[Library and Information Science] New Directions in Information Behaviour Volume 1 || Chapter 5 Into the Land of Adolescent Metacognitive Knowledge During the Information Search Process:

Chapter 5

Into the Land of Adolescent Metacognitive

Knowledge During the Information Search

Process: A Metacognitive Ethnography

Leanne Bowler

Abstract

This chapter examines adolescent metacognitive knowledge in a freshlight and answers some methodological questions related to theinvestigation of the deepest layers of thinking during the informationsearch. It does so by presenting a study that used an ethnographicapproach to investigate the metacognitive knowledge of 10 adoles-cents, aged 16 to 18, over the course of four months, and in a variety ofsettings –– home, school, public libraries –– as they searched for,collected, and then used information for a school project. The studywas framed by Flavel’s model of metacognition (1977) and Kuhlthau’sinformation search process (ISP) model, a six-stage, multidimensionalmodel of information problem solving (1991, 2004). The chapterbegins with a discussion about the ISP, metacognitive knowledge, andits potential for information seeking. The chapter then presents thefindings of the study as a set of gaps and strengths of adolescentmetacognitive knowledge, and concludes with commentary about thechallenges and rewards related to conducting research with youngpeople and suggestions for future areas of research.

New Directions in Information Behaviour

Library and Information Science, 93–125

Copyright r 2011 by Emerald Group Publishing Limited

All rights of reproduction in any form reserved

ISSN: 1876-0562/doi:10.1108/S1876-0562(2011)002011a008

94 Leanne Bowler

5.1. Introduction

Digital Natives. Net Generation. iGeneration. Generation Media. Despiteall the nomenclature to suggest that today’s young people are wizards atinformation seeking, evidence suggests just the opposite. Adolescents, forthe most part, are only adequate information seekers and in fact could usesome guidance in terms of best practices (Agosto, 2002; Branch, 2003;CIBER, 2008; Chen, 2003; Chung & Neuman, 2007; Dresang, 2005;Fallows, 2005; Fidel et al., 1999; Neilsen, 2005; Todd, 2003; Valenza, 2006;Watson, 1998; Williamson, McGregor, Archibald, & Sullivan, 2007).

But where do we begin? Adolescents, still adults-in-training, have manybarriers to conquer. Lacking life experience and a deep level of domainknowledge, what skills and knowledge can they use to guide them throughthe information search process (ISP)? The problem is not so simple. First,information problems are, by their very nature, idiosyncratic, contextual,and meaningful in very personal ways, so it is difficult to provide one-size-fits-all solution. Second, the information environment is vast and growingexponentially every year. And finally, the tools to enable informationseeking, discovery, and gathering are in a constant state of flux, leavinginformation seekers in a perpetual state of novice-hood when it comes tolearning how to navigate new information technology. In this environment,information seekers need skills that are adaptable and transferable, aninternal guidance system as it were, rather than a set of discrete skills.

Metacognition –– a form of self-aware thinking that can help one monitorone’s own thinking –– has been held up as one solution, the potential magicbullet of information seeking. This is reflected in the American Association ofSchool Librarian’s standards for learners in the 21st century which highlightself-assessment strategies such as monitoring information-seeking processesand seeking feedback from teachers and peers to guide one’s own inquiryprocess (AASL, 2007). The problem with teaching young people aboutmetacognition is that we have relatively few studies that actually tell us what itlooks like in the context of adolescent information seeking. How can we planfor learning outcomes without a baseline understanding of the landscape ofthe phenomenon?

This chapter describes a study that used ethnographic methods to exploreadolescent metacognitive thinking during the ISP. The study was framed byFlavel’s model of metacognition (1977) and Kuhlthau’s ISP model, a six-stage, multidimensional model of information problem solving (1991, 2004).The use of Kuhlthau’s ISP model implies that the information behaviorunder investigation here goes beyond basic information retrieval (IR) andincludes the tasks related to initiation, topic selection, prefocus exploration,focus formulation, information collection, and presentation. The chapterbegins with a discussion about the ISP, metacognitive knowledge, and its

A Metacognitive Ethnography 95

potential for information seeking. The chapter then presents the findings asa set of gaps and strengths of adolescent metacognitive knowledge andconcludes with commentary about the challenges and rewards related toconducting research with young people and suggestions for future areas ofresearch.

