-making to understand and/or improve specific
messages, such as software manuals.  Here is the respondent is asked to
focus on elements of a message which involved gap-de fining and/or
gap-bridging in some way.  The element may have led the respondent to have
an idea...or face a confusion, and so on.
	These four examples of interviewing approaches do not constitute
all approaches developed to date. They do provide the major examples,
however.  An important element in all of them is that the respondent is
conceptualized as a colleague.  There is no ele ment of the study purpose
which is hidden in any way.  Frequently, the respondent is involved in a
quite lengthy process in which he/she is taught many aspects of the
interviewing approach so he/she can control the pace of the interview
itself.  At firs t, some respondents and some interviewers balk at the
approach.  Interviewers balk because they want to be more directive and
need assistance in learning how to redirect the respondent to the
sense-making metaphor and how to assist the respondent in reco nstructing
his/her sense-making.  Respondents, on the other hand, need assistance in
learning how to present all the personally-important details that they
wish while at the same time utilizing the interview structure.  The
relationship is presented as a quid pro quo and results to date suggest
that the large majority of respondents accept it as such.  Using the
approach, studies have successfully utilized phone interviews which lasted
well beyond the accepted 15-20 minute average which survey researchers
 say is the maximum tolerated time.  In many studies, respondents have
volunteered to be interviewed again.  Results suggest that there is, when
sense-making interviews are at their best, a conscientizing and
therapeutic value to the process for responden ts. 

	Exemplar studies are selected for presentation here in order to
illustrate important elements in the discussions above.  Exemplars will be
presented to illustrate the genres of research to which sense-making has
been applied, namely studies of informatio n needs, images, and
satisfactions.  Exemplars will also be presented to illustrate the use of
sense-making in descriptive work, in system design, and in
hypothesis-testing work.  In addition, exemplars will be presented to
illustrate the use of the appro ach to yield data which is dealt with
primarily in what has been traditionally called qualitative terms(e.g. as
case studies) as well as to yield data which were then further analyzed
via content analytic and other tools to yield systematic results.  Furt
her, while it is a rare application, exemplars will be presented of
studies which have been primarily close-ended, asking respondents to
describe which of a series of gap-bridging and gap-defining categories
describe their situations best. 
	Any given study can serve more than one exemplar function, of
course, so the selected studies are presented below one after the other
with each described in terms of the different ways in which it is seen as
illuminating points made above.  Six studie s in all are described from
the set of forty or more studies done to date. 

	In a study for the California State Library, the research purpose
was to identify the "information needs" of Californians and to elaborate
ways libraries could usefully help with these needs.6 Black, Asian,
Hispanic, and Anglo individuals were asked to
 describe step by step what happened to them in the most important recent
troublesome situation they faced.  For each step, they were then asked to
indicate what questions they had.  They were then asked to select the most
important question they had and
 for this question to indicate whether they got an answer and how the
answer helped.  Each of these elements was seen as the implementings of
communicatings by the individual -- definings of gaps and actings to face
gappiness.  Here are excerpts from a s ample interview of an 18-year-old

The time-line
Step 1:	I quit school because I got pregnant.
Step 2:	I had the baby one month ago.
Step 3: I didn't know whether to go back to school or not.
Step 4:	I am only 18 and my folks thought it was important.
Step 5:	I live at home so I have no expenses and my mother babysits for me.
Step 6:	So I am going back to school.

Questions asked
At step 3:  	How important is returning to school?
At step 4:  	How much do I really want to go back?

