Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Abraham

•Download as DOC, PDF•
•
A
anesah

This doctoral research explores how users make sense of and use information when seeking information from web-based resources. The research examines user behaviors and strategies as they interact with information sources to identify examples of how users collect, evaluate, understand, interpret, and integrate new information. Three empirical studies were conducted involving participants completing information seeking tasks. The results provide insights into users' information interaction strategies and sensemaking activities. Implications for the design of technologies to better support sensemaking are discussed.

1 of 4
Information Seeking from Web-based Resources: Sensemaking in Web-based environments: Implications for Interaction Design Ann Abraham a.m.abraham@open.ac.uk Supervisors Professor Marian Petre Professor Helen Sharp Department/Institute Computing Status Full Time /Part-time transition during write-up Probation viva After Starting date April 2005 End-users are more and more turning to the World Wide Web (WWW) for information to help them complete work, leisure or formal learning related everyday tasks; indeed the majority of search tasks currently undertaken in web contexts are ‘informational’ (e.g., Kobayashi and Takeda, 2000, Morrison et al, 2001). Typically such searches result in large volumes of information, presented in a variety of different formats from many different web-sites; making sense of and using the found information is often a non-trivial task. This research explores ‘How users make sense of and use or represent their sensemaking whilst seeking information in web-based environments’ To this end, the research investigates user behaviours and strategies as they interact with information sources in web-based contexts; its aims are to identify examples evident when users collect, evaluate, understand, interpret, and integrate new information for their own specific problem/task needs. The relationship between their sensemaking and information seeking is examined. From the ensuing understanding of sensemaking, implications for interaction design are inferred. The project is informed from information behaviour and sensemaking research: Information behaviour research is mature; hundreds of empirical studies are reported in the literature and many of the seminal behaviour models were informed from earlier studies of specialist user groups deploying customised or bibliographic database search engines. As a direct response to the impact of the World Wide Web (WWW) and the common use of general search engines, a few of these models have been re-examined in recent years to determine to what extent they continue to represent user’s information behaviour; for example, Choo et al (2000) re-evaluated Ellis’ earlier model(s) (1989, 1993) and Vakkari (2001) re-examined Kuhlthau’s earlier Information seeking Process model (ISP). Despite over 50 years of research and this recent resurgence of interest, the use aspect of information behaviour has been largely overlooked or at best notionally mentioned in the models, e.g. Wilson (1999) and Vakkari (2001), explicitly identify a use component but do not explore it. This ongoing lack of attention to the use of the located information has been criticized by many, e.g., Wilson, 1999; Kuhlthau, 2005, (pg 5). Information use is fundamental to sensemaking activities. This project is
particularly interested in the information behaviour which encompasses the purposive evaluation and ‘use’ made of located information sources: information searching is reserved for discussions which focus on interactions with the search query interface. There is a smaller body of reported empirical sensemaking research representing at least four different perspectives; one of these, i.e., Russell et al’ empirical work is particularly pertinent to this research. Their early, well cited model (1993) was derived from studies of professional workers using specialist equipment, but like other models it is a high-level description; as yet, little is known about the micro-level sensemaking. However, as with information behaviour, there is a resurgent interest in sensemaking, particularly micro-level everyday sensemaking in web-based contexts. Further, although a relationship between information seeking and sensemaking has been acknowledged, e.g. Dervin (1983) and Kuhlthau (1991), there are renewed calls to investigate and better understand this relationship (e.g. Qu and Furnas 2005). Design and methods In order to answer the broad research question (stated above), five sub questions were generated and three main empirical studies were designed to specifically address one or more of these sub-questions: • the first study was designed to identify a range of typical broad information strategies and associated sensemaking activities deployed by users during a research related information seeking task; nine PhD students were observed and interviewed as they undertook their own research related activities; • the second study was set-up to examine, in more detail, particular strategies emergent from study 1; eight experienced information processors ‘talked aloud’ as they were observed undertaking a time-constrained web-based topic comprehension task; • the third study was designed to be compared with and probe the results from the second study; eight experienced information processors ‘talked aloud’ as they were observed undertaking a time-constrained purchase-based task in a web context. A range of data collection tools were deployed across the three studies; these included − questionnaires applied pre and/or post observations; − think aloud protocol with audio or audio+video recording − evidence gathered post observations Throughout the studies, analysis was qualitative and inductive, providing opportunities to iteratively examine and explore different facets of the emergent results. Results The significant outcome from study 1 was an emergent empirical framework which highlighted four broad categories of user information-interaction strategies. These results corroborated and extended some of the published information behaviour literature, for example, by distinguishing (notionally) different phases within user’s evaluation of information sources, and offering some further insight into the little reported usage strategies. Examples of implicit and explicit sensemaking activities during the task cycle were identified for further scrutiny. These broad indicative results provided a baseline for the second study.