5.2. The Information Search Process

This study began with an assumption that information seeking is a problem-solving process, a set of decisions and actions that culminate in a solution.The notion of process implies sequence –– a ‘‘moving picture’’ that takesplace over a period of time, rather than a ‘‘still photograph’’ of one searchincident (Dervin & Nilan, 1986, p. 14). To capture the true image of thephenomenon we call information seeking, a holistic and long-term approachis helpful. Various models of the problem-solving processes in informationseeking exist. This study is framed by one process model in particular ––Kuhlthau’s ISP (1991, 2004). Despite its name, the ISP model falls under thecategory of ‘‘information seeking’’ and is not particularly concerned with IRper se. Rather, the ISP model takes a long view of information seeking and isnot restricted to the study of IR systems.

The ISP model identifies six stages, each representing specific informa-tion-seeking tasks: task initiation, topic selection, prefocus exploration, focusformulation, information collection, and search closure (presentation) (1991).A seventh stage –– reflection and self-assessment –– was mentioned inKuhlthau’s book –– Teaching the Library Research Process: A Step-By-StepProgram for Secondary School Students (1985). Most writing and researchon the ISP model, however, does not acknowledge this stage.

In the ISP model, cognition, affect, and behavior are intertwined.Kuhlthau found a pattern of feelings that paralleled the specific stages ofknowledge integration during the search process. As information seekersmove through the process, their feelings reflect their understanding of theirresearch topic. High anxiety is associated with cognitive uncertainty and isrelated to difficulty integrating information from various sources into ameaningful whole. A turning point in the process comes when informationseekers are able to find a focus for their information-seeking mission andcognitive certainty turns to feelings of confidence. The central lesson forboth information seekers and information providers is that formulating afocus is a prerequisite for moving forward in the search process.

The ISP model was selected as a framework for this study for severalreasons. To begin with, the ISP model is relevant to the adolescent populationinvestigated in this study, the ISP model having emerged from a study with

96 Leanne Bowler

high school students. Second, the model offered some reliability, having beenvalidated in a series of five studies over a span of six years, using both smalland large scale sample sizes, in a diverse range of settings, within a variety ofpopulations (Kuhlthau, 1985, 1988, 1989; Kuhlthau, Turock, George, &Belvin, 1990). Subsequent research by others supports the behavior predictedby Kuhlthau’s model, although some modifications have been made to it(Byron & Young, 2001; Friel, 1995; Kracker, 2002; Kracker & Wang, 2002;Shamo, 2001; Swain, 1996). Finally, the ISP model provides a multi-dimensional view of the information-seeking process. Modeling informationseeking from the perspective of cognition (thinking), behavior (tasks), and,importantly for this study, affect (feelings). Metacognition is known to havelinks to motivation and affect and so the ISP model helped to set the stage forthe investigation of affect in relation to metacognitive knowledge.

5.3. Metacognitive Knowledge

Metacognitive knowledge is a critical piece of the information literacypuzzle. In a world of exploding information and communicationspossibilities, the difficulty for users of information systems and servicesmay not lie in finding information but in filtering and integrating it into acohesive whole. To do this, they must be able to make sense of it, an act thatassumes knowledge about one’s own information needs, goals, and abilities.This type of self-knowledge –– called metacognitive knowledge –– has threebasic components: knowledge of one’s self, knowledge of the nature of acognitive task in relation to one’s own cognitive abilities, and knowledge ofhow and when to effectively use cognitive strategies to complete a cognitivetask (Anderson & Krathwohl, 2001; Flavell, 1979). Such knowledge, whenused in information seeking, may help users to solve complex informationproblems. Land and Greene, in their study on project-based learning withthe Web, found that metacognitive knowledge compensated for a lack ofsystem and domain knowledge, suggesting that metacognitive knowledgecan act as a scaffold in knowledge integration from Web sources (2000).Interactive IR systems that model metacognition for users may be oneapproach for providing support in the development of metacognitiveknowledge (Beheshti, Bowler, Large, & Nesset, 2005; Gorrell, Eaglestone,Ford, Holdridge, & Madden, 2009).