Analysis of most important question (question at step 3)
How got answer:  My own thinking and parent's advice.
How did answer help:  Made me feel better about me.  Got me started toward
going back to school. 
	The study classified, using content analysis, ways in which
individuals conceptualized themselves as stopped or hindered in these
situations and hypothesized that these conceptualizations would be a
better predictor of information seeking (question askin g) and using (how
the individual applied answers to gaps, i.e. how they were helped) than
	Thus, for example, it was assumed that two individuals -- one
white, one black -- both of whom saw themselves as being dragged down
roads not of their own choosing, would be like each other communicatively
at that moment.  They contrasted with two indivi duals -- both black --
one of whom saw self as being dragged down a road not of his/her own
choosing and the other of whom saw self as faced with a host of
alternative roads from which to choose.  The hypothesis was confirmed.
	In line with the expectations that there are some kinds of
communicating that are highly constrained by the kind of socio-economic
forces for which race is constructed as a descriptor of human entities, it
was found that while race did not predict inform ation seeking in terms of
questions asked it did predict, in interaction with gap-definings, what
channels or sources individuals used to get answers. 
	In addition to the data used for hypothesis testing, results of
this study provided a large amount of descriptive data regarding the
nature of need situations, the kind of questions people had, the barriers
they saw to question-answering, the strategies they used to get answers,
the success they had with different strategies, the helps they wanted from
answers, the barriers they saw as standing between them and getting help. 
Essentially the same approach to information needs assessment has been
appl ied in a wide variety of contexts -- for example:  with Hmong
refugees regarding their health information needs; with cancer patients
regarding their information needs while underdoing treatment; and with
blood donors regarding their information needs per taining to donating
	When researchers are interested in quantification of their
qualitative results, each specific study context can involve its own
detailed set of content analytic categories.  Blood donors often ask "Will
I faint?" while cancer patients ask "Will I die?" 
 and citizens facing everyday situations ask "How long will it take me to
handle this?"  Much of the quantitative work of sense-making studies, to
date, has also focused on developing generic categories to describe needs,
barriers, helps wanted -- categor ies which are universal in the sense
that they pertain to gap-bridging and gap-defining across situations while
at the same time they capture important aspects of particular situations.
	Across studies, these category schemes have stabilized.  Several
examples will be given here.  To capture the ways in which humans see
their journeys blocked, a set of categories have been developed under the
label "situation stops".  Among the catego ries in this schema are these
stops:  the decision stop, where the human sees two or more roads ahead;
the barrier stop, where the human sees one road ahead but something or
someone stands on the road blocking the way; the spin-out stop, where the
human s ees self as having no road; the wash-out stop, where the human
sees self as on a road that suddenly disappears; the problematic stop,
where the human sees self as being dragged down a road not of his/her own
	To capture the questions people ask in situations in terms of
universal gap-definings, these categories have been developed to capture
human attempts to bridge gaps relating to: characteristics of self,
characteristics of others, characteristics of objec ts or events, the
reasons or causes of events, the consequences of actions or events, the
connections between things.  The foci of these questions change depending
on whether the human is defining gaps he/she sees in the past, present, or
	To capture the ways in which people put answers to questions to
use, these "help" categories have been developed:  creating ideas, finding
directions or ways to move, acquiring skills, getting support or
confirmation, getting motivated, getting connected
 to others, calming down or relaxing, getting pleasure or happiness,
reaching goals. 
	These three examples illustrate content analysis schemes that have
been developed across a series of studies to capture the gap-defining and
gap-bridging aspects of situation-facing.  Sense-making theory leads to
predictions, for example, that the ways i n which people see their gaps
will be related to the ways in which they try to bridge them and not to
characteristics of persons independent of the gaps.  Results to date
confirm this prediction.  Such findings are seen as potentially applicable
to system
 design not only in terms of how practitioners interact with users to
determine their needs but also in terms of actual elements of the design
of information storage and retrieval systems.  For example, a portrait of
how a book was seen as helpful to the past ten readers may be a more
useful retrieval device for the next reader than traditional categories. 
Or, as another example, an information agency should have better success
in "educating" its mandated public if it gives the public control of the
ent ry point of the information exchange.  This becomes possible if the
user can enter the system in terms of the question he/she has. 

	In a study of information seeking by blood donors7, donors were
asked to describe their donating situations in terms of what happened,
what questions they had during the process, and how they hoped to answers
to their questions would help them.  Again, responses from the donors were
conceptualized as strategies of gap-defining and tactics of gap-facing. 
Excerpts from a sample interview follow. 

The time-line and questions at each step
Step 1: We were told we could get extra credit in health class for donating.
Q1:	How much did I have to give?
Q2: 	What are the procedures?
Step 2: A friend who had donated told me about it so a friend and I 
        decided to donate. 
Q1: 	How long would it take?
Q2: 	Would it hurt?
Q3: 	How big is the needle?
Q4: 	How much blood do I have to give?
(Skipping down to step 7)
Step 7: The nurse called me in and I didn't know what was going on.
Q1:  	What are they going to do?
Q2: 	What is all this equipment for if they are just going to take my blood?
(Skipping down to last step 11)
Step 11:  After eight minutes I went to the canteen for cookies and juice.

In-depth analysis of Q2, step 7
How easy was it to get an answer to this question? 1 (on a 1-10 scale)
Did you ever get an answer to this question?  no
Why?  Because they didn't tell me anything.  They just did it.  
How would the answer have helped?  I wouldn't have been scared or in
suspense wondering what they were going to do? 

