Abstract. This article represents an emerging research genre which focuses on the study of information seeking and use in context. Context is explored using the Sense-Making Methodology which requires consideration of time, space and movement in defining and studying information seeking and use. This approach differs from most traditional studies that have conceptualized context and situation as pre-defined conditions that predict across time. This study's respondents included three professional groups: auditors, engineers, and architects. A Sense-Making interviewing approach was used to collect qualitative data focusing on information seeking and use in recent job-facing situations. An iterative comparative analysis involving the deductive application of Sense-Making's meta-theory with the inductive force of the data led to the development of ten information need situations, called information seeking situation types in this study. Seven of these situations were common to all three professional groups. These commonalities are noteworthy because previous studies have mainly highlighted differences between workplace domains. The conclusions from this study point to the utility of taking into account information seeking situation types which are conceptualized as embedded in time and arising out of sense-making activities as information users construct their movements through time-space.

This research report [1] is anchored in the tradition of user studies in the field of library and information science studies, It belongs to a long-standing tradition in library and information science field -- the study of user information needs and seeking. The first review of this genre appeared in the Annual Review of Information Science and Technology in 1978 (Crawford, 1978) and reviews have appeared regularly since (e.g. Dervin & Nilan, 1986, Hewins, 1990).

This study represents an emerging research genre within that tradition, a genre which focuses on research of information seeking and use in context [2] The Sense-Making Methodology developed by Dervin and colleagues has been an important force in this emerging research genre (Vakkari, 1997). During the 1970s and 1980s when Sense-Making was in its earlier stages of development, Dervin and others applied the approach in a large number of empirical studies on information seeking.[3] In the 1990s, however, the contributions of Sense-Making have been primarily meta-theoretical and methodological [4]. One important purpose of this study is to apply the philosophic attentions of recent Sense-Making work to empirical investigation. A second purpose is to provide an empirical example of a Sense-Making driven investigation of information seeking and use in context which highlights in important ways the differences between how Sense-Making approaches the definition of context in comparison to other approaches [5].

A review of the library and information science literature showed the emergence in the 1990s of a contextual (sometimes referred to as situational or dynamic) approach to conducting user studies of information seeking and use. However, what researchers have defined as "situational" differs markedly. For example, since the 1950s, workplace has generally been regarded as a "situational" factor in explaining differences in information seeking and use. Typical user studies have concluded that professionals differ in their preferred information sources because their workplaces attend to different informational domains. In one study, engineers preferred to gather information using internal reports (Allen, 1977); in another, academicians preferred to consult journals (Hogeweg, 1983). In essence, then, these studies assumed workplace to be a kind of global context or situation which could be conceptualized as constant across time-space.

In the 1990s, increasing numbers of researchers have recognized that even within a given workplace context, people's information seeking and use behaviors can differ. They have argued that there are other "situational" factors that influence information seeking and use. A wide variety of other factors have been presented as alternative contextual predictors such as work task; user demographic characteristics (e.g. education, income); and personality and other characteristics (e.g. locus of control, cognitive styles) (Ellis & Haugan, 1997; Borgman, 1996; Bystrom & Javelin, 1995).

Regardless of the "situational" factors being proposed, the bulk of traditional user studies have four methodological characteristics in common that contrast with the methodological mandates of Sense-Making:

1) "Situational" factors have been conceptualized as largely pre-defined from researchers' perspectives and usually represent aspects of the situation salient to the system and/or represent dominant approaches in the applied social sciences. This is what Dervin has referred to as "using a system mirror in conceptualizing the user" (Dervin 1992).

2) Researchers have hypothesized constancies in information seeking and use behaviors for people as a group across time, and have predominantly used across time-space factors (e.g. gender, demography, work roles, etc.) as their predictive/explanatory factors. As a result, numerous objective findings have been generated to highlight differences in how people seek and use information across workplaces. However, these findings seem to have limited utility for developing higher level generalizations across contexts (Reneker, 1993).

3) User behaviors have been conceptualized as atomistic rather than holistic, assuming that researchers can collect data on the nature of the situation, the information need, and the outcomes of information seeking as isolated components.

4) The concept of information need has, on the one hand, been conceptualized from the user-perspective but, on the other hand, operationalized from the system perspective. It has generally been accepted that an information need is the factor that triggers information seeking and use behaviors. The information need is typically defined as an "information gap" or an "anomalous state of knowledge" (Belkin, 1983). While these conceptualizations have clearly focused on the user's phenomenological world, in research practice the information need is mostly translated into a lack of subject knowledge (i.e. an absence of knowledge in a specific subject domain, usually defined in terms of the content and structure of the knowledge base as given). For example, typical user studies have assumed that engineers-as-a-group have the same needs for certain subject knowledges or information. Dervin (in press) emphasizes that this contradiction between the conceptualization of user need and its operationalization is particularly striking when applied to information seeking and use in particular because, by definition, at least a sub-set of information seeking behaviors must pertain to the regions beyond accepted knowledge structures.

5) The large number of individual characteristics that have been used as predictive factors (e.g. gender, education, cognitive styles, work task, and so on) have provided a body of findings which make people appear individualistic and capricious in information seeking and use behaviors. When predictive statistics are used, for example, the amount of variance accounted for by these across time-space factors describing users are typically very low, and the line of research pertaining to users' information seeking environments has not shown important gains over time (Fidel, 1987).

It is important to note in concluding this introduction on normative studies of information seeking in context that there have recently emerged two information seeking and use models (Kuhlthau, 1993; Ellis, et al., 1997) that address some of the methodological criticisms just mentioned. Neither of these models uses Sense-Making's methods for data collection or analysis, and unlike Sense-Making, they have emphasized the identification of specific stages or behaviors in very specialized information seeking contexts. Kuhlthau's model addresses information search processes of students and proposes six stages of information seeking and use: initiation, selection, exploration, formulation, collection, and presentation. Ellis et al.'s model examines "physical information seeking behaviors" in academic and workplace environments and proposes eight distinct information seeking behaviors which information seekers exhibit, regardless of what information sources they use. These eight behaviors are: starting, browsing, chaining, monitoring, differentiating, extracting, verifying, and ending.

It is beyond the purpose of this paper to provide a comparison of the Sense-Making approach employed in this study with the models offered by Kuhlthau and Ellis et al. Suffice it to say that all three attempt to identify situations in process-oriented ways and because of this, all three offer alternative interrogations of normative approaches to studying information seeking and use.

The purpose for developing the Sense-Making derived alternative proposed here is based on an assumption that the highly-developed meta-theoretic and methodological framework offered by Sense-Making offers tools that will permit more comprehensive and generalizeable analyses across diverse domains and diverse information seekers. This study is only the first step in what is expected to be an extended project.

The Sense-Making Methodology incorporates meta-theoretic assumptions, a foundation of methodological guidance, specific research methods (both for data collection and for question-framing and analysis), and a set of communication practices, all generated from a philosophic perspective which posits information as a human tool designed for making sense of a reality assumed to be both chaotic and orderly. [6]

Sense-Making is based on a central organizing metaphor of a person walking through time-space, facing a gap, bridging the gap to make sense, and moving on to the next moment in time-space. This metaphor is referred to as the Sense-Making Triangle. Its central meta-theoretic concepts include: time, space, horizon, movement, gap, and power. Its central operational concepts include situation, history, gap, barrier, constraint, force, bridge, sense-making strategies, outcomes, helps, and hurts.