There is perhaps no other user group who could benefit more from thedevelopment of metacognitive knowledge than adolescents, aged 16 to 19.On the cusp of adulthood, they face many of the complex informationproblems of adults, but as ‘‘novice adults,’’ their depth of knowledge onmost topics may be shallow simply because they have only experienced life


for a handful of years. Metacognitive knowledge may be a particularlyuseful support for young people who are unraveling their first complexinformation problems. There is a wide body of research that has looked atinformation seeking in relation to critical thinking, higher order thinking,mental models, reflection, self-monitoring, and control processes –– allconstructs related to but not synonymous with metacognitive knowledge.Bertland (1986) introduced the concept of metacognition to the schoollibrary audience in her review of research that could have implications forinformation skills instruction. At that point in time, research in this area wasnew and principally connected to text comprehension, attention, andmemory rather than information seeking per se. Interestingly, Bertlandnoted that very little was known about the metacognitive processes ofadolescents, most of the research having focused on children under the ageof 12 or on college-level students.

Fitzgerald looked at K-12 students’ ability to evaluate information andargued that one area of information literacy program development could bein the area of metacognitive awareness (1999). More recently, Wolf hasdrawn a theoretical and operational link between two areas of inquiry ––information literacy and self-regulation (an aspect of metacognition) ––concluding that it ‘‘might prove useful to create and implement assessmentsor inventories that can measure the self-regulatory or information literacydevelopmental stage of students’’ (2007).

Regardless of the connection between information literacy and metacog-nition, a relatively small set of studies has explicitly investigated themetacognition of young people during the ISP and for the most part, thosethat exist were conducted before young people had broad access to the Web.Kuhlthau, in her study into the ISP, touched on the importance ofmetacognition when she argued that one of the tasks of librarians is to raisein information seekers a self-awareness about their own cognitive state ––specifically, their states of uncertainty –– and to teach this alongside thetraditional skills of locating and evaluating information (1991, 2004).

In their study of grade-six children in a New Zealand school, Mooreand St. George found that although all the students were guided bymetacognitive knowledge during the research process, the nature of thisknowledge was incomplete and flawed (Moore, 1995; Moore & St.George, 1991). McGregor explored the higher order thinking skills ofgrade-eleven students in a gifted program in a Canadian high school andfound that the students’ thinking was carried out at an intuitive level,without awareness of their own thinking or the ways they could modifytheir thinking to advance their learning (1994a, 1994b). Wolf, Brush, andSaye (2003) tested the Big6 instructional model grade-eight students tosee if it could be used to scaffold specific metacognitive strategies in aidof information problem solving.

98 Leanne Bowler

5.4. Methods

The study used a naturalistic, qualitative approach –– identified in thischapter as metacognitive ethnography –– to uncover the metacognitiveknowledge of 10 adolescents, ages 16 to 18, from a Canadian school as theysearched for, selected, and used information for a school-based inquiryproject on a topic related to the history of western civilization (Bowler,2010a, 2010b). The primary question guiding this research was What is therole of metacognitive knowledge during the ISP of adolescents? Twosecondary questions framed the study and provided conceptual categoriesfor data analysis:

1. Within the context of the search process, what are the qualities ofadolescent metacognitive knowledge?

2. How does the metacognitive knowledge of adolescents map on to thecognitive, affective, and behavioral dimensions of the search process?

For all the reasons that one-size-fits-all approaches to information skillstraining may not be helpful (e.g., the deeply personal and contextual reasonsfor information seeking which may in turn determine deeply personal andcontextual approaches to seeking information), it is equally difficult to use aone-size-fits-all research approach (like a survey or questionnaire) touncover the landscape of the thinking that may underlie the cognition ofyoung people during the ISP. For this reason, an ethnographic approachwas used in this study to address the particularly challenging problem ofrevealing, in a naturalistic, authentic way, the deep layers of adolescentthinking during the ISP knowledge.