	A battery of different predictors were compared in terms of their
power to predict how donors wanted to be helped by information:  across
time-space measures (e.g. demography);  apriori time-space measures (e.g.
interests and focus of attention at the mo ment of entering the donating
situation); and time-space bound measures (e.g. how the donors defined
their gaps and how they acted to face them).  Results showed that across
time-space and a priori time-space measures accounted for an average of
only 3% o f the variance in several criterion measures tapping how donors
wanted information to help.  In contrast, time-space bound measures
accounted for an average of 17%. 
	Detailed qualitative analysis of the primary statistical patterns
showed that each step in the donating process had its own characteristic
pattern of information seeking and using.  Based on these results, a
design was constructed to provide a user-frie ndly computerized
question-answering system at five different points during the donating
process.  Typical questions at each point were to be displayed on the
screen of a terminal.  Donors would simply touch the question of interest
to them.  They could then choose from a variety of question-answering
strategies.  For example, two donors, each asking the same question --
Will I faint? -- could choose to select answers from doctors, or other
donors, or statistical counts, or any combination. 
	A public library with a large number of Hispanic citizens on its
mandated patron roster searched for ways to entice the Hispanics into the
library.8 Previously, a number of study approaches were tested and met
with hostility.  The traditional methods of
 publicity had failed.  The sense-making project focused on users of the
library's audiovisual services and serendipitously provided a breakthrough
for serving the Hispanic community.  One study asked 30 randomly selected
recent users what happened that b rought them to the library, what they
got while there, and how they were helped., A second study asked 64
randomly selected borrowers of library video-tapes how the specific
videotapes they used helped them.  Here are excerpts from two examples
from the second study: 

Example #1 
What was the title of the video taped borrowed?  Ghandi. How
did the video help?  It helped me set some positive goals and not give up
until we succeed. 

Example #2
What was thew title of the video tape borrowed?  Rumpelstiltskin
How did the video help?  This move let me sit down and watch television
with my children.  It was a movie they picked.  They decided on the movie
and the time for themselves.  The movie makes my children ask questions
about what is real. 

	Librarians reported that for the first time they had conducted
interviews with Hispanic patrons and not been met with hostility.  They
learned that their video-tape checkout service was providing these patrons
with important helps.  Further, the videotap es pointed to a link for
these patrons with other library services -- literacy training, for
example, and how-to-do-it books.  One librarian summed it up:  "it helped
us see patrons from a different point of view, to understand them better,
and to be bett er able to tolerate the crowds around the audio-visual
desk."  The library staff decided to move funds from other services to
video services. 

	In a study of images of an institution, a random sample of 1006
residents of a city were asked about their last contact with that
institution and what that contact involved.9 They were asked what ideas
that contact led them to have and what impact the y saw each idea as
having on their thinking, talking, and relating to the institution.  The
study was an attempt to understand public images regarding the institution
and the "effects" interactions with and/or awareness of the institution
had.  All elements -- situation definitions, images, and effects --- were
conceptualized in consonance with the sense-making formulation as hows --
the hows of the situation-defining and gap-facing.  Two sample interviews

Example #1
Nature of last contact:  I passed the building when I was downtown.
Idea that resulted from this contact:  How come they get to have such a 
big building?
Impact from this idea:  I think of them as not being good members of the 
community.  I wouldn't be lenient with them in a pinch.

Example #2
Nature of last contact:  I talked to someone who worked there.
Idea that result from this contact:  They treat the people who work there badly.
Impact from this idea:  I don't want to use their service or have any 
contact with them.

	A comparison of what predicted "images" and effects" showed that
situation was far more powerful as a predictor than demography.  Thus, for
example, two citizens, one old and one young, both of whom had contact
involving interpersonal interactions with e mployees who worked at the
institution, were likely to share the same ideas about the institution and
to see themselves as affected by these ideas in the same way. 
	For this particular institution, results showed that citizens
whose last consciousness of the institution involved coming to the idea
that the institution treated its employees badly were significiantly more
likely to report that they were explicitly avo iding use of the
institution as a result.  The results pointed, therefore, to specific
changes the institution had to make in itself if it were to change its