It is important to emphasize that the Sense-Making Triangle is a highly abstract meta-theoretic tool and not assumed to be a picture of reality. It does not assume all sense-making (e.g. information seeking) is purposive or linear. Nor does it assume that sense-making requires resources outside of self, or even the conscious use of resources. Rather, purposiveness, linearity, and conscious seeking would all be sub-sets of what Sense-Making expects researchers to study.

The importance of this observation is this: Sense-Making is not an information seeking model, rather it is a model of a methodology to study information seeking and use (Wilson, 1999). It differs markedly from other information seeking models (e.g. Ellis, 1989; Kuhlthau 1993) which focus on specific information seeking and use behaviors in specific situations (e.g. an engineering company or a school environment). In brief preview, the essential components of the Sense-Making approach to context, which make it stand apart from other approaches, are (Dervin, in press):

1) The use in both data collection and analysis of conceptualizations which focus not on nouns nor substances but rather on verbs and processes. Sense-Making's focus on what Dervin calls "verbings" (Dervin, in press) is an attempt to free the approach from the use of "system mirrors" for studying users. By making the interface between researcher and researched one based on assumed universals of the human condition -- moving through time-space, facing gaps, bridging barriers, and so on -- Sense-Making removes the researchers' imposition of nouns.

2) An emphasis on movement and change across time-space. By introducing this assumption, Sense-Making is not suggesting that there will not be constancies across time-space. Rather, the assumption is that researchers should not impose constancy by applying methodological frameworks which do not allow change across time-space to emerge. Dervin explicitly emphasizes that Sense-Making assumes that there are patterns to be found in changes in movement across time-space and that when research relegates all such changes to error by not explicitly allowing the potential to exist methodologically, the result is the reduction of our understanding of users to solipsistic individualism.

3) The conceptualization of the underlying information need as defined by both the material and phenomenological horizons of the actor, past, present, and future. By employing the Sense-Making Triangle, "information need" is re-conceptualized as a discontinuity in movement through time-space whereby people perceive themselves as being in a moment of gap and needing to bridge that gap in some way. The information need is, thus, defined by a combination of the three components of the Sense-Making triangle, namely, gap (i.e. the questions or muddles people have), situation (i.e. the situation as embedded in time both past and present) and use/help (i.e. the kind of help people construct out of bridges across gaps). In the context of the interface between users and information systems mandated to serve their needs, the bridge may be information stored among the systems' informational resources. From the perspective of Sense-Making, however, the bridge is anything that informs and assists sense-making and can include both internal phenomena (e.g. ideas, emotions, feelings, memories, hunches) and external phenomena (e.g. a helping hand, a fact retrieved from a data base, a change in someone's behavior).

4) Expansion of the focus on user "information needs" to any gap in movement, and forces and constraints which impede movement. This is an important aspect of how Sense-Making mandates treating information needs holistically in terms of users' lived experiences and how users see both the so-called objective and the so-called subjective aspects of those experiences. In this way, Sense-Making attempts to transcend the objective-subjective polarization that Dervin suggests has led to a portrait of users as chaotically solipsistic.

This primarily qualitative study examined how 24 professionals constructed their information need situations in recent on-the-job task contexts. The professionals consisted of eight auditors, eight engineers and eight architects, all working in Singapore. Adopting the Sense-Making Methodology, this study investigated information seeking and use of each respondent situationally. The focus of the entire study allowed examination of how information seeking and use differs across professional groups and within professional groups by situation. The guiding research interest was the question of what can be gained by studying information seeking and use as embedded in specific time-space bound situations within workplace contexts. The specific aspect of the study emphasized in this report is the measurement of the information need situations faced by these professionals, including a comparison of these information need situations both across and within professional groups. Future analyses are planned to compare the efficacy of using these information need situations as predictors of information seeking (e.g. source using, diversity of source using) compared to professional domain.

The research question for this study, then, started with two assumptions of Sense-Making: (1) as users move through time-space their information seeking situations change moment by moment; and (2) from a verbing perspective, there will be commonalities in information seeking situations across diverse users. Our specific research question was: What information needs did these professionals have and how did these information needs compare across and within professions?

Employing Sense-Making to Reconceptualize Information Needs

There are two primary implications for the conceptualization of information needs to be drawn from Sense-Making. One is that the Sense-Making Methodology requires researchers to conceptualize the "information need" as a gap in the user's sense-making faced in the process of the movement through time-space which is assumed by Sense-Making to be a characteristic of the human condition. This conceptualization of information need contrasts markedly with the usual definition that focuses on the information need as a gap in the user's knowledge of and access to a given knowledge structure. In Sense-Making, the information need is always situational and subject to change. Further, it is always a gap that is defined at least partially outside any given knowledge structure and, thus, at least partially in terms that are beyond the capacities of any knowledge structure.

The second implication for the conceptualization of information need drawn from Sense-Making is the specification of the information need as an intersection of the points of the Sense-Making Triangle -- i.e. as a situation with a gap and a potential bridge over the gap which will have consequences or outcomes. The Triangle and its foundational use in this study is pictured in Figure 1.

==========================================================================

Figure 1. The Sense-Making triangle picturing an information seeking situation -- an information need -- in the micro-moment at the intersection of a situation, a gap, a potential bridge, and potential helps/uses which users get from information.

==========================================================================

It is important to not interpret the Triangle metaphor as implying linearity or purposiveness. For this study, the Triangle is operationalized as how the professionals saw their situations as "stopped," what questions they saw themselves as needing answers to, and how they wanted to put answers to use. In this study, the phrase "information seeking situation" is used in place of "information need" to avoid confusion. The phrase refers specifically to an intersection between a specific situation, a question it gave rise to, and the hoped for utility of a possible answer. The important point of contrast to traditional information needs studies is that in order to use Sense-Making to inductively identify information needs, the researcher has to take into account the three components of the Sense-Making Triangle -- situation, gap/bridge, and outcome.

The Sense-Making Methodology was first applied to the study of information seeking and use by everyday citizens in the 1980s, and later extended to public library users and other public service users. [7] From these studies, numerous statements have been elicited from users describing their information need situations. For these situations, actors have described the gaps they faced in situations, usually defined as the questions they asked or confusions they faced. Actors have also described the helps/uses they hoped to get from answers. These statements have been collected and content analyzed with conjoint attention to deductive use of Sense-Making's meta-theory and inductive use of the understandings being offered by users. The results of these efforts have been the development of a stable set of categories for coding situations, gaps and helps/uses into what are assumed to be universally relevant dimensions of human movement, and which are all conceptualized in process terms, that is, in terms of movements through time-space, or what Sense-Making calls verbings. [8]

Thus, Sense-Making studies have generated category schemes for focusing on how users see their situations, their stops in movement, the barriers and gaps they face, the uses they would make of sense-making, and the ways in which actual bridges over gaps help and/or hinder their movement. For example, in a typical Sense-Making study a user might be described as experiencing a problematic situation where she sees herself as being dragged down a road not of her own choosing; where she has little power but feels highly compelled to move towards some sort of resolution; where her primary questions include "Why me?" and "When will this stop?"; where the primary way she wants to be helped is by learning how others coped with similar situations; and where the actual bridge she manages to construct across her gap helps because she got the ideas of others and this helped her get control of the situation. A large number of such categories have been generated in Sense-Making studies. Dervin (in press) emphasizes that the purpose of these categories is not to develop an isomorphic map of "reality" but rather to provide a verbing interface that allows the researcher to understand the researched on their own terms. These categories also have the potential for those designing systems that are relevant across individual differences previously construed as solipsistic.

