Decision making framework for thinking about
information seeking behavior
Irene Lopatovska
Rutgers University, School of Communication, Information and Library Studies, 4
Huntington St., New Brunswick, NJ 08901 irenelo@eden.rutgers.edu
irenelopa@comcast.net
At the time of constantly emerging technologies, information professionals need
frameworks for modeling users’ preferences as well as understanding adoption and
usability issues. One such framework, the decision making Expected Utility theory, is
reviewed for its ability to offer the logic of mathematical modeling with graphical aids
for analyzing information seeking decisions. Major information seeking decisions,
including decisions to seek information, select information sources and search
strategies, select relevant information objects, and decisions to stop seeking, are
analyzed using the four expected utility components: alternatives, outcomes,
likelihoods and values. Application of decision theoretic framework helps to identify
under-studied areas of information seeking research, as well as offers new ways of
analyzing information preferences.
1. Introduction
Decision making is a major component of information seeking behavior (ISB). Information
seekers face choices to seek or not to seek information (Timmermans, 1993), to select some
information objects over others (Wang, 1994), end or continue search (Kantor, 1987).
Understanding information seeker’s decisions and preferences for various information objects
and processes is at the center of most of the information retrieval and human-computer
interaction studies. Yet, few information scientists have considered applying decision theoretic
approach in information seeking context. Kraft (1973) suggested applying decision theory to
describe information retrieval (IR) process. According to Kraft, the IR system alternatives are to
retrieve or not to retrieve a document; the outcomes of the system “decisions” are documents,
the values associated with document outcomes are relevance of documents to a searcher’s
query, and likelihood equates to the estimations of probability of a document being relevant to
a query. While Kraft’s article offers a good introduction to the decision theory and interesting
application of this theory to the IR process, it addresses the issues of machine “judgments”
rather than human judgments. Fidel (1991) used decision tree technique for analyzing term
selection routines of online searchers. Kantor (1986) suggested applying decision theoretic
approach for understanding IR user stopping behavior. Blackshaw and Fishhoff (1988) applied
decision theory in the analysis of online searching behavior.
All the above mentioned studies suggest a narrow application of the decision theories to the
specific ISB domains. A broad application of the decision making framework in the ISB
research can offer quantitative modeling techniques for understanding seekers’ preferences of
information objects and processes. This paper illustrates how some of the ISB phenomena can
be viewed through the lens of decision making theory and suggests directions for future
research.
2. Decision making components
This paper refers to a normative Expected Utility theory developed by von Neumann and
Morgenstern (1967). While normative models are criticized for not accurately describing
behavior (Gigerenzer, 2004), they have a virtue of being formally explicit and tractable, and can
be used for making quantitative predictions in various contexts (Sanfey, Loewenstein, McClure,
& Cohen, 2006). Most of the ISB research is quite successful in describing various information
seeking decisions phenomena. However, it is lacking a benchmark theory that would unify
scattered concepts, discriminate between normative (e.g., common, beneficial to seeker) and
deviant (e.g., exceptional, detrimental) behavior, and offer ways to quantify, model and predict
behavior.
Decision making theories attempt to describe and model situations in which a choice is made
using four major components:
Alternatives or all the strategies available to a decision maker for a given situation
Outcomes or potential consequence of the decision if a certain alternative is chosen;
Likelihood or probability of various outcomes based on the states of nature;
Values that are associated with outcomes and are based on a formula or a cognitive
process (Roberts & Sonnenberg, 2000).
Figure 1. Decision tree (from Huber, 1997).
We illustrate the connection between decision components using a decision tree - one of the
common visual modeling techniques in decision research. In Figure 1 (Huber, 1997), the
square on the left represents a choice node; it is the start of the decision making process when
decision maker has full control over the alternatives. Branches represent different alternatives
which lead to different outcomes. Selection of an alternative narrows the outcome range to the
branches below the chosen alternative. . Outcomes depend on various events, or states of
nature, which decision maker cannot control (chance node is represented by a circle). Different
states of nature determine probabilities of the outcome occurrences. Each outcome is
associated with a value a decision maker assigns to it.
We can draw a simple ISB process using the decision tree technique. Lets imagine that a
person has an information need. The alternatives for addressing the need include 1) seeking
information on the internet, 2) seeking information in a library, or 3) not seeking information at
all. Any alternative can lead to finding or not finding the needed information, however, the
likelihood of finding information varies between alternatives. For example, the probability of
finding information on the internet is 50%, finding it in the library is 80%, and finding it without
seeking is 1%. Note, that the probabilities within alternative’s branches should add up to
100%. For example, if probability of finding information in a library is 80%, the probability of
not finding it in a library should be 20% so that the probability of all the states in “seek
information in the library” alternative adds up to 100% (80%+20%=100%). Note as well, that
the probabilities of finding information across the entire search decision tree need not add-up
to one if we can be expected to obtain the correct information from more than one source.