5.4.1. Pilot Study

A pilot study was conducted six months in advance of the larger study. Thepilot study provided an authentic platform for testing the protocols andpracticing their implementation with an adolescent student. It also helped tohighlight some issues related to navigating throughout the school system.Although seven students initially expressed interest in participating in thepilot study, only one student, aged 17, stayed with the pilot study frombeginning to end, suggesting a potential retention problem. The pilot studywas not held in the same location as the principal study so as not to taint thedata. Ethical approval to run the pilot study was received from the schooland written consent was received from both the student and her parents.Access to the student was gained through a personal connection with ateacher, and while the teacher was able to point to interested students, the


actual assignment in her class did not have a research component. Thestudent who participated in the study had to identify another course whichincluded a research component. The research project was small, extendingover less than a month, with the outcome being a class presentation ––nothing comparable to the research project followed by the students in themain study (see below). As a result, the data derived from the pilot study hasnot been included in the analysis. Nevertheless, the pilot project did teachthe researcher some important lessons.

There were three principal outcomes from the pilot study. First,problems with the question prompts for the written or audio journal cameto light. They did not shine a big enough spotlight on thinking processesand seemed to lead the participant to focus on procedural matters.Furthermore, a prompt asking the participant to identify the date was notincluded, making it difficult to corroborate the timeline. As a result, thewording for the question prompts was refined and the final number ofquestion prompts increased from four to seven. Second, the researcher wasalerted to potential problems accessing and retaining participants in thestudy, suggesting that a generous approach should be adopted whenrecruiting participants and that every effort should be made toaccommodate their lifestyles. A final outcome of the pilot was that theresearcher was alerted to potential logistical problems (such as meetingrooms being locked or student schedules changing) and as a result,approached the principal study armed with possible solutions to problemsrelated to scheduling and location.

5.4.2. The Setting

The setting for the principal study was an English language, junior college(or more commonly called a CEGEP) in Montreal, Canada. The acronymCEGEP stands for ‘‘College d’enseignement general et professionnel’’ or‘‘College of General and Professional Education.’’ There are two programstreams in the CEGEP system –– a two-year preuniversity program and athree-year professional program. In either case, the first year is roughlyequivalent to Grade 12 elsewhere in Canada because high school in Quebecends at Grade 11. New CEGEP students have little prior experience ortraining in information skills as the position of teacher–librarian did notexist in Quebec public high schools nor in most private high schools at thetime that this study was conducted. These students may have had little elsebut their metacognitive knowledge to guide them through the newinformation environment they encounter at CEGEP.

100 Leanne Bowler

5.4.3. The Participants

Ten adolescents, aged 16 to 18, participated in the study –– eight femalesand two males. They came from a variety of linguistic backgrounds ––Montreal being a diverse city in terms of language. All of the participantsspoke both English and French (according to the participants). At least twohad a third language as their mother tongue (Hungarian and Bulgarian).They had all graduated from high school the previous year and all wereacademic achievers in their first term at CEGEP, having been accepted to achallenging ‘‘Arts and Sciences’’ program. All the participants were in thesame required humanities course at the same CEGEP.

As the study was inductive in nature, it was preferable that the samplesize be small in order to allow for in-depth exploration of variables.Nevertheless, the study was open to all students in the class wishing toparticipate and no constraints were set on the individual characteristics ofparticipants drawn from these classes. This was to alleviate concerns thatpotential benefits of the study would only accrue to a few (such as learninghow to keep a journal of search behavior). All volunteers, therefore, wereaccepted irrespective of their gender, language, or ethnicity or of the numberof volunteers. In the event, only 10 students participated and all 10 carriedthrough with the study, from beginning to end.