Study 2 produced several different visual representations as outcomes of the inductive analysis. For example individual participant timelines which highlight the multiples of resources in-play (visible and in use) and other characteristics of their resource interactions throughout the comprehension task. Along with the empirical framework from study 1, these timelines were used as a baseline for further investigation into participant interaction behaviours. Participants deployed a combination of five different interaction behaviours to solve the topic comprehension task; four of these broadly reflected the behaviour classes as identified in the empirical framework (study-1). The fifth, i.e. plan, was an emergent behaviour which helped relate the other behaviours to the task-in-hand. Typically, participant’s behaviours were complex and varied; frequently two or more resources were in-play as they switched between resources, extracting and compiling information. Each user session is captured as a unique sequence of conceptual episodes of behaviour; these algebraically expressed sequences indicate how participants collected and made sense of located information for the task-in-hand. Instances of broad sensemaking strategies set in the context of the empirical framework (study 1) are inferred from these individual ‘episode sequences’. Whilst each instance of broad sensemaking strategy’ was unique, some patterns and commonalities were evident. For example, − Each participant’s problem solving style can be classified with reference to their underlying planning strategies; top down, bottom-up or a mixed approach; − Planning is a dominant activity across most participants; − Seek is integral to sensemaking; − Almost all adopt a sequential approach to source processing; − There was a strong preference for pen&paper used as an intermediate workspace for representation and organisation purposes; − etc.. Each instance of ‘broad sensemaking strategy’ visually indicates the relationship between sensemaking and information seeking; this is to be explored. A collection of instances of broad sensemaking strategies is offered as one model of sensemaking. Instances of this model can be compared from two or more comparative studies. These results are of interest to information behaviour generally and in particular information interaction and exploratory search communities, and sensemaking communities. Indicative Implications for interaction design of sensemaking technologies Current everyday systems do not fully support sensemaking tasks; for example navigation errors & representation failures were evident, there was a notable preference for pen & paper for representation (storage & organisation), resources were often deployed in alternative roles, i.e., not default roles. Whilst it is clear that each interaction behaviour strategy has own design needs, there appears to be three groups of implications, namely • Improved Navigation support needed to address the high level of switching and navigation errors; interestingly apart from one case, no-one made use of multiple windows
• More robust solutions for representation are needed to support the range of representations emergent from usage, planning activities. An intermediate workspace reflecting the properties of pen&paper is needed. Further, an integrated solution would help reduce/address the switching errors from the iterative switching between source and target representation; • Management support would enhance the planning activities which occur regularly and throughout the session. These implications are of interest to designers of sensemaking support technologies. Selected References: 1. Choo, C. W., Detlor, B., & Turnbull, D. (1998). A Behavioral Model of Information Seeking on the Web -- Preliminary Results of a Study of How Managers and IT Specialists use the Web. In Proceedings of the 61st Annual Meeting of the American Society for Information Science. Pittsburgh, PA: edited by Cecilia Preston, 290-302. Medford, NJ: Information Today, Inc. 2. Choo, C.W.(1999) Closing The Cognitive Gaps: How People Process Information. Financial Times, March 1999, Reprinted August, 2001, in National Post of Canada. http://128.100.159.139/FIS/respub/FThis/default.html [last accessed 08/2006]. 3. Choo, C. W., Detlor, B., & Turnbull, D. (2000). Working the Web: An Empirical Model of Web Use. In Proceedings of HICSS 33 (Hawaii International Conference on Systems Science). Hawaii. 