The outcomes of the decisions represent desirability of goals that decision maker wants to
achieve (Baron, 2000b). The goal in our example could be obtaining the most accurate
information with the least effort. The outcomes are associated with utility (u), a measure of
extend of goal achievement (Baron, 2000b). The hassle of finding information by going to a
library is significant. However, if going to the library is the best way to obtain the most accurate
and complete information then the utility of the outcome of going to the library and finding
information represents the difference between the value of quality information and the hassle
of obtaining it (u=.8). Using similar logic, we assign utility values to all outcomes: utility of
finding medium quality information with medium effort on the internet is .7, not finding
information on the internet is .3, not finding information in a library after making an effort of
going there is .1, utility of finding information without any effort is maximum value of 1, and
not finding it without searching is .2.
Now we can decide on the best alternative by considering probabilities and desirability of the
outcomes. The goodness of alternative is measured in expected utility (EU). It is calculated for
each alternative by multiplying the probability of each outcome by its utility, summed across
outcomes:
where EU stands for expected utility, ui sand pi stand for utility and probability of the ith
outcome. The premise of the expected utility theory is that rational decision makers are trying
to maximize EU and therefore select options with the highest EU value. In our example,
The largest EU value of seeking information in the library (EU=.66) indicates that this choice
would be the most optimal, followed by seeking information on the internet (EU=.5) and not
seeking information at all (EU=.208).
A reader is advised to remember this example throughout the following sections of the paper.
Every reference to alternatives, outcomes, likelihoods or values in the context of ISB decisions
can be associated with numbers that can be plugged into the model for calculation of the
optimal decision, decision with the highest EU value.
We used Expected Utility theory to analyze our information seeking choice example. However,
there are other decision theories, including Multi-Attribute Decision theory (Baron, 2000a) and
Prospect theory (Kahneman & Tversky, 1979). The differences between these theories,
including techniques for assessing outcome utilities, likelihoods of outcomes and states of
nature, are beyond the scope of this paper. None of the decision making theories accurately
describe or predict human behavior 100% of the time. However, normative theories provide a
conceptual framework for thinking about behavioral phenomena.
3. Decision making components in information seeking behavior (ISB)
Information seekers make decisions at different stages of the seeking process. Based on the
information seeking research of Blackshaw and Fishhoff (1988), Wilson (1997) and others, we
can identify the following types of decisions:
decisions to seek or not to seek information,
decisions related to specific search strategies,
decisions related to selection of relevant information objects, and
decisions to continue or stop seeking information.
3.1 Decisions to seek or not to seek information
3.1.1 Alternatives
Alternatives that are available to a person with an information need are “to seek” or “not
to seek” information. The variations of these alternatives include to seek information
now or to seek it later. Most of the ISB research assumes that when a person faces a gap
that prevents her from making sense of a situation (Dervin & Nilan, 1986), an anomalous
state of knowledge (Belkin, 1978), need to confirm/ disprove an inconsistent idea
(Wildemuth, de Bliek, Friedman, & Miya, 1994), or uncertainty (Repo, 1989; Horne,
1983; Urbany, Dickson, and Wilke, 1989), the preferred alternative is to seek information
to bridge the gap, gain knowledge or resolve uncertainty. However, there is some
research indicating that seeking information to resolve uncertainty is not always the
preferred alternative. Allen (1996) and Wilson (1997) describe situations of health
information seeking when individuals prefer not to seek information. One of the
explanations for such preference is that once people receive information they will face
the burden of selecting a course of action. Another explanation is avoidance of
information related to a thread, such as a disease, in a stressful situation.
An individual might choose not to seek information, and rely on stereotypes or other
heuristics for closing the cognitive gap or resolving uncertainty. Timmermans (1993) and
Macrae, Hewston, and Griffiths (1993) found that increased decision complexity and
information processing reduces the amount of information seeking and increases
reliance on stereotypes’ heuristics.
3.1.2 Outcomes
The outcomes of seeking information could be grouped into two general categories:
cognitive and tangible. An example of a cognitive outcome would be changed knowledge
manifested in new word associations (Belkin, Cool, Stein, & Thiel, 1995), readiness to
make a decision (Hirshleifer & Riley, 1992) or readiness to make progress in tasks
(Kuhlthau, 1991). The tangible outcomes include research, dissertation, or exam
(Saracevic and Kantor (1997); pleasure from learning something new, staying current,
educating others or relaxing (Krikelas, 1983; Savolainen, 1995; Rioux, 2005).
3.1.3 Likelihood
It is hard to find literature that refers to the likelihood estimations in information seeking
judgments. A possible reason for the scarcity of literature is that people in general have
difficulty formalizing likelihood estimations, especially in terms of probabilities
(Gigerenzer & Hoffrage, 1995). Blackshaw and Fishhoff (1988) point to the fact that
online system users might not even address all stages of the decision making process.
However, the probability estimations can be inferred from information seekers’
preferences or believes about information seeking success or failure. For instance,
Wilson (1996) referrers to cases when a person decides not to seek information because
of the lack of belief in personal success. In such case, a person might realize the benefits
from seeking information and place high value on the outcome, but estimate very low
probabilities of a successful search. The low expected utility of a search (a sum product
of the outcome’s value and probability) leads to the choice to not seek the information. In
cases when probability estimates of successful information seeking are high enough to
contribute to high expected utility values, individuals prefer to engage in information
seeking.