Employing Sense-Making's Micro-Moment Time-Line Interviewing

Dervin's (1983; 1992) Sense-Making's Micro-Moment Time-Line interviewing approach was employed for data collection. Twenty-four respondents from three professions -- auditors, engineers and architects -- participated in the study by sharing their information seeking and using experiences in the workplace. Each respondent was asked to recall a typical task he or she had accomplished at work. The respondent then described in detail each step or event he or she experienced in the process of completing the task. Specifically, respondents were asked to focus on all micro-moments when they saw themselves as "running out of sense." These micro-moments are termed sense-making instances in this study and it is these instances that are analyzed with the procedures described below.

After eliciting of the sense-making instances, the remaining interview, in fact the brunt of it, focused on what Dervin calls the "triangulation" of each sense-making instance using Sense-Making's central metaphor -- the Sense-Making Triangle. For each selected time-space moment, these questions were posed:

Questions on situations:
* What was the situation that led you to look for information?
* What was your objective? What did you feel?

Questions on gaps:
* What was the question in your mind?
* What did you want to find out more about?

Questions on uses/helps:
* How did you think answers could help (or fail to help) you?
* What were you trying to achieve/accomplish?

These 3 sets of questions were the data collection approaches for identifying information seeking situations in this study. In the context of each respondent's answers to the triangulation of each sense-making instance, these additional questions were asked to identify information seeking and use behaviors.

Questions on information seeking and use behaviors:
* What did you do (when you were in the reported situation)?
* What did you do to get answers to your questions? Why?
* How did you handle the answers you obtained?
* How did you decide whether the answers you got were acceptable?
* What problems/barriers did you have in getting answers?

It should be noted that in any one interview, the respondent described many different sense-making instances that she went through to complete her tasks. For clarity of presentation, the above questions were listed in a linear order. However, to encourage respondents to express themselves freely, the actual interviews were conducted in a more natural and less linear manner. For instance, some questions were skipped when the answers were apparent from prior questions. Also, respondent non-verbal clues were carefully monitored, and follow-up questions were posed (e.g. What else? Can you explain more on? What do you mean by? Can you give me an example of?).

In addition, use of the term information was avoided in the interview, as mandated by Dervin (in press). In Sense-Making terminology, "information" is a word that pertains to established structures which limit the experiences actors will share with the interviewer. For example, when an architect was asked what information he needed at a certain stage in his work process, he answered "I was not looking for information. I was looking for approval from the government authorities." When the term "information" was used, the interviewer [9] made it clear to the respondent that "information" can mean different things to different people, and she encouraged the respondents to elaborate further.

Conduct of the Interviews

All interviews were conducted by the senior author either in the respondent's or senior author's home or office. Each individual interview lasted for approximately ninety minutes. Before the interview, participants were briefed about the purpose of the research. They were also briefed about the nature of the interview, in particular, to let them know that the same set of questions would be asked repeatedly throughout the interview about different events or steps in a task-facing situation. An additional ten minutes of the interview were used to build rapport with the participants by chatting freely about issues related to their careers, work tasks and companies. All interviews were fully audiotaped and transcribed, except for one participant (Auditor #4) who preferred not to be audiotaped. In that case, the senior author took notes during the interview and typed out the transcript right after the interview. An excerpt of a sample interview transcript is presented in Figure 2.

==========================================================================

Figure 2. Excerpts from an interview with an architect.

==========================================================================

Data Analysis: Sense-Making Instances as Units of Analysis

Although the individual respondent is the data collection unit as reported above, the unit of analysis is not the individual as is normative in most user studies but, rather, the sense-making instance -- the moment in facing a work task situation where the professionals reported they ran out of sense. This follows Sense-Making's mandate which proposes that important dimensions of the systematic patterns in information seeking and use are not repeated inflexibly by individuals across time-space but are instead practiced reflexively in response to characteristics of given moments in time-space. The 24 respondents in this study identified a total of 626 sense-making instances during their task-facing situations. Of the 626 sense-making instances identified in this study, 256 were from auditors, 155 from engineers, and 215 from architects.

It is these 626 sense-making instances that were analyzed in this study in terms of what types of information seeking situations each represented and then what kinds of information seeking behaviors were used. It would be possible to call each of these sense-making instances an information need situation, or an information seeking situation. While this terminology would be conceptually correct, it becomes confusing when differentiating, for example, between all sense-making instances and those that were coded as falling into a specific information seeking situation type. For this reason, the term sense-making instance is used to describe the units of analysis.

For each of these sense-making instances elicited from respondents, the senior author examined all words and word phrases attached to each and placed them in broad, preliminary categories labeled "situation," "gap," and "uses/helps" -- i.e. in the major categories mandated for attention by the Sense-Making Triangle. Some data seemed to fit into more than one of these three broad categories. Thus, multiple codes were attached to sense-making instances when necessary.

When doing the content analysis, the author was informed by categories of situation-gap-help/uses developed from past Sense-Making studies (Dervin, 1983; Dervin, 1991). However, because this particular application of Sense-Making was a qualitative investigation, the categories from prior studies were not used a priori as pre-defined categories. Instead, the categories of situation-gap-help/uses were derived inductively. Thus, at the highest level of abstraction Sense-Making's meta-theoretic premises guided the category development; at the lowest level of abstraction the words of the 24 respondents as they described their 626 information seeking situations guided the category development.

To ensure the credibility of the analysis, the senior author's biases were probed, meanings explored and bases for interpretation clarified in a series of face-to-face debriefing sessions with academic advisors. In a typical debriefing, the author walked the debriefers through a randomly chosen sense-making instance and the debriefers described what they saw as the gap, situation and help/use. The categories inferred by the author were consistently the same as those coded by the debriefers. To further test the stability of the analysis, the author re-coded the interview transcripts four months after the initial findings had been stabilized in the first round of analysis. This re-coding demonstrated the stability of the findings, as only minor adjustments were needed to tighten the categorization schemes.

Figures 3,4, and 5 present the categories developed to tap situations, gaps, and helps/uses in this study. For the purposes of this paper, the delineation of the three different groups of categories was a preliminary step leading to the characterizing of each specific sense-making instance in a particular information seeking situation type. Thus, in defining information seeking situation types, the categories describing situation, gap, and helps/uses were assumed to play an intersecting and defining role. This is in keeping with Dervin's definition of the information need as being in the middle of the Sense-Making Triangle as pictured above in Figure 1.