	College student information seeking and use in a series of 12
different situations was compared.10 The situations were created by
pitting three situational dimensions against each other.  One dimension
focused on how the individual defined his/her statu s in the situation --
low or high.  A second focused on how the individual defined the openness
of communication in the situation -- open or closed.  A third focused on
how the individual saw self as stopped -- as having to choose between two
or more road s seen ahead, as having to cope with being dragged down a
road not of his/her own choosing; or, as having to follow another more
experienced person down the road. 
	Each person was asked to recall a time when he/she was in a
situation of each type:  for example, where he/she had low status, was
making a decision, and communication was open; or, a time when he/she was
in a situation where he/she had low status, was m aking a decision, but
communication was closed.  In this way, each individual was asked to
report on 12 situations and for each was asked to indicate what questions
they had in the situation and what ways they wanted to be helped. 
	Again, all elements were conceptualized as communicating
strategies and tactics.  A predictive analysis showed that situational
measures predicted information seeking and use.  The important finding was
that neither demography nor the individual as state
 entity did so.  No statistically significant consistency was found in
information seeking and use in terms of individual behavior across the 12
situations.  Rather, how the individual defined the situation predicted
how the individual faced it, thus supp orting a hypothesis derived from
sense-making theory. 

	In a study of southeast Asian refugee health information needs11,
intact groups of southeast Asians were interviewed in bi-linqual settings
using the sense-making format. Respondents were allowed to talk in their
native language or in English, as they wi shed.  The group facilitator
translated.  The refugees were asked to recall their recent visits to a
hospital or clinic: to describe the events and for each event the
questions they had, whether they got answers, and how.  Here is the
results of the inter view with one respondent, a 35-year-old female whose
first visit to a US hospital was to have a baby. 

Situation 1:  The last time I went to the hospital is I have my baby and
then after I have my baby the doctor and nurse bring me cold water.  So
that in my culture that's different and I keep asking them about the
question :  Why the people that has new baby they keep drinking very cold

Situation 2: After I have my baby, I am very new, my body is changing and
they let me take a walk every two hours or three hours.  I keeping
thinking that my body is new and that I'm so tired so that they're doing
these things and it's so hard for me to u nderstand.  And also I think
that many things in my body is not wrong and there is no illness but I
just have a bay and I am thinking that in a few days I'll get better. 
I'll get strong but the doctor say you have to walk and I was thinking : 
Why he say this?  

	Case study results of the project were useful in helping medical
practitioners serving the new refugees answer their questions about US
medical services.  Alternatively, they were useful in tempering and
changing medical practitioner demands on the new p atients. 

	The six exemplars presented are only a few of the some 40
different sense-making studies conducted to date.  Studies have been
conducted with a wide variety of populations (e.g. pre-school children,
teenagers, doctoral students, developmental disabled a dults) pertaining
to their information needs in a wide variety of situations (e.g. health,
environment, politics, science, childcare, education, finances, leisure
time. everyday life, job) and their interactions with a wide variety of
communication syste ms (e.g. libraries, data bases, media, books,
newspapers, software manuals) for a wide variety of institutions (e.g.
California State Library, National Cancer Institute, Ohio Department of
	The intent here has been to present a representative sampling of a
methodological approach which has been widely applied to research
questions relating to human use of information and information systems but
which is itself conceptualized as a generalize d methodological approach
for the study of any situation in which one wishes to focus on how people
construct sense of their experiences. 
	Having presented the discussion and exemplars above, it is now
possible to address the aspect of sense-making which brings it into this
volume -- the qualitative nature of the approach.  The issue of
quantitative vs qualitative approaches to the study of
 human behavior is one of a set of inter-related issues which have come to
be known as the ferment in the social sciences, or the paradigm crisis.12
These include among others, the issues of theoretic vs applied research,
individualistic vs structural research, and administrative (i.e. serving
established institutions) vs critical(i.e. criticizing institutions)
research as well as qualitative vs quantitative.
	The presence of "versus" in these ferment descriptions generally
implies that the researcher must choose between one or the other
--theoretic or applied, individualistic or structural, administrative or
critical.  While the arguments can not be fully de veloped in this
chapter, sense-making as an approach has explicitly cast itself in the
middle: as all of the above.  It is theoretic because it sets out to test
hypothethical propositions and is itself based on a coherent set of
theoretic premises and ass umptions.  It is applied because the work is
seen as directly applicable to information system management, design, and
practice.  In fact, in its focus on communicating behaviors, sense-making
sets out explicitly to develop theoretic understanding directl y useful to
practice;  to be, in effect, a theory of practice.  Sense-making is
individualistic in its focus because it acknowledges that individual
humans are the carriers of communicative action -- the acts by which
meaning is made and systems are ener gized.  But, it is also structural
because it acknowleges that individuals live in and embody structures and
have varying degrees of consciousness of this.  The approach is
administrative in that it sets out to improve systems and critical because
it ser ves as a vehicle for the users of those systems to speak to those
systems on their own terms.  In essence, the approach attempts to provide
a vehicle for giving voice to users and potential users of systems so that
the systems can be responsive to them. 