3.1.4 Values
Discussion about the value of information seeking is linked to the discussion about the
value of information. There is no consensus about the nature of information value among
information scientists, economists and other researchers dealing with the issue (Bates,
1988; Repo, 1989). However, most of the researchers agree that the value of
information derives from its usefulness at present or future times. The value of
information seeking can be derived by examining the implications of finding information
and addressing the question: does improved knowledge make an individual better or
worse off considering the costs of obtaining the information?
One of the approaches to estimating the value of information seeking is risk and reward
model. Stigler (1961) suggests that decisions to seek information are influenced by
evaluations of risks and rewards involved in seeking. Information is sought when the
rewards from seeking are higher than from not seeking. The risk is measured in terms of
cost, time, and convenience. Godbold (2006) lists more risks that might lower the value
of information seeking, among them, emotional risks, physical risks, or political and
social risks.
The value of seeking and the level of investment in seeking are largely determined by the
perceived importance of the task or the decision requiring information (Zach, 2005).
3.2 Decisions related to specific search strategies
3.2.1 Choices of information sources
3.2.1.1 Alternatives
There are various alternatives for obtaining information. People can learn from personal
communication with professionals, such as financial advisors, physicians. Alternatives
include consulting the media, such as television or newspaper (Allen, 1996), a reference
desk, electronic references resources, book or film, or purchasing material (Saracevic
and Kantor, 1997).
3.2.1.2 Outcomes
The outcomes in judgments related to selection of information sources are potential
consequences of consulting IR systems, and/or libraries, an expert, purchasing a book or
using other source for obtaining information. An example of the desired outcome would
be obtaining relevant information at the minimum cost and the maximum convenience.
3.2.1.3 Likelihood
Availability and accessibility of sources are two researched criteria pertaining to users’
assessments of probabilities of finding information (Schamber, 1991; Saracevic &
Kantor, 1997; Saracevic, 2004). If a system is not accessible or available, the probability
of it being useful is low. So even in situations when a user knows that the search of a
particular source might be very fruitful (high value search, U = 1), she will not select the
source due to its low availability and low total expected utility of the choice.
3.2.1.3 Values
There seems to be no literature on a single measure of information source goodness,
something that could be compared to the concept of information object utility discussed
in section 3.3. However, there is abundant literature on the various aspects contributing
to the overall value of information source. Some of these aspects include quality, trust,
authority, affective reactions associated with the source (Halpern & Nilan, 1988; Belkin,
1978; Zach, 2005; Auster & Choo, 1991; Palmer, 1991).
Other aspects contributing to the evaluation of information sources include time and
costs considerations (Taylor, 1986; Marchionini, Plaisant, & Komlodi, in press) and
knowledge of the system content (Hert, 1996).
3.2.2 Choices of search tactics
3.2.2.1 Alternatives
In the context of the mediated information seeking, alternatives include deciding on
different ways to seek an advice: consult informal mediators (e.g., family and friends) or
formal mediators (e.g., teachers and librarians) (Kuhlthau, 1993); meeting with
mediators face to face, or getting in touch with them using telephone, email or fax.
Different formats of the information request, such as formal or information questions,
would also be considered alternatives.
In the context of human-machine information seeking, alternatives include various
strategies of searching for information, and more specific considerations related to query
terms selections (Marchionini, 1995; Bates, 1989).
The decision theoretic approach is not new to the research in query term formulation.
Fidel (1991) used decision tree technique for analyzing term selection routines. Fidel
(1991) notes that while query formulation is an intuitive process, searchers use rules that
can be mapped onto the formal models.
3.2.2.2 Outcomes
In the context of interpersonal information seeking, the outcomes would be
consequences of written or verbal expert responses. In the context of IR system
interaction, the desired outcome would include relevant information object (e.g.,
document) or a set of retrieved information objects. Information object outcomes are
described in Section 3.3. We did not find literature analyzing information object sets as
the outcomes of the search tactic judgments.
3.2.2.3 Likelihood
We did not find literature on the estimations of probabilities of success for various search
strategies. However, probability estimates can be inferred from the information seeker’s
preferences. For example, most of the information seekers would place a high value on
an expert advice (Belkin, 1978). However, in some instances the probability of this
outcome might be low due to the lack of expert’s availability or approachability,
contributing to the low overall expected utility of this strategy. Another example when
probability estimates might change the overall preference is the use of Boolean
operators. While an online searcher might be aware of the benefits of Boolean operators,
she might not believe in her abilities to use them correctly, and consequently avoid using
Boolean operators in online search (Savage-Knepshield & Belkin, 1999).
3.2.2.4 Values
Discussion about the value of information tactics’ outcomes follows the discussion about
the value of information. Information seekers put high value on relevant information that
leads to changes in cognitive (Belkin, et al., 1982) and contextual (Harter, 1992) states,
and low values on irrelevant information. Individuals select tactics that have high
expected utility values derived from high outcome values and/or probabilities.
The values of various search terms depend on user preferences for minimal cognitive
effort and energy required to formulate the query. The ultimate value of a query is
determined by its effectiveness in producing quality information (Wacholder & Lui,
2006).