==========================================================================

Figure 3. Categories inductively derived to characterize the situation angle in the application of the Sense-Making Triangle to the process of identifying information seeking situation types.

===========================================================================

Figure 4. Categories inductively derived to characterize the gap angle in the application of the Sense-Making Triangle to the process of identifying information seeking situation types.

===========================================================================
Figure 5. Categories inductively derived to characterize the helps/uses angle in the application of the Sense-Making Triangle to the process of identifying information seeking situation types.

==========================================================================

To arrive at the typology of information need stuation types, after identifying the contiquities of situation, gap, uses/help in each of the 626 sense-making instances, the author used a method of iterative comparison to investigate how each sub-category of situation, gap, helps/uses co-existed within a a given sense-making instance. This was compared against the next unit of analysis to find out whether it shared the same sub-categories or exhibited new ones. Having completed the iterative analytic process, ten major themes, or information seeking situation types, were identified. In terms of the Sense-Making conceptualization driving this study, these information seeking situation types became a proposed typology of information needs. Figure 6 provides an example of how the information seeking situation types that were constructed inductively in this study were mapped using the points of the Sense-Making Triangle -- situation, gap, uses/helps. Figure 6 applies Figure 1 to one of the ten information seeking situation types identified in this study -- the task initiating situation.

==========================================================================

Figure 6. An example for the task initiating information seeking situation type of the application of the Sense-Making Triangle in the coding of the information seeking situations.

==========================================================================

Results are presented in two sections below. First, the 10 different information seeking situations that emerged from the iterative comparison are described. This presentation proceeds without regard to how the different kinds of information seeking situation types were distributed across the three professional groups. Secondly, quantitative tools are then used to portray how the three professional groups compared in terms of both similarities and differences in their construction of these 10 information seeking situation types.

The Ten Information Seeking Situation Types

The findings showed that the auditors, engineers and architects commonly experienced seven information seeking situation types at work; and an additional three information seeking situation types were experienced primarily by architects. Figure 7 presents an overview of the results of the analysis in which the Sense-Making Triangle was used as a tool for examining the contingent relationships between the three points -- situation, gap, helps/uses -- which were assumed to form at their intersections an information seeking situation. What Figure 7 provides is an overview of the intersections of sub-categories of situation, gap, helps/uses which defined each information seeking situation type. The presentation below draws on Figure 7 in presenting a thematic description of each information seeking situation type. These descriptions are then illustrated with verbatim quotes from the interviews which, where possible, provide an example from each of the three professional groups.

==========================================================================

Figure 7. An overview of the contingent patterns that emerged between situation, gap, and helps/uses for the ten information seeking situation types identified in this study.

==========================================================================

Information Seeking Situation Type #1: Task Initiating Situation. As shown in Figure 7, when respondents saw themselves in a task initiating situation, they defined themselves as facing gaps which required getting started on new assignments. They expressed their information gaps in terms of asking factual questions about their assignments. For example, auditors asked which specific areas need to be looked into in more detail, engineers asked for the facts related to product failure, and architects asked how many units the developer wanted to build. When respondents were in task initiating situations, they saw answers as potentially helping to provide overviews of things (e.g. scope and objectives) as well as specific facts about the tasks they were facing (e.g. work-related facts, work schedule, work requirements, customer requirements, etc.). The following quotations illustrate some of the typical task initiating situations identified by the three professional groups:

For this client, December is their year end. So by September or October, I needed to know what the audit assignment was all about. The senior auditor, that is myself, was given the portfolio by my boss. Using this portfolio, I prepared the memorandum (Auditor #4).

The first thing we need to do is to narrow down the stages. First of all we need to get the schedule fixed. So within each schedule we need to define what task, what objectives must be met, what defines the completion of the task. Then the third step will be the resources allocated for each test and by when we must complete the task that was given or assigned (Engineer #5).

...for the Housing and Development Board we have new rules and regulations. These new rules and regulations will be e-mailed to me by the contract manager. All the liaison architects. Then, we make changes. We incorporate them in our drawing. We don't seek information, they provide it. Like in the first part, we want to win, so I need to know what people's needs are. And who I should be designing for. So our target is very clear initially . . . . (Architect #1).

Information Seeking Situation Type #2: Focus Forming Situation. As shown in Figure 7, when respondents saw themselves in a focus forming situation, they defined themselves as not having clear ideas of exactly what information they needed to complete their assignments. They saw unclear and foggy paths ahead. They asked many questions in order to get a better focus for their assignments so that they could better define and plan their work scope. They expressed their information gaps in terms of asking questions about the past and the future (e.g. What has been done in the past? What is the assignment expected to be like?). Also, they asked questions about direction, so as to get pointers to the right information sources (e.g. Who should I talk to? Where should I get the files?). Further, they asked for opinions of others (e.g. What do people think about the task? What are people's opinions on what should be done to tackle the problem?).

By getting answers to their questions, when respondents were in focus forming situations, they hoped answers would provide them with general understandings (e.g. getting any potentially interesting raw data that was not yet interpreted); the ability to gain control (e.g. by being able to predict); and a sense of the opinions of others (their advice, thoughts, and suggestions). The following quotations illustrate some examples of typical focus forming situations identified from the various professions:

. . . You know this is roughly the risk area, but how far should I go into how deep should you get into it when should you stop? There is no end to it. You can study the whole operation. It's like deciding what's the limit. That's one thing. If you don't go deep into it, maybe there's risk area in there, and you don't know. This is one area, sometimes, you have to decide there is some area of uncertainty (Auditor #6).

. . . I think the way it works is when at the beginning of the things, when you are faced with a completely new situation and you do not know what on earth is going on. The first thing you do is to try and find out what the gist of the problem is . . . of the situation, and what is expected of you. Then . . . you kind of get a feel knowledge of what you are expected to do (Engineer #6).

The other thing, of course, is to consult a senior architect before you do that. And I have quite a senior architect in the office, and I usually consult him before I do any of these and that helps . . . Usually I will tell what is the problem or what exactly is it I am trying to find out. And they will say oh, why don't you just call up this person, or look at this, or check out this book we have in our library or whatever (Architect #4).

Information Seeking Situation Type #3: Idea Assuming Situation. As shown in Figure 7, when respondents saw themselves in idea assuming situations, they defined themselves as having formed clear ideas about their destinations and exactly what information they were seeking. They saw the need to get access to specific information required to improve their specific task knowledge and get going toward their destinations.

Respondents expressed their information gaps in terms of asking factual and evaluative questions (e.g. What does all the information tell me?). By getting answers to their questions, respondents wanted to get answers which they saw as providing them with value-added information (e.g. getting specific, vital details and current information and knowledge). The following quotations illustrate some examples of typical descriptions of idea assuming situations as identified by the three professional groups:

. . . Normally you just know what you want to look for, then you know who are the persons in charge. You go to that particular person. This is definitely standard (Auditor #7).

I guess like any plan, you need to know what resources are available to you. For this, you talk to the people who are in charge of the resources. You can talk to your manager (Engineer #6).