5.4.4. The Information-Seeking Task

The information-seeking task was created and assigned by the teacher. The10 participants were asked to write a seven- to eight-page argumentativeessay exploring continuity and change in western civilization, on a topic oftheir choice. To do so, they searched for, selected, evaluated, and usedinformation from a variety of sources over the course of the fall 2006 term.The search process was a critical element to the success of the assignmentbecause the students were to use information sources to defend theirposition. The class had an assigned textbook, but it was not to be included inthe list of sources for the research paper. The participants were first asked toidentify a topic and to locate it within a specific geographic location andtime frame. No specific guidelines or boundaries as regard the topic wereprovided, as long as it was related to the history of western civilization. As aresult, the 10 participants investigated a wide array of topics –– from Greekarchitecture to classical music.

The participants were told about the assignment during the first class(last week of August) and the assignment then unfolded in four stages overthe course of four months. By the second week of school, they were to


identify the topic. Two weeks later, they were to present a short annotatedbibliography of five sources. They were then asked to critically evaluateone Web site that would be helpful in their research, and finally, they wereasked to write a seven- to eight-page research paper, due just before theexam period at the end of the term. The bibliography for the final paperhad to have at least eight information sources, only three of which couldbe Web sites.

5.4.5. Data Collection Protocols

The study used a combination of think-aloud and think-after verbalprotocols in order to provide as many venues possible for the expressionof thoughts, feelings, and actions experienced by the participants during thesearch process. In this way, the data could be triangulated. Five types ofdata collection protocols were used in this study: (1) a series of threetelephone interviews; (2) written and/or audio journals kept by theparticipants over the course of the semester; (3) an in-person interviewimmediately following the final submission of the essay; (4) a visualizingexercise (a timeline); and (5) a follow-up interview conducted several monthslater. Metacognitive knowledge was specifically targeted by questionsrelated to why and probes designed to reveal the self-prompting questionsthe participants may have asked themselves.

5.4.5.1. Telephone Interviews Three telephone interviews were conductedwith each participant over a four-month period. The telephone interviewswere typically no more than 10minutes long. The researcher took notes byhand and transcribed the interviews into a text document immediately afterthe interview. Interviewing by telephone, rather than in person, waspreferable because it could be conducted in the evening and on weekends,and made the interview process easier for the participants. The telephoneinterviews caught the participants ‘‘live,’’ in the middle of their researchprojects, and were therefore as close as data collection could get to anauthentic ‘‘think-aloud’’ protocol without actually being in the participants’homes. A semi-structured interview schedule was followed during thetelephone interviews so as to be assured that specific aspects of the searchprocess were discussed in as naturalistic a way as possible.

5.4.5.2. Written or Audio Journals Each participant was asked to record ina written or audio-taped journal, in English or French, their thoughts,feelings, and actions whenever they searched for information for theirresearch paper. The option to choose either writing or talking into a tape

102 Leanne Bowler

recorder (or both) was offered in order to facilitate participation. Eachparticipant was provided with a kit that included a notebook, a smallhandheld tape recorder, and a card with question prompts. While thejournals were conceived as possible sources for ‘‘think-aloud’’ data –– thatis, information about an event that is recorded at the time it occurs –– inactual fact, there is no way to know at what point the journals werecompleted.

The journals, whether completed in the middle of the process or after thefact, did provide a rich source of data on the participants’ thoughts, feelings,and actions. Of the 10 participants, 7 chose to keep a written journal, 2 keptboth a written and an audio journal, and 1 used the audio journalexclusively. The length of journal depended on the participant. For example,one participant wrote lengthy essays almost nightly while another chose toprovide notes written in the course notebook which included thoughts aboutthe research topic. Overall, most wrote one to two pages per week. Whilemost participants found writing in the journal more comfortable thanspeaking into a tape recorder, the three participants who did use thehandheld recorder said it was more convenient when they were pressed fortime. The audio journals were transferred to a digital format and thentranscribed to a text document, using iTunes software as the audio output.