3.3 Selection of relevant information objects.
3.3.1 Alternatives
After conducting a search or inquiring about information, the seeker chooses information
that is perceived as the most relevant to the need. The literature on the alternatives such
as abstracts or full-text articles is particularly abandoned in the context of information
retrieval (Harter, 1992; Mizzaro, 1997; Wang, 1994).
3.3.2 Outcomes
The outcomes of selection process are information or information objects that are
chosen to address information needs. The outcomes of information selection are not
known until information is found, examined and evaluated. An example of a decision tree
would include document n1 through nx as alternatives. Each of these alternatives would
have “use “or “not-use” branches resuting in outcomes similar to “how better off will I be
by using/not using this information”. Selection of relevant objects is an unusual decision
since multiple or all alternatives can be considered for use. The uncertain situation of not
knowing whether any of the information will be useful contributes to the feeling of
anxiety often observed during information seeking (Kuhlthau, 1993).
3.3.3 Likelihood
Users’ familiarity with systems and perceptions of their quality contribute to their
assessments of probabilities of obtaining useful results. For example, knowledge of
system’s recall/precision (Tague-Sutcliffe, 1992), response time, capacity and error rate
(Saracevic, 2004) might contribute to high or low probability assessments of quality
outcomes. Searcher might not select information objects that are associated with low
perceived quality due to the poor authority and/or credibility estimations (Wilson, 1983).
Searchers are more likely to select information objects that they deem more likely to be
trustworthy and of good quality (Reih, 2002).
3.3.4 Values
Research in values of information objects is the most prolific ISB area related to
judgment and decision making. The value of information objects is usually associated
with its relevance as judged by a searcher (Schamber, 1994). There are two treatments
of relevance (or relevance-like) concepts in the literature: multi-dimensional and holistic.
Multi-dimensional relevance can be viewed as a sum of values assigned to different
aspects of information objects (Wang, 1994). ISB authors identified a number of relevant
information objects’ features: Barry (1994) identified 23 relevant criteria pertaining to
the content of a document, user’s knowledge, beliefs and preferences; Cool, Belkin, and
Kantor (1993) identified 60 criteria pertaining to documents’ topicality (being on the
searched topic), content, format, and presentation; Schamber (1991) identified 10
criteria grouped in categories pertaining to information, source, and presentation. Other
authors who analyzed multi-dimensional relevance include Ingwersen (1992), Vickery
(1959), Eisenberg and Schamber (1988), Regazzi (1988), Lancaster and Warner (1993),
Rees and Schultz (1967).
Understanding specific information object characteristics valued by users is important for
improving information systems and objects. However, for the analysis of searchers’
judgments, one value measure suffices; particularly since people intuitively know what
information objects are good or bad without necessarily differentiating between various
relevant aspects of an object (Saracevic, 1996; Mizzaro, 1997). Holistic approach to
relevance treats information object’s value as a measure of overall information object
goodness (Tague-Sutcliffe, 1992). The measures of the overall goodness include
satisfaction (Tessier, Crouch, & Atherton, 1977; Su, 1994; Gluck, 1996), satisficing
(Agosto, 2002), affective relevance as an emotional response to the document (Cosijn &
Ingwersen, 2001), value of search results as a whole (Su, 1998), emotional value (Wang
& Soergel, 1998), utility as usefulness (Saracevic, 1996; Saracevic & Kantor, 1997),
utility as global usefulness or goodness of objects (Cooper, 1973; Brookes, 1980;
Saracevic, Kantor, Chamis, and Trivison, 1988), and instant utility as a measure of
instant episode goodness (Rorvig, 1985; Lopatovska & Mokros, (in press). The overall
value concept closest to the decision making utility theory is Cooper’s (1973) utility,
defined as whatever the user finds to be of value about the system output: usefulness,
entertainment or aesthetic value. It is also a measure of the system’s worth and
usefulness to the user.
3.4 Deciding to continue or stop seeking information.
To some extend, decisions to continue or stop searching are similar to the decisions to
engage in searching in the first place. The differences between start and end decisions
are related to cost-benefit estimations: at the beginning of search the benefits of
searching outweigh the costs, at the end of the search costs of further searching
outweigh the benefits.
3.4.1 Alternatives
At any point of search, information seeker has an alternative to stop searching. Stopping
decision depends on a selected search strategy. In some situations, information seekers
decide to stop soon after the search begins. Kruglanski, Peri and Zakai (1991) have
found that people who are confident in understanding the situation and want to decide
quickly do little information seeking. Blair (1996) found that lawyers search just enough
information to win the case. Stopping when “just enough” information to satisfy the need
is found (Zach, 2005) is usually explained by the concept of satisficing developed by
Simon (1956).
In the circumstances that require extensive research, such as working on school papers
(Kuhlthau, 1993) or projects that have high organizational impact (Zach, 2005),
information seekers arrive at stopping decisions at later points.