We need all the structural input, we need all the beams, and columns and all that which will affect the design. And how big is the space required to be . . . . We have meetings. We give them a set of drawings, and they will give us their design, their layout. They will provide feedback to us. If they have encountered certain problems, they will highlight [these] to us. And we will check through the drawing (Architect #3).

Information Seeking Situation Type #4: Idea Rejecting Situation. As shown in Figure 7, when respondents saw themselves in idea rejecting situations, they defined themselves as being in situation movement states which in Sense-Making research have been called problematic, spin-out, or wash-out. Here, respondents saw themselves as being out of control, having no road, or being dragged down a road not of their own choosing. These were all situations where respondents reported they were unable to make sense of conflicting and unexpected information as it did not conform to their plans, expectations or personal understandings. In this situation type, respondents also reported not getting required information as expected.

Respondents expressed their information gaps in terms of questions seeking confirmation (e.g. Am I really seeing something abnormal?), reasons (e.g. Why would this happen?), history (e.g. Has this occurred in the past?), human error (e.g. Has someone made a mistake?) and evaluation (e.g. Should I spend more time on this issue?).

By getting answers to their questions, respondents wanted to get confirmation and explanations (e.g. getting clarification and reasons regarding problematic issues), as well as alternatives (e.g. other ways to complete the task instead of following the original plan). The following quotations illustrate some examples of typical professional narratives in idea rejecting situations:

So should we go down to the audit field and if we come up with any findings that weaken these foundations or destroy whatever we have laid down as the basis, we need to check back further and see if we misunderstood something. Have we missed some essential information that causes us to view the business differently at the planning stage? (Auditor #3).

Actually there were a few issues, especially concerning the power line testing: it's failing by a certain margin. In the US, it's failing by a lesser margin. So we needed to find out what the differences were. So when I shared my results with them, feedback to them my results, I tried to see what kind of setup they were using to do their power line testing -- what were the wave shapes of the discharge product -- to try and make sure that the actual test itself in Singapore was being done in the same way (Engineer #8).

Usually we go for the lowest. My question would be, why would the price be so low? Did he miss anything...if all are around the same rate? So ok, we look for something to list out in the tender interview, to ask them the questions and all that. But if one tender price is so far off from another, then you would like to question more, you like to check for more detail, like why? Did he miss something? Did he offer a cheaper product? Did he make a mistake? Did he forget a zero? (Architect #2).

Information Seeking Situation Type #5: Idea Confirming Situation. As Figure 7 reports, when respondents saw themselves in idea confirming situations, they defined themselves as connecting to truth and having to confirm, verify and double-check certain assumptions, ideas, issues and suggestions including those generated by self as well as those proposed by others.

Respondents expressed their information gaps in terms of asking factual, closed-ended questions about confirmation (e.g. Is it correct?). By getting answers to their questions, respondents hoped to get the confirmation they sought (e.g. evidence, third party opinion). The following quotations illustrate some examples of how professionals described idea confirming situations:

Normally you go to higher level, and maybe to the person in charge. At least, to seek a last clearance from the ultimate person in charge, before you raise the issue. Of course, if you are not satisfied, that will be our findings and our recommendation. But before that, we want to get clearance from them, at least our finding is factual. That is very important. So we need to clarify, is that what you are doing.? Just to clarify before we put up our findings (Auditor #7).

Just like the zinc coating thickness, so when we say coating thickness is below specification, it is possible, but it is not the actual root cause, it might cause... So we need to do some testing to confirm, so we need more data in this case. We prove that the part from the vendor is below spec, but we can only say it's possible it will rust faster. We cannot say that it will guarantee to rust in a certain time. So we need to do testing to double confirm, to pick up the data (Engineer #2).

Then we fax over [our building plans] to the registered architects, they have some guys to vet it, to see if we are on the right tracks, If everything is ok, fax it back to us and then we will continue on. So once the base is set, we'll carry through (Architect #5).

Information Seeking Situation Type #6: Idea Finalizing Situation. As Figure 7 reports, when respondents saw themselves in idea finalizing situations, they defined themselves as forming whole pictures in order to make decisions and/or to reach conclusions. Respondents saw themselves as using personal or collaborative judgments, so as to come up with value-added decisions/conclusions.

Respondents expressed their information gaps in these situations in terms of asking questions about conclusions (e.g. What is the final decision?) and confirmations (e.g. Will the manager agree with my decision?). Respondents wanted their answers to help them by leading them to decisions and providing them with the ability to make value-added judgments for the benefits of their companies. The following quotations come from typical respondent narratives describing idea finalizing situations:

Sometimes, only after when you consolidate all the information then you begin to think, do you begin to ask, think, do they tie together to each other? So there is no specific way, it depends. Sometimes you are able to piece them together. But sometimes what you gathered previously somehow contradicts what gather now. So you will need again further clarification from the same people (Auditor #7).

... you've got to create a mental framework on how you want to fit all the pieces of information together, and once you have done that it is in your head, that's when you can put it on paper (Engineer #6).

We would just proposed a solution on the spot to the inspector. We asked, proposed "Can we put the railing on this side?" They said, "Okay." So then we'll do it and we take a photograph and show him. It's a bit of at that moment, what are the questions that are raised, and what are the alternatives we to come up with. Then you did it fast, and re-submit.... We may discuss them with the boss or [other] architects. I told the boss, we did not comply with this, The boss may suggest something because he is the person who signs . But I am quite senior, and on simple things, I can just decide and let him know. For things which affect costs and have other implications, we let him know. For this circumstance, I decided myself. (Architect #5).

Information Seeking Situation Type #7: Idea Sharing Situation. As shown in Figure 7, when respondents saw themselves in idea sharing situations, they defined themselves as open to communication and needing to share their information and knowledge with other people. They perceived themselves as information providers (with the roles of advice giving, consulting, and debating), and they needed to learn how to ensure good communication with their audiences.

The idea sharing situation was predominantly characterized by respondents who expressed having no information gaps in the traditional sense since they already had all the knowledge they felt they needed. The most asked questions focused on learning about how to communicate with others (e.g. What will the audience like to hear? What is the most important information to be presented to the audience?). By getting answers to their questions, respondents wanted to get control, be able to predict, and specifically to prepare effective presentations which achieved their goals. The following quotations illustrate some examples of typical idea sharing situations as described by the three professional groups:

I need to consider what is best to include in the report. I think a good audit report should tell the management how the department is doing, and it should summarize everything that is found during audit. It should be short because management doesn't have time to go into details . . . We must know who's reading the report. Whether what I'm writing inside serves any value, gives any useful information. That's important (Auditor #7).

. . . thing that pops up in my mind is whether we are talking too mean to them. Since they are the design center, we can't say your design is lousy...we should not be too mean to them. Other than that, we'll be talking about engineering or scientific data, there is nothing to argue. If the test is okay, it's okay; if it's not okay, it's not okay (Engineer #1).

. . . usually the important thing is the clarity of information, no matter to the clients or to the contractors. You have to be clear to the clients so you won't misunderstand each other. You agree with each other. To the contractor is the same thing, because once you give out information and it is not clear and it's built according to that information, there will be a lot of argument so clarity is important. So what would be in my mind is to make sure that we understand their questions, what the clients want from me, and after that, whatever information I give to the person is correct (Architect #4).