4. Dervin, B. (1983 & 2002). An overview of sense-making research: concepts, methods and results. In: International Communications Association Annual Meeting. Dallas, Texas. 5. Kuhlthau, C. (1991). Inside the Search Process: Information Seeking from the User’s Perspective. Journal of American Society for Information Science 42(5), 361-371. 6. Kuhlthau, C. C. (2005). "Towards collaboration between information seeking and information retrieval" Information Research, 10(2) paper 226 [Available at http://InformationR.net/ir/10-2/paper226.html] 7. Kobayash, M., & Takeda, K. (2001). Information Retrieval on the Web. ACM Computing Surveys, 32(2). 8. Morrison, J. B., Pirolli, P., & Card, S. K. (2001). A Taxonomic Analysis of What World Wide Web Activities Significantly Impact People's Decisions and Actions. In Interactive Poster in CHI 2001 conference proceedings, pg 163-4: ACM. 9. Russell, D. M., Stefik, M. J., Pirolli, P., and Card, S. K. (1993). The Cost Structure of Sensemaking. In Proc.of ACM INTERCHI'93, 269-276. 10. Qu, Y. and Furnas, G. (2005). Sources of Structure in Sensemaking. In Proc CHI 2005, ACM Press, 1989-1992 11. Vakkari, P. (2001). Changes in Search Tactics and Relevance Judgements when Preparing a Research Proposal: A Summary of the Findings of a Longitudinal Study. Information Retrieval 4 (3/4), 295-310. 12. Wilson, T. D. (1999). Models of Information Behaviour research. Journal of Documentation, 55(3), 249-270.

More Related Content

Abraham

  • 1. Information Seeking from Web-based Resources: Sensemaking in Web-based environments: Implications for Interaction Design Ann Abraham a.m.abraham@open.ac.uk Supervisors Professor Marian Petre Professor Helen Sharp Department/Institute Computing Status Full Time /Part-time transition during write-up Probation viva After Starting date April 2005 End-users are more and more turning to the World Wide Web (WWW) for information to help them complete work, leisure or formal learning related everyday tasks; indeed the majority of search tasks currently undertaken in web contexts are ‘informational’ (e.g., Kobayashi and Takeda, 2000, Morrison et al, 2001). Typically such searches result in large volumes of information, presented in a variety of different formats from many different web-sites; making sense of and using the found information is often a non-trivial task. This research explores ‘How users make sense of and use or represent their sensemaking whilst seeking information in web-based environments’ To this end, the research investigates user behaviours and strategies as they interact with information sources in web-based contexts; its aims are to identify examples evident when users collect, evaluate, understand, interpret, and integrate new information for their own specific problem/task needs. The relationship between their sensemaking and information seeking is examined. From the ensuing understanding of sensemaking, implications for interaction design are inferred. The project is informed from information behaviour and sensemaking research: Information behaviour research is mature; hundreds of empirical studies are reported in the literature and many of the seminal behaviour models were informed from earlier studies of specialist user groups deploying customised or bibliographic database search engines. As a direct response to the impact of the World Wide Web (WWW) and the common use of general search engines, a few of these models have been re-examined in recent years to determine to what extent they continue to represent user’s information behaviour; for example, Choo et al (2000) re-evaluated Ellis’ earlier model(s) (1989, 1993) and Vakkari (2001) re-examined Kuhlthau’s earlier Information seeking Process model (ISP). Despite over 50 years of research and this recent resurgence of interest, the use aspect of information behaviour has been largely overlooked or at best notionally mentioned in the models, e.g. Wilson (1999) and Vakkari (2001), explicitly identify a use component but do not explore it. This ongoing lack of attention to the use of the located information has been criticized by many, e.g., Wilson, 1999; Kuhlthau, 2005, (pg 5). Information use is fundamental to sensemaking activities. This project is