3.4.2 Outcomes
The outcomes of choices to stop searching are implications of stopping. By studying
information behavior of administrators in a non-profit organization, Zach (2005)
concluded that the outcomes of searching are associated with the arbitrary level of
comfort with the found information. The level of comfort is largely determined by the
nature of task: for important tasks, the comfort level tended to be high; for simple tasks,
the threshold is often quite low. If the found information does not address searcher’s
needs, level of comfort or other predetermined goal, the search continues; when the
goals are met and needs are addressed, the search ends (Zach, 2005).
3.4.3 Likelihood
Kantor (1987) refers to the Bayesian model of probability estimations for analyzing
stopping decisions. Per author, the estimates of the probabilities of success change
during search. As a searcher finds good information objects, she estimates that it is more
probable to find more good objects; as a searcher starts finding bad objects, estimates of
probabilities of successful search drop. A decision to stop is made when the probabilities
of success are very low, or when in other words, searcher does not expect to find useful
information. At the end of the search, individual still values the relevant information
higher than the irrelevant. However, because the probabilities of finding relevant
information are now small, they make the overall expected utility (EU) value of searching
very low or negative (Kantor, 1987). Lets imagine that information seeker places a high
value on relevant information (U = 1). At the beginning of the search, a person is
optimistic about finding relevant information and has high probability estimates for
finding this information (with p=1 the EU of initial search is EU = 1(1) = 1). However, as
the search progresses and information seeker lowers her expectation of finding the
relevant information, the overall expected utility of the search diminishes and becomes
unattractive (the value of relevant information is still 1, but the probability is now .1, so
that EU =.1(1) = .1).
3.4.4 Values
Values of information objects have been described in Section 3.3.4. At any point of a
search, including the end of a search, information seeker values relevant information
more than irrelevant information. According to Cooper's (1976) utility theory approach,
searchers stop looking for new objects when the effort of the new search outweighs the
benefit of the new information, or in other words, when the effort of searching is valued
higher than potentially discovered new information.
4. Examples of applying EU framework in HIB situation
Imagine a situation in which an information system administrator is trying to understand or
predict a system’s popularity (e.g., increase or decrease in system’s use). Decision making
framework can help administrator to conceptualize important variables impacting user’s
decisions to use the system. An administrator will have to identify:
information system alternatives available to users (e.g., expert, library, internet,
bookstore);
users’ goals, or desirable outcomes from the use (e.g., entertainment, successfully
performing a job);
the likelihoods of success that users associate with various systems (e.g., confidence in
finding information in the library, lack of confidence in finding it on the web);
values users associate with various outcomes (e. g., finding information for work is more
valuable than finding it for entertainment purposes).
By considering the number and values associated with all these decision components,
administrator, for example, can determine that the system usage changed due to the increase
of the alternatives, changes in user’s goals (user might have used system primarily for work,
but now uses it for entertainment purposes), changes in user’s estimates of success (they are
more familiar with the system and therefore more confident in its successful use), or changed
values.
Let’s review another example. A system designer is deciding which search options to
incorporate into IR system. A designer knows that Boolean option will retrieve accurate search
results. However, does it mean that the search feature should be incorporated? Not
necessarily. If the designer determines that user’s expected utility from other search options is
significantly higher (e.g., due to higher familiarity (value) or higher success estimates
(probability)), the designer might choose not to invest into developing a search feature that will
likely go underused.
Imagine a searcher who is involved in a costly search. Is there any ‘scientific’ way of
determining the best course of action? A searcher can use expected utility theory to determine
the best database(s) to search. She can use her prior experience with a database to derive
probability estimates of success (e.g., high if system retrieved good results in the past, low if it
did not), expected outcomes and their values (e.g., high if system produced relevant,
authoritative and otherwise good results, low if it did not) and plug these values into the model.
A searcher can use similar framework for deciding to end the search (e.g., use estimated
probabilities of finding relevant items and value of addition information when considering
invsting additional resources).
5. Conclusion
The aim of this review was to stimulate discussion around potential uses of decision theory
framework in ISB research. Our catalog of ISB decision making concepts is illustrative, not
exhaustive. We found a wide array of ISB literature concerned with judgments and decisions.
Some researchers suggest using decision-making framework in the context of information
system’s design (Blackshaw and Fishhoff, 1988). However, most of the information scientists
continue to navigate information seeking decisions without explicitly relying on decision
making theories, terminology or techniques.
Mapping some of the ISB literature onto the decision making framework allows the analysis of
various decisions-making situations using standardized, and potentially quantifiable
components. The ISB literature sufficiently covers value concepts, especially in the realm of
judgments related to information objects’ relevance. More work is needed in developing
quantitative techniques for representing values in behavior modeling. Insufficient work is done
on the role of probability estimations in judgments to proceed with search, to use particular
search tactics, or to select particular information objects. Some work has been done on
investigating information seeker’s estimations of success with different ISB choices; however,
more work is needed in order to understand seeker’s preferences for information objects and
processes.
References
Allen, B. L. (1996).
Information tasks.
Academic Press San Diego
Auster, E., & Choo, C. W. (1991). Environmental scanning: A conceptual framework for
studying the information seeking behavior of executives. Proceedings of the 54th ASIS
Annual Meeting (ASIS '91) J. M. Griffiths (Ed.), (pp. 3-8). Medford, NJ: Learned Information.