Information Seeking Situation Type #8: Approval Granting Situation. As Figure 7 displays, when respondents saw themselves in approval granting situations, they defined themselves as observing or monitoring as they had to give permission to others who could then carry out decisions and tasks. Only four architects and one engineer reported that they constructed this information seeking situation type in facing their tasks.

These five respondents expressed their information gaps in terms of asking questions about facts (e.g. What are all the facts to support the application?) and confirmation (e.g. Is the information given correct?) so they could make decisions regarding giving approval to and supervising others. The following quotations illustrate some examples of typical approval granting situations:

Even when the product is being manufactured, we have to keep an eye on it to make sure there aren't any process changes. Even a change in the position of the wiring can affect the performance. Sometimes, procurement has changed the parts -- we have to be made aware of that. Everything is controlled. There are supposed to be no changes to the product unless we approve it. For anyone to change the design of the product, we need to be informed, and secondly we have to give an approval. It depends on whether there is a design change. If it's a part that we feel will affect the performance of a product...for this product, I'll do a test. For this product, it's now on hold. But there will come a day when the suppliers are changed so we still have to test (Engineer #8).

To verify that the . . . checks were correct. We are telling the clients, this has reached the first story, and they can collect money from the purchaser. We are liable (Architect #2).

Information Seeking Situation Type #9: Design Generating situation. As Figure 7 displays, when respondents saw themselves in design generating situations, they defined themselves as coming up with new ideas and concepts. Only architects reported that they this saw themselves in this information seeking situation in facing their tasks.

The architects expressed their information gaps in terms of asking questions about design (e.g. How to make use of the contour in the design?), alternative ideas (e.g. What are the various possibilities to use this piece of land?) and opinions of others (e.g. What ideas do my boss and colleagues have?). By getting answers to their questions, they saw themselves as gaining intuitive ideas and interpretations. The following is a verbatim quote from an architect describing a design generating information seeking situation:

We're supposed to always be reading. Because architecture is always changing, there are new things, new ideas, new concepts being done. It helps to keep you creative, I suppose...you shouldn't really read only when you start a project, I suppose. That would be copying. When it comes to design, there is no answer. It's not like an exam. It helps if you are very imaginative and a creative person. (Architect #4).

Information Seeking Situation Type #10: Approval Seeking Situation. Figure 7 shows that when respondents saw themselves in approval seeking situations, they defined themselves as being led by others. Only architects and one engineer reported that they saw themselves in approval seeking situations while facing their job tasks. These architects and engineer expressed their information gaps in terms of asking closed-ended questions about confirmation (e.g. Have I complied with all regulations? Is the distance of the car park from the road within legal requirements?) Answers to these questions were seen as helping by getting the approvals needed. The following quotations illustrate some examples of typical approval seeking situations as described by the architects and one engineer:

So in that case [realizing that because of higher priority demands he would not have time to complete all the measurements for the design center] I won't have time, say, to complete all the measurements. So what I would do is maybe I can get four out of the eight measurements. I'll go back and talk to the requestor of the test they are people from the design center. They wanted me to test something to verify it. Then I talk to them and say well I know you wanted me to test all eight, but in this case, will it do if I test four, because I have this and this and this constraint. Then if it's yes, it's fine, If it's no, then...it is a situation I can't deal with. And the person really needs it, then we'll see what other alternatives are. Can someone else do it? (Engineer #6).

In this stage, the anxiety is more, because all the technical requirements are there. Different technical departments have different requirements, and we are worried that we cannot comply....The problem is there are a lot of departments that have different requirements, and they keep changing. So we try to update ourselves....So sometimes, because there are so many things you cannot remember by heart, you may miss out on one or two items. So what we did was, we submitted proposals to all these departments. And when they came back, we looked and saw whether all the things could be complied with (Architect #2).

Comparison of the Three Professional Groups

The analytic mandate for the presentation of the 10 information seeking situation types described above was maximum homogeneity within a situation type in terms of the contingent intersection of the three points of analysis -- situation, gaps, and help/uses. Thus, the analysis above proceeded without regard to the substantive differences in the task worlds of auditors, engineers, and architects. The intent of the section that follows is to attend to the substantive worlds of the three professional groups and ask specifically what evidence emerged that professional context impacted information seekers such that the three professional groups differed in their likelihood of naming information seeking situations of different types. Table 1 and Figure 8 display the percentages of the information seeking situations of each type reported by the three professional groups. Table 1 also shows the baseline ns on which the percentages were calculated. The intent in presenting quantitative data at this point is to provide an organizing tool for qualitative presentations and not to use quantitative data in the more traditionally accepted sense of prediction and inference. This task is, however, slated for future analyses.

==========================================================================

Table 1. Percentages of the sense-making instances experienced by the three professional groups which were coded into each of the ten information seeking situation types.

===========================================================================

Figure 8. A graphic portrayal of the data presented in Table 1 showing percentages of sense-making instances experienced by the three professional groups which were coded into each of the ten information seeking situation types.

===========================================================================

In reading Table 1, the ns present the total number of each information seeking situation type as reported by auditors, engineers and architects. For example, the 8 auditors altogether reported 256 sense-making instances, 11 of which were coded as task initiating information seeking situations. In the table, the ns are reported in brackets to distinquish them from the percentage calculations. Percentages were calculated based on the number of information seeking situations of each type divided by the total number of sense-making instances that a particular professional group encountered. For instance, for auditors 4.3% of their sense-making instances were coded as task initiating information seeking situations.

An examination of Table 1 and Figure 8 showed that for the most part -- in fact for 90% of all sense-making instances -- the three professional groups shared common constructions of information seeking situations. Each group described at least an average of one sense-making instance in each of the first seven information seeking situation types -- task initiating, focus forming, idea assuming, idea rejecting, idea confirming, idea finalizing, and idea sharing. The commonality between the professions is most clearly shown in Figure 8 where one sees that for the first seven information seeking situation types there is remarkable similarity in the rise and fall of the bars indicating the percentage of sense-making instances for each professional group which were coded into each of the situation types.

There were three information seeking situation types which were reported primarily by architects -- approval granting, design generating, and approval seeking. These situations appeared to be built into the architects' work processes. For instance, generating design concepts represented a typical work task for the architects. With regard to approval granting, the architects were required to submit their building plans for government approval before they could proceed to construction work. Further, they were required to oversee the construction process and issue certificates and letters to the contractors which placed them in approval granting situations. These work processes are specific to architects (but not to auditors and engineers), and thus impacted architect reports of the types of information seeking situations which they saw themselves encountering. For architects, 38 (17.3%) of 215 sense-making instances fell into these three situational types. For engineers, the comparable percentage was 1.2%, for auditors, 0.0%.

Among the seven information seeking situation types that were common to all three professional groups, there were remarkably few differences in the frequencies with which the professionals saw themselves in these situations. There were, however, a few exceptions.