5.4.5.3. In-Person Interview At the end of the term, after the research paperhad been submitted to the teacher, the researcher met with each participantfor approximately 90minutes in order to conduct a semi-structuredinterview, followed by a visualizing exercise (details about the exercise areprovided below). The interview questions emerged from the telephoneinterviews and were framed by the six stages of the ISP model. This structurehelped to correlate the qualities of metacognitive knowledge with the specifictasks, phases, and dimensions of the ISP model. While the interviewfollowed a general structure, points of interest did emerge and the researcherwas not hesitant to ask questions that helped to reveal thinking processesand feelings experienced during the information-search process. Theinterview was audiotaped, later transferred to a digital format and thentranscribed to a text document, using iTunes software as the media player.

5.4.5.4. Visualizing Exercise (Timeline) The visualizing exercise (a timeline)occurred during the same meeting as the in-person interview. For thevisualizing exercise, participants were asked to draw a timeline and alongthis line describe four elements of their search process using differentcolored pens. The four elements were (1) their actions, (2) the reasons whythey took these actions, (3) the questions they asked themselves, and (4) thefeelings they experienced. They were also asked to identify the point at


which they felt they had a focus on the topic –– the point at which they knewwhat they were looking for.

5.4.5.5. Follow-Up Interviews Ten months after the participants hadfinished their research projects, and midway through data analysis, 4 ofthe 10 participants agreed to be interviewed again. The interviews were semi-structured and were conducted in-person on campus. The purpose of thefollow-up interviews was to present the participants (or at least four ofthem) with early results from data analysis and ask for feedback. Since theevents under study had taken place almost one year before, the researcherasked the participants to consider the results in light of their information-seeking behavior in general (but in terms of school-based tasks). The fourparticipants elaborated on the themes emerging from the study, offeringfurther insight into adolescent metacognitive knowledge during the ISP.

5.4.6. Data Analysis

Analysis was inductive and grounded in the data, using the words andactions of the 10 participants to tell the story of their journey through theISP. The data was transcribed and imported to a qualitative data analysissoftware application (Atlas.ti 5.2), and one hermeneutic unit was created(the ‘‘container’’ for all the data and coding). During coding, the data wassegmented, or chunked, into meaning units that lay at the sentence andparagraph level. Three levels of codes were developed –– descriptive,interpretive, and pattern (Miles & Huberman, 1994, p. 57) –– each levelbuilding toward a broad model of essential themes. At the descriptive levelof coding, the language of the participants was reflected in the code name asa way to identify, describe, and locate the chunk of data. The descriptivecoding sorted the data into three broad categories –– thinking (includingstatements that showed self-awareness or self-questioning), feelings (includingcertainty/uncertainty), and tasks in Kuhlthau’s ISP model.

At the next level of analysis, interpretive codes emerged. Interpretivecodes add meaning to data. In this study, data that was previously codeddescriptively as ‘‘metacognitive knowledge,’’ or even more broadly as‘‘thoughts,’’ took on a new, deeper meaning when recoded interpretively.For example, participant 3 spoke of ‘‘whipping’’ himself into shape andignoring information that was interesting but not relevant because he simplydid not have the luxury of time –– evidence of his awareness of theimportance of self-control, the role of motivation, and the demands of thetask. This text, previously coded under a broad category of metacognitive

104 Leanne Bowler

thinking, was coded interpretively as ‘‘control,’’ which relates to self-regulation, a key characteristic of metacognitive behavior.

A total of 291 descriptive and interpretive codes were assigned to thedata. The relationship between interpretive codes was then tested byquerying the data set and the creation of ‘‘super codes’’ or pattern codes.Thirteen pattern codes were identified. They are as follows: knowing yourstrengths and weaknesses, knowing what you do not know, scaffolding,building a base, parallel thinking, understanding curiosity, communicating,changing course, understanding time and effort, balancing, understandingmemory, pulling back and reflecting, and connecting. These codes pulltogether a lot of data and are more inferential and explanatory than thelower level codes. For example, the control category mentioned in theprevious paragraph was related to a larger pattern of thinking related tounderstanding time and its relation to cognitive effort. The pattern code waslabeled, not surprisingly, Understanding time and effort, and includes otherfeatures such as, for example, understanding how persistence interacts withcognition and the ability to use cognitive short cuts in timely and effectiveways. Given the vastness of the information environment, this pattern codeexplains the importance of understanding the relationship between time andone’s own cognitive effort is critical to the successful completion of aninformation problem. The pattern codes were also tested against the stagesin the ISP model. Although there was some alignment between thecategories of metacognitive knowledge and stages in the ISP, a grandpattern was not uncovered. The reasons for this are discussed below.