  • 2. particularly interested in the information behaviour which encompasses the purposive evaluation and ‘use’ made of located information sources: information searching is reserved for discussions which focus on interactions with the search query interface. There is a smaller body of reported empirical sensemaking research representing at least four different perspectives; one of these, i.e., Russell et al’ empirical work is particularly pertinent to this research. Their early, well cited model (1993) was derived from studies of professional workers using specialist equipment, but like other models it is a high-level description; as yet, little is known about the micro-level sensemaking. However, as with information behaviour, there is a resurgent interest in sensemaking, particularly micro-level everyday sensemaking in web-based contexts. Further, although a relationship between information seeking and sensemaking has been acknowledged, e.g. Dervin (1983) and Kuhlthau (1991), there are renewed calls to investigate and better understand this relationship (e.g. Qu and Furnas 2005). Design and methods In order to answer the broad research question (stated above), five sub questions were generated and three main empirical studies were designed to specifically address one or more of these sub-questions: • the first study was designed to identify a range of typical broad information strategies and associated sensemaking activities deployed by users during a research related information seeking task; nine PhD students were observed and interviewed as they undertook their own research related activities; • the second study was set-up to examine, in more detail, particular strategies emergent from study 1; eight experienced information processors ‘talked aloud’ as they were observed undertaking a time-constrained web-based topic comprehension task; • the third study was designed to be compared with and probe the results from the second study; eight experienced information processors ‘talked aloud’ as they were observed undertaking a time-constrained purchase-based task in a web context. A range of data collection tools were deployed across the three studies; these included − questionnaires applied pre and/or post observations; − think aloud protocol with audio or audio+video recording − evidence gathered post observations Throughout the studies, analysis was qualitative and inductive, providing opportunities to iteratively examine and explore different facets of the emergent results. Results The significant outcome from study 1 was an emergent empirical framework which highlighted four broad categories of user information-interaction strategies. These results corroborated and extended some of the published information behaviour literature, for example, by distinguishing (notionally) different phases within user’s evaluation of information sources, and offering some further insight into the little reported usage strategies. Examples of implicit and explicit sensemaking activities during the task cycle were identified for further scrutiny. These broad indicative results provided a baseline for the second study.
  • 3. Study 2 produced several different visual representations as outcomes of the inductive analysis. For example individual participant timelines which highlight the multiples of resources in-play (visible and in use) and other characteristics of their resource interactions throughout the comprehension task. Along with the empirical framework from study 1, these timelines were used as a baseline for further investigation into participant interaction behaviours. Participants deployed a combination of five different interaction behaviours to solve the topic comprehension task; four of these broadly reflected the behaviour classes as identified in the empirical framework (study-1). The fifth, i.e. plan, was an emergent behaviour which helped relate the other behaviours to the task-in-hand. Typically, participant’s behaviours were complex and varied; frequently two or more resources were in-play as they switched between resources, extracting and compiling information. Each user session is captured as a unique sequence of conceptual episodes of behaviour; these algebraically expressed sequences indicate how participants collected and made sense of located information for the task-in-hand. Instances of broad sensemaking strategies set in the context of the empirical framework (study 1) are inferred from these individual ‘episode sequences’. Whilst each instance of broad sensemaking strategy’ was unique, some patterns and commonalities were evident. For example, − Each