A model of uncertainty and its relation to information seeking and retrieval (IS&R)
| Date | 08 July 2014 |
| Pages | 575-604 |
| DOI | https://doi.org/10.1108/JD-05-2013-0060 |
| Published date | 08 July 2014 |
| Author | Sudatta Chowdhury,Forbes Gibb,Monica Landoni |
| Subject Matter | Library & information science,Records management & preservation,Document management |
A model of uncertainty and its
relation to information seeking
and retrieval (IS&R)
Sudatta Chowdhury and Forbes Gibb
Department of Computer and Information Sciences, University of Strathclyde,
Glasgow, UK, and
Monica Landoni
Faculty of Informatics, University of Lugano, Lugano, Switzerland
Abstract
Purpose – The purposeof this paper is to show thatuncertainty may becaused not only by a knowledge
gap in the mind of auser with respect to a given subject or topic, but also by the various complexities
associated with the information seeking an d retrieval (IS&R) process in a digital environment.
Design/methodology/approach – Both quantitative and qualitative studies were conducted to
collect data from users in the higher education sector regarding whether or not they experienced
uncertainty in relation to the IS&R process. Analysis: a correlation analysis was undertaken to
establish whether there were any relationships betwe en information-seeking activities an d
information-seeking problems.
Findings – The findings of this research show that uncertainty existed at different stages of the IS&R
process amongst users. It was establishedthat uncertainty was caused by a number of information-seeking
activities and information-seeking problems, and that such uncertainty could continue over the course
of successive search sessions, leading to the proposal of a new model of uncertainty.
Research limitations/implications – The proposed model of uncertainty should contribute to a
better understanding of the issues related to IS&R in a digital environment.
Practical implications – A number of benefits could be realised in systems design from the
application of this model in terms of reducing the negative impact of uncertainty, while at the same
time helping users to gain from the positive aspects of uncertainty in IS&R.
Originality/value – The general consensus is that uncertainty is a mental state of users reflecting a
gap in knowledge which triggers an IS&R process, and that the gap is reduced as relevant information
is found, and thus that the uncertainty disappears as the search process concludes. However, in the
present study it is argued that some form of uncertainty is always associated with some part of
the IS&R process and that it also fluctuates throughout the IS&R process. Users may therefore feel
uncertain at any stage of the IS&R process and this may be related to: the initial information need
and expression of that need, the search process itself, including identification of relevant systems,
services and resources; and the assessment of, and reaction to, the results produced by the search
process. Uncertainty may be unresolved, or even increase, as the user progresses, often iteratively,
through the IS&R process and may remain even after its completion, resulting in what may be called
a persistent uncertainty. In other words, this research hypothesises that, in addition to the uncertainty
that triggers the information search process (Wilson et al., 2000), users suffer from varying deg rees of
uncertainty at every stage of the information search and retrieval process, and that in turn, triggers
different information-seeking behaviours.
Keywords Uncertainty, Information seeking and retrieval, Information seeking model, IS&R,
Uncertainty model
Paper type Research p aper
1. Introduction
Users look for information from a vast range of sources and it is often qui te difficultfor
them to determine which of the various channels could provide infor mation in the most
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0022-0418.htm
Received 8 May 2013
Accepted 24 May 2013
Journal of Documentation
Vol.70 No. 4, 2014
pp. 575-604
rEmeraldGroup Publishing Limited
0022-0418
DOI 10.1108/J D-05-2013-0060
575
A model of
uncertainty and
its relation
to IS&R
effective and efficient manner (Chowdhury and Landoni, 2006). In addition, despite
a significant amount of research into the development of sophisticated search tools,
the identification of appropriate info rmation sources remains a very challenging
task (Savolainen, 2007; Fritz and Schiefer, 2003). It is also widely recognised that there
are many complexities involved in informatio n seeking and retrieval (IS&R) (Dillon
and Hahn, 2002; EPIC, 2001; Tenopir et al., 2003; Blake and Pratt, 2002; Rieh, 2002;
Prabha et al., 2007), and all of these factors can lead to a feeling of uncertainty in users.
Twokey questions therefo reare: what sp ecificallycauses unce rtainty; and what arethe
implications of such uncertainty in a digital environment?
The concept of uncertainty is understood and applied in different ways in different
fields such as philosophy, statistics, economics, finance, insurance, psychology,
engineering, and science. Within the context of infor mation science it may relate to
a number of different aspects of a user’s engagement with an information system. For
instance, in the model of Shannon and Weaver (1949) uncertainty is inversely related
to the amount of information received in a communication system. Belkin (1980), on the
other hand, argues that that the information-seeking process begins with a level of
uncertainty which is caused by a gap in users’ knowledge and that this impacts on
their ability to formulate effective queries to retrieve the very information that will
reduce their uncertainty. The general consensus is that this uncertainty g radually
decreases, sometimes iteratively through positive feedbac k, as the user discovers
information that addresses this gap during the IS&R process (Kuhlthau, 1993; Wilson
et al., 2000, 2002). It is also suggested that uncertainty, like risk, is not an entirely
negative influence (Anderson, 2006): without taking risks progress can be limited, and
without embracing uncertainty users may not be led down unexpected but useful
information pathways.
However, the basis for this research is the hypothesis that it is not only the gap in
knowledge that causes uncertainty in a user, but also the complexities associated
with the IS&R process itself, and that uncertainty may not always disappear at the end
of an information search. It is argued here that uncertainty may be caused in particular
by: the growth of diverse electronic information channels and sources and the
associated mechanisms to access them; lack of awareness of, and unfamiliarity with,
the appropriate channels and sources; and questions related to the quality of the
channels and sources. From this perspective uncertainty can be categorised as being
related to topicality and information needs, the IS&R process itself, and satisfaction
with the results of the IS&R process. The context of the user is also a factor, as
discussed by Bystro
¨m and Ja
¨rvelin (1995) and will influence both the range and
complexity of tasks which a user must undertake. Moreover, information needs,
and the pressure to satisfy them, may create psychological uncertainty in users.
For instance, workers (i.e. those engaged in economic activity) will be influenced
by an awareness of performance monitoring by supervisors which will, in tur n, be
influenced by the effectiveness of task completion and the ability to search for
accurate information and make appropriate decisions. Academic researchers (who
were the focus of the research reported here) will be influenced by peer pressure
and acceptance in their subject field, which will in turn be influence d by their abilityto
demonstrate both their awareness of existing contributions to their field and the
novelty of their work. Ironically, this requires them to satisfy concurrently an
information need based around the presence and absence of research results thus
creating a particular type of uncertainty. The relations between these factors are
summarised in Figure 1.
576
JDOC
70,4
To continue reading
Request your trial