Baron, J. (2000a). Decision analysis and values.
Cambridge University Press:NY, pp. 335-355.
Thinking and Deciding, J. Baron (Ed.),
Baron, J. (2000b). Normative theory of choice under uncertainty.
Baron (Ed.), Cambridge University Press:NY, pp. 223-244.
Thinking and Deciding, J.
Barry, C.L. (1994). User-defined relevance criteria: an exploratory study.
American Society for Information Science, 45(3), 149-159.
Bates, B. (1988).
Information as an economic good.
Journal of the
The Political Economy of Information
V. Moscoe & J. Wasco (Eds.), Madison, WI: University of Wisconsin Press, pp. 76-95.
Bates, M. J. (1989).
Search Interface.
Belkin, N. J. (1978).
science.
The Design of Browsing and Berrypicking Techniques for the Online
Online Review 13(5), 407-424.
Process of documentation: information concepts for information
Journal of Documentation, 34, 55-85.
Belkin, N.J., Cool, C., Stein, A. & Thiel, U. (1995).
Cases, scripts and information-seeking
strategies: On the design of interactive information retrieval systems.
Applications, 9(3), 379-395.
Belkin, N.J., Oddy, R.N. & Brooks, H. (1982).
Expert Systems with
ASK for Information retrieval Part I.
Journal of
Documentation, 38(2), 61-72.
Bell, D., Raiffa, H., & Tversky, A. (1988). Descriptive, Normative, and Prescriptive
Interactions in Decision Making Decision Making: Prescriptive, Normative, and Descriptive
Interaction D. Bell, H. Raiffa & A. Tversky (Eds.), (pp. 9-33). New York, NY: Cambridge
University Press.
Cool, C., Belkin, N. J., & Kantor, P. B. (1993). Characteristics of texts affecting relevance
judgment. Proceedings of the 14th National Online Meeting M. E. Williams (Ed.), New York,
NY. Medford, NJ: Learned Information, Inc.
Cooper, W. (1973). On selecting a measure of retrieval effectiveness. Part I. The “subjective’
philosophy of evaluation. Journal of the American Society for Information Science, 24,
87-100.
Cooper, W. (1973). On selecting a measure of retrieval effectiveness.Part II. Implementation
of the philosophy. Journal of the American Society for Information Science, 24, 413-424.
Cooper, W. (1976). The paradoxical role of unexamined documents in the evaluation of
retrieval effectiveness. Information Processing and Management, 12(5), 367-375.
Cosijn, E. & Ingwersen, P. (2001).
Management, 36(4), 533-550.
Dimensions of relevance.
Information Processing &
Dervin, B. & Nilan, M. (1986). Information needs and uses. Annual Review of Information
Science and Technology, 21, 3-33, White Plains, NY: Knowledge Industry Publications.
Eisenberg, M. & Schamber, L. (1988). Relevance: The search for a definition.
American Society for Information Science, Atlanta, GA, pp.164-168.
Ellis, D. (1989). A behavioral approach to information retrieval system design.
Documentation, 45(3), 171-212.
Ferguson, M. A., & Valenti, J. M. (1991).
The
Journal of
Communicating with environmental risk takers: An
individual differences perspective.
Fidel, R. (1991).
Health Education Quarterly, 18(3), 303-318.
Searchers’ selection of search keys: I The selection routine. Journal of the
American Society for Information Science,42(7), 490-500
Blackshaw, L., & Fischhoff, B. (1988).
Decision making in online search.
Journal of
American Society for Information Sciences, 39, 369-389.
Gigerenzer, G. (2004).
Fast and frugal heuristics: The tools of bounded rationality.
Blackwell Handbook of Judgment and Decision Making D.J. Koehler & N. Harvey (Eds.), (pp.
62-88). New York: Blackwell Publishing.
Gigerenzer, G. & Hoffrage, U. (1995)
Frequency formats.
How to improve Bayesian reasoning without instruction:
Psychological Review, 102(4), 684-704.
Gluck, M. (1996). Exploring the relation between user satisfaction and relevance in
information systems. Information Processing and Management, 32(1), 89-104.
Godbold, N. (2006). Beyond information seeking: towards a general model of information
behaviour. Information research, 11(4). http://informationr.net/ir/11-4/paper269.html
Halpern, D. & Nilan, M. S. (1988). A step towards shifting the research emphasis in
information science from the system to the user: An empirical investigation of
source-evaluation behaviour in information seeking and use. Proceedings of the American
Society for Information Science (pp.169-176), Medford, NJ: Learned Information.
Harter, S. (1992). Psychological Relevance and Information Science.
American Society for Information Science, 43(9) 602-615.
Hayes, R. M. (1993). Measurement of information.
Management, 29(1), 1-11.
Journal of the
Information Processing and
Hert, C. (1996). User Goals on an Online Public Access Catalog.
Society for Information Science, 47(7), 504-518.
Journal of the American
Hirshleifer, J. and Riley, J. G. (1992). The analysis of uncertainty and information: Cambridge
surveys of economic literature. Cambridge University Press Cambridge
Horne, E. (1983). Question generation and formulation: an indication of information need.