participant’s problem solving style can be classified with reference to their underlying planning strategies; top down, bottom-up or a mixed approach; − Planning is a dominant activity across most participants; − Seek is integral to sensemaking; − Almost all adopt a sequential approach to source processing; − There was a strong preference for pen&paper used as an intermediate workspace for representation and organisation purposes; − etc.. Each instance of ‘broad sensemaking strategy’ visually indicates the relationship between sensemaking and information seeking; this is to be explored. A collection of instances of broad sensemaking strategies is offered as one model of sensemaking. Instances of this model can be compared from two or more comparative studies. These results are of interest to information behaviour generally and in particular information interaction and exploratory search communities, and sensemaking communities. Indicative Implications for interaction design of sensemaking technologies Current everyday systems do not fully support sensemaking tasks; for example navigation errors & representation failures were evident, there was a notable preference for pen & paper for representation (storage & organisation), resources were often deployed in alternative roles, i.e., not default roles. Whilst it is clear that each interaction behaviour strategy has own design needs, there appears to be three groups of implications, namely • Improved Navigation support needed to address the high level of switching and navigation errors; interestingly apart from one case, no-one made use of multiple windows
  • 4. • More robust solutions for representation are needed to support the range of representations emergent from usage, planning activities. An intermediate workspace reflecting the properties of pen&paper is needed. Further, an integrated solution would help reduce/address the switching errors from the iterative switching between source and target representation; • Management support would enhance the planning activities which occur regularly and throughout the session. These implications are of interest to designers of sensemaking support technologies. Selected References: 1. Choo, C. W., Detlor, B., & Turnbull, D. (1998). A Behavioral Model of Information Seeking on the Web -- Preliminary Results of a Study of How Managers and IT Specialists use the Web. In Proceedings of the 61st Annual Meeting of the American Society for Information Science. Pittsburgh, PA: edited by Cecilia Preston, 290-302. Medford, NJ: Information Today, Inc. 2. Choo, C.W.(1999) Closing The Cognitive Gaps: How People Process Information. Financial Times, March 1999, Reprinted August, 2001, in National Post of Canada. http://128.100.159.139/FIS/respub/FThis/default.html [last accessed 08/2006]. 3. Choo, C. W., Detlor, B., & Turnbull, D. (2000). Working the Web: An Empirical Model of Web Use. In Proceedings of HICSS 33 (Hawaii International Conference on Systems Science). Hawaii. 4. Dervin, B. (1983 & 2002). An overview of sense-making research: concepts, methods and results. In: International Communications Association Annual Meeting. Dallas, Texas. 5. Kuhlthau, C. (1991). Inside the Search Process: Information Seeking from the User’s Perspective. Journal of American Society for Information Science 42(5), 361-371. 6. Kuhlthau, C. C. (2005). "Towards collaboration between information seeking and information retrieval" Information Research, 10(2) paper 226 [Available at http://InformationR.net/ir/10-2/paper226.html] 7. Kobayash, M., & Takeda, K. (2001). Information Retrieval on the Web. ACM Computing Surveys, 32(2). 8. Morrison, J. B., Pirolli, P., & Card, S. K. (2001). A Taxonomic Analysis of What World Wide Web Activities Significantly Impact People's Decisions and Actions. In Interactive Poster in CHI 2001 conference proceedings, pg 163-4: ACM. 9. Russell, D. M., Stefik, M. J., Pirolli, P., and Card, S. K. (1993). The Cost Structure of Sensemaking. In Proc.of ACM INTERCHI'93, 269-276. 10. Qu, Y. and Furnas, G. (2005). Sources of Structure in Sensemaking. In Proc CHI 2005, ACM Press, 1989-1992 11. Vakkari, P. (2001). Changes in Search Tactics and Relevance Judgements when Preparing a Research Proposal: A Summary of the Findings of a Longitudinal Study. Information Retrieval 4 (3/4), 295-310. 12. Wilson, T. D. (1999). Models of Information Behaviour research. Journal of Documentation, 55(3), 249-270.