First, auditors reported more frequently seeing themselves in focus-forming situations than engineers and architects -- 26.2% of auditor sense-making instances were coded as focus forming information need situations compared to 10.3% for eningeers and 9.8% for architects. As pointed out by the auditors, they act as external parties who check on business entities once or twice a year. Thus, they are not as familiar with how the business entities function, nor as aware of the changes evolving within the company over time. A great portion of their audit time is used to come up with an audit scope, in order to decide on which specific areas to investigate in depth within the audit period. The following quotations provide some examples to show the emphasis that auditors placed on forming a better focus for of the audit scope:

As an auditor, you always constantly face setbacks . . . . If you come across any setback, you go back and reconsider your objectives and reformulate your strategy and go in again. It's like driving a car -- start, in, stop, and change gears again. It is always changing. It's the . . . puzzle of internal auditing. If you go by one approach across the board for all industries, I think it's going to be very tough. Tough because of different cultures, different sets of people -- they react differently. Different kinds of industry, different inherent problems with the company, in fact industry. The worse thing is the management is different, because the management plays a very important part. So there are many factors involved. Then perhaps through trial and error, you will pick up on that (Auditor #3).

[For] internal auditing, your job is very varied. You may be auditing accounts one year, and next year you're auditing a settlement department. All this information you need to get it from different people, from different sources. There is no one standard source which provides all the information you need. So as an auditor, you have to use your own initiatives and discretion where to get information, whether you can get it here or there (Auditor #6).

There is a prior stage, which is called the audit planning stage. We have to plan the audit, and we've got to understand the client's business, which is the most important thing in any audit. If you do not understand the client's business, you would not know what is reasonable, what is the focus of the audit, and what you are looking for. If you know what you are looking for, you don't have to do unnecessary work to gather evidence (Auditor #8).

A second distinction between the three professions was that engineers reported somewhat more frequently seeing themselves in idea rejecting situations involving the handling of conflicting and unexpected information. Thus, 18.1% of the engineers' sense-making instances were coded as idea rejecting information seeking situations compared to 12.9% and 12.6% respectively for the auditors and architects sense-making instances. As pointed out by the engineers, their work role is to ensure that the electronic products manufactured by their companies are up to specified quality standards. Most engineers chose to share as their typical tasks how they tackled problems related to product failure. They explained that a great portion of their working time goes to investigating the causes of product failures, and to exploring why testing results were unexpected. The following quotation exemplifies how the engineers experienced these idea-rejecting situations:

For this project, I am dealing with an external vendor, and the vendor is in the US. So my task is to do the testing on the product before we accept their proposal to buy their chip. First, I do the testing. Second, I tell the vendor my test findings. Then I wait for their update, and then I do the test again. For example, for my sound card, there will be a driver update if there are any bugs in the driver. So after I receive the new driver, I'll do another round of testing. So this process is actually repeating (Engineer #7).

Thirdly, architects reported less frequently seeing themselves in idea confirming situations than auditors and engineers. For architects, 9.3% of sense-making instances were coded as idea confirming situatons compared to 16.4% and 17.4% for auditors and engineers respectively. This was due to the architects' structured work process as defined by laws in Singapore. Especially when confirming building plans and design concepts, architects reported that they had little to no power. Instead, they saw themselves as being in approval seeking situations -- being led by others and having to wait for others' approval. For example, one architect mentioned she wanted to confirm whether a piece of state land in front of her client's property could be used for a carriage way to her client's house. However, as her narrative shows, the Land Office in Singapore would not entertain her questions:

The Land Office doesn't entertain questions. It's only after we make submission the Development Control Division will consult the plan, then submit the plan to the Land Office to consult them on this piece of land. Say there is a piece of state land in front of the development, so the land office will give the advice and say the client has to make this piece of state land into his land....There is a risk behind every project. Other than that, sometimes certain areas are under comprehensive development guide plan, DGP as they call it. Those are confidential until they are published to the public. Those are also the things that we are very worried about....We might be proposing, working very hard on something. At the end of the day, this piece of land is acquired for North-East MRT line. So that's it. That's the end of the project (Architect #7).

In summary, this study used the Sense-Making Methodology to develop a typology of ten information need situations, which were referred to in this study as information seeking situation types. These situation types were extracted inductively by examining descriptions of sense-making instances faced by employees in work tasks in three professions -- auditing, engineering, and architecture. The Sense-Making Triangle metaphor was applied deductively to isolate situations, gaps, and uses/helps as described by each respondent for each sense-making instance. The identification of information seeking situation types, then, involved an intersection of the application of a deductive analytical tool from Sense-Making with the inductive force of the data.

From this analysis, seven information seeking situation types were found to be both common and similiarly frequent for all three professional groups. However, some differences across professional groups were noted. First, the architects reported three additional "information seeking situation" types, which were not reported by auditors, and scarcely reported by engineers. These additional types were approval granting, design generating, and approval seeking. Second, auditors reported more often facing focus forming situations while engineers more often faced idea rejecting situations. Finally, architects reported proportionately fewer idea confirming situations because their occupational roles rarely gave them final say.

Thus, by employing the Sense-Making Methodology to study information seeking and use as embedded in time and constructed in movement through time-space, an important finding emerged: that professionals belonging to three different workplaces encountered seven common "information seeking situation" types. The extraction of these commonalities is valuable because previous studies have mainly highlighted differences rather than commonalities between workplace domains.

This study represents, of course, only a modest beginning. Further qualitative and quantitative research needs to examine whether the information seeking situation types which emerged in this study are applicable to larger samples of these professional groups as well as to professionals involved in other contexts. Further research also has to be conducted to explore additional possible information seeking situation types which have not appeared in this study. And, quantitative predictive analyses need to be conducted that explicitly test the basic proposition which drives this entire study: that there are systematic ways to examine information seeking and use that go beyond the traditional approaches which assume that information seeking and use varies across time-space by domain, or across time-space based on enduring characteristics of information seekers. The systematic entry proposed here focuses on attending to and categorizing systematically micro-moments of gap-facing (i.e. sense-making instances) by utilizing Sense-Making's verbing conceptualizations as methodological tools.

Of course, in proposing the idea of the seven common and three unique information seeking situations, we must emphasize that these are a result of the intersection between the realities of these information seekers as they described their situations and the deductive organizational tools derived from Sense-Making. The question remains: what is the most useful way to typologize these information seeking situation categories? In this study, we allocated separate category status to sense-making instances involving the seeking and granting of approval. Conceptually, however, these categories could be redefined as idea confirming and idea finalizing situations respectively. By going up this level of abstraction we find a commonality between architects, auditors and engineers which the lower level data analysis did not. An important empirical question to ask is what level of abstraction in categorizing the information seeking situation types yields the most useful typology in terms of research prediction and, even more importantly, potential practice applications. This is an important illustration of a central premise that underlies the Sense-Making Methodology -- that our methodological moves have as much to do with our research results as the warrants of our data collections. These are inextricably tied to each other and must be addressed simultaneously.

One of the major limitations of traditional approaches to conducting user studies of information seeking and use is that despite numerous studies in different contexts, researchers still have limited clues as to how to make use of the scattered findings to arrive at higher levels of generalization across contexts (Kuhlthau, 1998; Reneker, 1993). In most of these studies, researchers have adopted a "domain" approach as they aimed to find constancies of information seeking and use behaviors across time for people belonging to particular workplace domains. As a result, many independent findings have emerged that highlight differences between domains, but the findings seem to have limited utility in developing higher level generalizations across contexts.