Journal of the American Society for Information Science, 34(1), 5-15.
Huber, O. (1997). Beyond gambles and lotteries. Decision Making: Cognitive Models and
Explanations: Frontiers of Cognitive Science. R. Ranyard, W. R. Crozier & O. Svenson (Eds.),
New York, NY: Routledge.
Ingwersen, P. (1992).
Kahneman, D. (2003).
Information retrieval interaction.
Taylor Graham London, England
A perspective on judgment and choice: Mapping bounded rationality.
American Psychologist, 58, 697-720.
Kahneman, D., & Tversky, A. (1979).
Prospect theory: An analysis of decisions under risk.
Econometrica, 47, 313-327.
Kantor, P. (1987).
A model for the stopping behavior of users of online systems.
Journal of
the American Society for Information Science. 38(3), 211-214.
Keren, G. & Teigen, K. (2004).
Yet another look at the heuristics and biases approach.
Blackwell Handbook of Judgment and Decision Making D.J. Koehler & N. Harvey (Eds.), (pp.
89-109). New York: Blackwell Publishing.
Kraft, D. H. (1973).
A decision theory view of the information retrieval situation: an
operations research approach.
24(5), 368-376.
Journal of the American Society for Information Science,
Krikelas, J. (1983). Information seeking behaviour: Patterns and concepts.
Quarterly, 19(2), 5-20.
Drexel Library
Kruglanski, A. W., Peri, N. and Zakai, D. (1991). Interactive effects of need for closure and
initial confidence on social information seeking. Social Cognition, 9(2), 127- 148.
Kuhlthau, C. C. (1991). Inside the search process: Information seeking from the user’s
perspective. Journal of the American Society for Information Science, 42(5), 361-371.
Kuhlthau, C.C. (1993). Seeking meaning: A process approach to library and information
services. Ablex Norwood, NJ
Lancaster, F. W. & Warner, A. J. (1993).
Resources Press Arlington, VA
Information retrieval today.
Information
Lopatovska, I. and Mokros, H. B. (in press). Willingness to pay and experienced utility as
measures of affective value of information objects: users’ accounts. Information Processing
and Management.
Macrae, C. N., Hewston, M., & Griffiths, R. J. (1993). Processing load and memory for
stereotype based information. European Journal of Social Psychology, 23(1), 77-87.
Marchionini, G. (1995). Information Seekers and Electronic Environments. Information
Seeking in Electronic Environments pp. 11-26. Cambridge, MA: Cambridge University Press.
Marchionini, G. & Komlodi, A. (1998). Design of interface for information seeking.
Review of Information Science and Technology (ARIST), 33, 89-130.
Annual
Marchionini, G. and Liebscher, P. (1991). Performance in electronic encyclopedias:
Implications for adoptive systems. ASIS ’91, Systems understanding people, Proceedings of
the 54th Annual Meeting of the American Society for Information Science J. Griffiths (Ed.).
(Vol. 28, pp. 39-48). Medford, NJ: Learned Information.
Marchionini, G., Plaisant, C. & Komlodi, A. (in press).
The people in digital libraries:
Multifaceted approaches to assessing needs and impact.
Digital library use: Social practice
in design and evaluation A. Bishop, B. Buttenfield, & N. VanHouse, (Eds.) Retrieved Oct. 28,
2006 http://ils.unc.edu/~march/revision.pdf
Mizzaro, S. (1997). Relevance: the whole history.
Information Science, 48 (9), 810-832.
Palmer, J. (1991).
style.
Journal of the American Society for
Scientists and information: I. Using cluster analysis to identify information
Journal of Documentation, 47, 105-226.
Rees, A. M. & Schultz, D. G. (1967). A field experimental approach to the study of relevance
assessments in relation to document searching: Final report: Vol. 1. Case Western Reserve
University, School of Library Science, Center for Documental and Communication Research.
Cleveland, OH
Regazzi, J. J. (1988). Performance measures for information retrieval systems - An
experimental approach. Journal of the American Society for Information Science, 39(4),
235-251.
Repo, A. (1989). The value of information: approaches in economics, accounting and
management science. Journal of the American Society for Information Science, 40(2),
68-85.
Reih, S. Y. (2002). Judgement of information quality and cognitive authority in the Web.
Journal of the American Society for Information Science, 53(2), 145-161.
Rioux, K. (2005). Information acquiring-and-sharing. Theories of information behavior K. E.
Fisher, S. Erdelez & L. McKechnie (Eds.), (pp. 169-173). Medford, N.J: Information Today, Inc.
Roberts, M. S. & Sonnenberg, F. A. (2000). Decision modeling techniques. Decision
Making in Health Care: Theory, Psychology, and Applications G.B. Chapman & F. Sonnenberg
(Eds.) New York: Cambridge University Press, 20-64.
Robinson, J. P. & Davis, D. K. (1990). Television news and the informed public: an
information-processing approach. Journal of Communication, 40(3), 106-119.
Rorvig, M. E. (1985). An experiment in human preferences for information in a simulated
information system. Unpublished doctoral dissertation, The University of California, School
of Library and Information Studies, Berkeley, CA.