	In the argument between qualitative and quantitative approaches to
research, sense-making likewise refuses to choose a side.  It is
explicitly both qualitative and quantitative.  Even more important,
sense-making does not see any of these choices (e.g.  qualitative vs
quantitative, administrative vs critical) as legitimate or always binding
choices but rather as polarizations that have at least in part resulted
from specious understandings of the nature of research growing out of the
application of unpow erful and too limiting assumptions of the nature of
human information behaviors. 
	To illustrate this argument, return again to the idea that the
behavior of individuals can look chaotic if one keeps looking for
constancy in the wrong place -- as carried by the entity rather than by
process.  Given an assumed chaotic individuality, it
 becomes rather easy to frame qualitative research approaches,
particularly those forms of qualitative research which are proposed as
having to be systematically unsystematic, as the necessary response. 
Sense-making, on the other hand, assumes there is something systematic
about individual behavior to be found by pursuing process orientations. 
In this way, then, sense-making casts itself as systematic qualitative
research -- an approach with qualitative sensitivity which is amenable to
the systematic power of quantitative analysis.
	To understand this more fully, it is necessary to examine the
several meanings of the word "qualitative" which apply to sense-making. 
One of these is fundamental to the theory of sense-making -- the
assumption that human use of information and inform ation systems is
qualitative, not monolithic.  This implements the discontinuity
assumption.  Information is not seen as something which describes a given
reality in an absolute and potentially accurate way, which can be
transmitted from source to receive r through channels, which can be
counted by external standards and pigeon-holed for all time.  Rather,
information is constructed -- the act of constructing and the act of using
that which is constructed is a qualitative act.  It varies in kinds. 
	Sense-making assumes that the essential aspects of information use
can be captured by looking at qualities of gap-defining and gap-bridging. 
A person in a moment defines that moment as a particular kind of gap,
constructs a particular strategy for facin g the moment, and implements
that strategy with a particular tactic.  Gap-defining and gap-bridging
become, therefore, the essential qualitative aspects to be examined. 
	A second way in which sense-making is qualitative is that its
implementation in method is at least in part what we usually term
qualitative.  This is a given based on the assumption set-forth earlier
that method is residual of theoretic assumption.  Sen se-making methods,
therefore, yield data which are identifiable as qualitative.  It consists
of open-ended responses to questions, for example, and can be constructed
as case studies or records or interactions with messages and so on.  In
the few studies which have been entirely close-ended, the close-endedness
has involved respondent assessments of qualities of gap-defining and
	What is different about sense-making, however, is that the
qualitative data collection and analysis methods are all guided by the
same general theory of what is appropriate to capture in these qualitative
analyses.  This theory introduces, therefore, a means of systematization
across qualitative analyses. 

Figure 1
The Sense-Making Metaphor

Figure 2
The Sense-Making Triangle

Figure 3
An example of a micro-moment time-line interview


1 References to sense-making studies are incorporated throughout this
chapter.  For the most recent published pieces, see:  Brenda Dervin,
"Audience as Listener and Learner, Teacher and Confidante: The
Sense-Making Approach," in Ronald Rice and Charles Atk ins, ed. Public
Communication Campaigns, second edition (Newbury Park, CA: Sage, 1989),
67-86; Brenda Dervin, "Users as Research Inventions:  How Research
Categories Perpetuate Inequities," Journal of Communication 39:3 (Summer
1989), 216-232; Brenda Der vin and Michael Nilan, "Information Needs and
Uses," Annual Review of Information Science and Technology 21 (1986),
3-33.  The most comprehensive description to date is available from the
author as:  Brenda Dervin, "An Overview of Sense-Making Research: 
 Concepts, Methods, and Results to Date,"  paper  presented to International Communication Association, Dallas, May 1983.