Contrasting with the "domain" approach is an emerging "dynamic" approach, whereby researchers propose that even within a domain, there are differences in people's information seeking and use behaviors at different times, usually construed as in different situations and/or different contexts. Researchers have called for the need to study why a person might engage in one form of information-seeking behavior at a particular time (Borgman, 1996; Belkin, 1983; Dervin, 1992; Katzer and Fletcher, 1992). Research responding to this mandate has yielded a research genre focusing on information seeking and use in context. The mandate implies, of course, that what is neded is systematic classification of contextual and/or situational factors. This study aimed to provide one such classification by utilizing the meta-theoretical and methodological tools offered by Sense-Making.

[1] The data in this paper comes from the dissertation in progress of the senior author. A preliminary report was published as Cheuk (1998).

[2] Two international conferences focusing on "Information Seeking in Context" have been held recently: in August 1996 at the University of Tampere, Finland; in August 1998 at the University of Sheffield, United Kingdom. Another is scheduled for August 2000 in Goteborg, Sweden. Papers at these conferences illustrate the increasing number of contexts in which user studies are conducted and the increasing variety of conceptual approaches being brought to bear. Proceedings are available for the first conference: see Vakkari, Savolainen, and Dervin (1997).

[3] See, for example: Atwood & Dervin, 1982; Dervin, Nilan & Jacobson, 1981; Dervin, Jacobson, & Nilan, 1982; Dervin & Clark, 1987.

[4] See, for example: Dervin, in press; Dervin, 1998.

[5] The most recent comprehensive understanding of how Sense-Making meta-theory has informed methodology and methods to study human information needs, information seeking and use behaviors is found in Dervin (in press).

[6] See, in particular: Dervin, in press; Dervin, 1998.

[7] See, for example: Chen & Hernon, 1982; Dervin, 1992.

[8] The category schemes used here come primarily from: Dervin, 1983; Dervin, 1991.

[9] All interviews were conducted by the senior author.

Allen, T. J. (1977). Managing the flow of technology: technology transfer and the dissemination of technological information within the R&D organization. Cambridge, MA: MIT Press.

Atwood, R., & Dervin, B. (1982). Challenges to socio-cultural predictors of information seeking: A test of race versus situational movement state. In M. Burgoon (Ed.), Communication Yearbook: Volume 5, (pp. 549-569). New Brunswick, NJ: Transaction Books.

Belkin, N. J. (1983). Anomalous states of knowledge as a basis for information retrieval. Canadian Journal of Information Science, 5, 133-144.

Borgman, C. L. (1996). UCLA-NSF Social Aspects of Digital Libraries Workshop held at UCLA, February 15-17 1996: Final Report to the National Science Foundation. Los Angeles: University of California.

Bystrom, K., & Jarvelin, K. (1995). Task complexity affects information seeking and use. Information Processing and Management 31(2), 191-213.

Chen, C. C., & Hernon, P. (1982). Information seeking: Assessing and anticipating user needs. New York: Neal-Schuman Publishers, Inc.

Cheuk Wai-yi, B. (1998). An information seeking and using process model in the workplace: A constructivist approach. Asian Libraries, 7 (12), 375-390.

Crawford, S. (1978). Information needs and uses. Annual Review of Information Science and Technology, 6, 1-39.

Dervin, B. (1983, May). An overview of sense-making research: Concepts, methods, and results to date. Paper presented at the International Communications Association Annual Meeting, Dallas, TX.

Dervin, B. (1991). The content analysis of Sense-Making responses: An introduction to methodology and method. The Ohio State University: Department of Communication. Unpublished paper.

Dervin, B. (1992). From the mind's eye of the "user": The Sense-Making qualitative-quantitative methodology. In J. D. Glazier & R.R. Powell (Eds.), Qualitative research in information management (pp. 61-84). Englewood, CO: Libraries Unlimited.

Dervin, B. (1998). Sense-Making theory and practice: An overview of user interests in knowledge seeking and use. Journal of Knowledge Management, 2(2), 36-46.

Dervin, B. (in press). On studying information seeking methodologically: The implications of connecting metatheory to method. Information Processing and Management.

Dervin, B., Nilan, M., & Jacobson, T. (1981). Improving predictors of information use: A comparison of predictor types in a health communication setting. In M. Burgoon (Ed.), Communication Yearbook: Volume 5 (pp. 807-830). New Brunswick, NJ: Transaction Books.

Dervin, B., Jacobson, T. & Nilan, M. (1982). Measuring information seeking: A test of a quantitative-qualitative methodology. In M. Burgoon (Ed.), Communication Yearbook: Volume 6 (pp. 419-444). New Brunswick, NJ: Transaction Books.

Dervin, B., & Clark, K. (1987). Asking significant questions: Alternative tools for information need and accountability assessments by libraries. Report to California State Library, July 1987. Sacremento, CA: California State Library.

Dervin, B., & Nilan, M. (1986). Information needs and uses. Annual Review of Information Science and Technology, 21, 3-33.

Ellis, D. (1989). A behavioral approach to information retrieval system design. Journal of Documentation, 45(3), 171-212.

Ellis, D., & Haugan, M. (1997). Modeling the information seeking patterns of engineers and research students in an industrial environment. Journal of Documentation, 53(4), 384-403.

Fidel, R. (1987). What is missing in research about online searching behavior. The Canadian Journal of Information Science, 12(3/4), 54-61.

Hewins, E.T. (1990). Information needs and use studies. Annual Review of Information Science and Technology, 25, 145-174.

Hogeweg-DeHaart, H. P. (1983). Characteristics of social science information: A selective review of the literature, Part I. Social Science Information Studies, 3, 147-164.

Katzer, J., & Fletcher, P. T. (1992). The information environment of managers. Annual Review of Information Science and Technology, 27, 227-263.

Kuhlthau, C. C. (1993). Seeking meaning: A process approach to library and information services. Norwood, NJ: Ablex.

Kuhlthau, C. C. (1998). Investigating patterns in information seeking: Concepts in context. Paper presented at the Information Seeking in Context conference, August 1998, Sheffield, United Kingdom.

Reneker, M. H. (1993). A qualitative study of information seeking among members of an academic community: Methodological issues and problems. Library Quarterly 63(4), 487-507.

Vakkari, P. (1997). Information seeking in context: A challenging meta-theory. In P. Vakkari, R. Savolainen, & B. Dervin (Eds.), Information seeking in context (pp. 451-464), London, UK: Taylor Graham.

Vakkari, P., Savolainen, R., & Dervin, B. (Eds.). (1997). Information seeking in context. London, UK: Taylor Graham.

Wilson, T. D. (1999). Models in information behaviour research. Journal of Documentation, 55(3), 249-270.

Copyright 1999 Communication Institute for Online Scholarship, Inc.

This file may not be publicly distributed or reproduced without written permission of the Communication Institute for Online Scholarship, P.O. Box 57, Rotterdam Jct., NY 12150 USA (phone: 518-887-2443).