Rorvig, M. E. (1990). The simple scalability of documents.
for Information Science, 41(8), 590-596.
Journal of the American Society
Sanfey, A.G., Loewenstein, G., McClure, S.M., Cohen, J.D. (2006).
Neuroeconomics:
Cross-currents in research on decision making.
Saracevic, T. (2004).
Trends in Cognitive Sciences, 10, 108-116.
Evaluation of digital libraries: An overview.
Presentation at the DELOS
WP7 Workshop on the Evaluation of Digital Libraries, 4-5 October 2004, Department of
Information Engineering, University of Padua,
Italy. http://www.scils.rutgers.edu/~tefko/articles.htm
Saracevic, T. (1991). Individual differences in organizing, searching and retrieving
information. ASIS ’91. Systems understanding people. Proceedings of the 54th Annual
Meeting of the American Society for Information Science J. Griffith (Ed.), (Vol. 28, pp. 82-86).
Medford, NJ: Learned Information.
Saracevic, T. (1996). Relevance reconsidered. Information science: Integration in
perspectives. Proceedings of the Second Conference on Conceptions of Library and
Information Science. Copenhagen (Denmark): Royal School of Librarianship, 201-218.
Saracevic, T., Kantor, P., Chamis, A. & Trivison, D. (1988). A study of information seeking
and retrieving. I. Background and methodology. Journal of the American Society for
Information Science, 39(3), 161-176.
Savage-Knepshield, P. A. & Belkin, N. J. (1999). Interaction in information retrieval: trends
over time. Journal of the American Society for Information Science, 50(12): 1067-1082.
Savolainen, R. (1995). Everyday life information seeking: Approaching information seeking
in the context of 'way of life'. Library & Information Science Research, 17, 259-294.
Schamber, L. (1991). Users’ criteria for evaluation in a multimedia environment.
Proceedings of the American Society for Information Science. Washington, DC (pp. 126-133),
Medford, NJ: Learned Information.
Schamber, L. (1994). Relevance and Information Behavior.
Science and Technology (ARIST), 29, 3-48.
Annual Review of Information
Simon, H. A. (1956). Rational choice and the structure of the environment.
Review, 63, 129-138.
Stigler, G. J. (1961).
213-225.
The economics of information.
Psychological
Journal of Political Economy, 69,
Su, L. T. (1998). Value of search results as a whole as the best single measure of
information retrieval performance. Information Processing and Management, 34(5),
557-579.
Tague-Sutcliffe, J. (1992). The Pragmatics of Information Retrieval Experimentation,
Revisited. Information Processing and Management, 28(4), 467-490.
Taylor, R. (1986).
Value-added process in the information systems.
Ablex Publishing Corp.
Norwood, NJ
Tessier, J. A., Crouch, W. W. & Atherton, P. (1977).
computer-based literature searches.
Timmermans, D. (1993).
New measures of user satisfaction with
Special Libraries, 68(11), 383-389.
The impact of task complexity on information use in multy
attribute decision making.
Journal of Behavioral Decision Making, 6(2), 95-111.
Tversky, A. & Kahneman, D. (1974).
Judgment under uncertainty: Heuristics and biases.
Science, 185, 1124-1131.
Urbany, J. E., Dickson, P. R. and Wilke, W. L. (1989) Buyer uncertainty and information
search. Journal of Consumer Research, 16, 208-215.
Vickery, B. C. (1959). Subject analysis for information retrieval. Proceedings of the
International Conference on Scientific Information (Vol. 2, pp.855-865), Washington, DC:
National Academy of Sciences.
von Neumann, J., & Morgenstern, O. (1967).
Ed.). Wiley New York
Theory of Games and Economic Behavior (3RD
Wacholder, N. and Liu, L. (2006). User preference: A measure of query-term quality.
Journal of the American Society for Information Science and Technology, 57(12): 1566-1580.
Wang, P. (1994). A cognitive model of document election of real users of IR systems.
Unpublished doctoral dissertation Collage Park, MD: University of Maryland, College of Library
and Information Services.
Wildemuth, B. M., de Bliek, R., Friedman, C. P., & Miya, T. S. (1994). Information-seeking
behaviors of medical students: A classification of questions asked of librarians and
physicians. Bulletin of the Medical Library Association, 82(3), 295-304.
Wilson, T. (2000). Human information behaviour.
http://inform.nu/Articles/Vol3/v3n2p49-56.pdf
Informing Science, 3(2), 49-56.
Wilson, T. (1997). Information behaviour: An interdisciplinary perspective. Proceedings of
an International Conference on Research in Information Needs, Seeking and Use in Different
Contexts, 14-16 August, 1996, Tampere, Finland (pp. 39-50). Los Angeles, CA: Taylor Graham.
Wilson, T. D. & Järvelin, K. (2003). On conceptual models for information seeking and
retrieval research. Information Research, 9(1). http://informationr.net/ir/9-1/paper163.html
Zach, L. (2005). When is “enough” enough? Modeling the information-seeking and stopping
behavior of senior arts administrators: Research Articles. Journal of the American Society for
Information Science and Technology, 56(1), 23-36.