2 Sense-making as an approach rests heavily on the work of Richard Carter. 
See, in particular:  Richard F. Carter, "What Does a Gap Imply?"  Paper
presented to International Communication Association, San Francisco, May
1989;  Richard F. Carter, "Discont inuity and Communication," Paper
presented at East-West Center Conference on Communication Theory East and
West, Honolulu, November 1980;  Richard F. Carter, "Toward More Unity in
Science," Unpublished paper, University of Washington, Seattle, 1974;
Richa rd F. Carter, "Communication as behavior," Paper presented at
Association for Education in Journalism, Fort Collins, 1973; Richard F.
Carter et al., "Application of Signaled Stopping Technique to
Communication Research," in Peter Clarke, ed. New Models f or
Communication Research (Beverly Hills, CA: Sage, 1972) 15-44. 
Sense-making owes a debt, as well, to the works of:  Jerome Bruner, Beyond
the Information Given (New York: W.W. Norton, 1973); Jesse Delia,
"Alternative Perspectives for the Study of Huma n Communication,"
Communication Quarterly 25 (1977) 46-62; Paolo Friere, Pedagogy of the
Oppressed (New York: Seabury Press, 1974); Clifford Geertz, The
Interpretation of Culture (New York: Basic Books, 1973); Anthony Giddens,
The Constitution of Society :  Outline of the Theory of Structuration
(Cambridge: Polity Press, 1984; Jurgen Habermas, Theory of Communicative
Action II: Lifeworld and Systems, T. McCarthy, tr. (Boston: Beacon press,
1987); Jurgen Habermas, Theory of Communicative Action I:  Reaso n and the
Rationalization of Society, T. McCarthy, tr. (Boston: Beacon press, 1984); 
Harvey Jackins, The Human Situation (Seattle: Rational Island Publishers,
1973); Klaus Krippendorff, "On the Ethics of Constructing Communication,
in Brenda Dervin et a l. Rethinking Communication 1: Paradigm Issues
(Newbury Park, CA: Sage, 1989) 66-96. 

3 For information needs assessments directed to public libraries, see, in
particular:  Rita Atwood and Brenda Dervin, "Challenges to Sociocultural
Predictors of Information Seeking: A test of Race versus Situation
Movement State," Communication Yearbook 5 (New Brunswick, NJ: Transaction,
1982) 549-569;  Brenda Dervin and Kathleen Clark, "Asking Significant
Questions: Alternative Tools for Information Need and Accountability
Assessments by Libraries." Report to California State Library, July 1987. 
For info rmation needs assessments of blood donors, see, in particular: 
Brenda Dervin et al., Improving Predictors of Information Use: A
Comparison of Predictor Types in a Health Communication Setting, "
Communication Yearbook 5 (New Brunswick, NJ: Transaction, 1 982) 806-830; 
Brenda Dervin et al., "Measuring Aspects of Information Seeking: A Test of
a Quantitative/Qualitative Methodology," Communication Yearbook 6 (
Beverly Hills: Sage, 1982) 549-569.  The study of users of TV programming
guides is in progress.

4 For sense-making designed accountability studies of public libraries,
see, in particular:  Dervin and Clark, "Asking Significant Questions";
Brenda Dervin and Benson Fraser, "How Libraries Help," Report to
California State Library, October 1985.  The arc hdiocese study is in

5 Brief reports on image studies are included in references listed in 
note 1. 

6 Reported in:  Atwood and Dervin, "Challenges to Sociocultural 
Predictors of Information Seeking."

7 Reported in Dervin et al., "Improving Predictors of Information Use" 
and Dervin et al., Measuring Aspects of Information Seeking"

8 Reported in Dervin and Clark, "Asking Significant Questions", 211-230

9 Reported in Dervin, "Audience as Listener and Learner, Teacher and 

10 Reported in Michael Nilan,  "Structural Constraints and Situational 
Information Seeking: A Test of Two Predictors in a Sense-Making Context," Doctoral Dissertation, University of Washington, 1985.

11 Reported in:  Scott Wittet, "Information Needs of Southeast Asian 
Refugees in Clinical and Hospital Situations,"  Master's Thesis, University of Washington, 1983.

12 Particularly helpful to the author in developing this section have
been:  Richard F. Carter, "Comparative Analysis and Theory in
Communication," Paper presented to International Communication
Association, San Francisco, 1989;  Ferment in the Field [Spec ial Issue],
Journal of Communication 33:3 (Summer 1983);  Anthony Giddens, "The
Orthodox Consensus and Emerging Synthesis," in Brenda Dervin et al.,
Rethinking Communication 1, 53-65; Stuart Hall, "Ideology and
Communication Theory," in Brenda Dervin et al., Rethinking Communication
1, 40-52.