5.4.7. Limitations

Due to the small sampling size, generalizations beyond the context of thestudy will be difficult to infer. The 10 participants in this study were highacademic achievers in a Montreal area private school, and their behaviormay not reflect that of the general population. As well, some of theirbehavior may have been shaped by the type of information task assigned tothem by the teacher. The results of this study relate to a specific contentarea –– the history of western civilization –– and are not generalizable toother content areas or information-seeking tasks. In addition, the extent ofthe students’ prior domain, information system, and metacognitive knowl-edge, in relation to other students of their age, were not known, as thequalitative methods to be used in this study precluded the use of a controlgroup or wide sampling procedures. Only 2 of the 10 participants in thisstudy were male and therefore the study presents no findings regardinggender-based behavior.


5.5. Results

5.5.1. Taxonomy of Adolescent Metacognitive Knowledge Duringthe Information Search Process

The participants were hindered in their searches by a lack of proceduralknowledge related to information problem solving as well as a lack ofconceptual knowledge in the domain of history. The study found that theseroadblocks were mediated by metacognitive knowledge which was used asmuch as an emergency strategy as it was a deliberate line of attack. Whilethe young people in this study did show evidence of thinking ahead andplanning, their use of metacognitive knowledge was as often as not reactive,rather than predictive, exhibiting a short metacognitive horizon –– aparadox perhaps for a knowledge that is associated with planning.

Perhaps, more interesting was the characteristics of the participants’metacognitive knowledge. The metacognitive knowledge of the adolescentsin this study was surprisingly varied and wide ranging. The study uncovered13 attributes of adolescent metacognitive knowledge related to the ISP:balancing, building a base, changing course, communicating, connecting,knowing that you do not know, knowing your strengths and weaknesses,parallel thinking, pulling back and reflecting, scaffolding, understandingcuriosity, understanding memory, and understanding time and effort (seeTable 1).

Although most of the adolescents in this study demonstrated each of the13 attributes of metacognitive knowledge, they generally did so inidiosyncratic ways, so that it cannot be said that one pattern ofmetacognitive thinking overlays the entire ISP model. Rather, an array ofmetacognitive tools emerged, to be used as required to solve the informationproblem of the moment and it was clear that the patterns of thinking wereunique to each participant. In other words, not every participant applied all13 types of metacognitive knowledge uniformly –– some privileged oneattribute over another, reusing one type of metacognitive knowledge overand over, while utilizing other types of metacognitive knowledge infre-quently. In keeping with the ethnographic approach of this study and itsfocus on the totality of the participants’ information experiences, the 13attributes have been drawn together to create a taxonomy of adolescentmetacognitive knowledge. These forms of knowledge have been labeled ISPmetacognitive knowledge in recognition that they relate specifically to theISP. A description of each type of ISP metacognitive knowledge follows.

5.5.1.1. Balancing Balancing relates to the cognitive task of makingchoices, sometimes between two desirable options, and knowing thatmaking choices helps to move you forward in the search process because if

Table 1: Taxonomy of adolescent metacognitive knowledge during theinformation search process.

Balancing Parallel thinking� Filtering information� Finding an equilibrium� Awareness of the weakness

of a strategy� Weighing the options

� Double thinking� Future thinking� Planning

Building a base

Pulling back and reflecting

� Browsing� Doing the groundwork� Finding synonyms� Scanning� Broad to narrow search strategy� Seeing the big picture

� Reflecting

� Taking a break from the task� Following up a lead� Reviewing

Changing course

Scaffolding

� Adapting the topic� Ending exploration� Shifting the search strategy� Simplifying the search strategy

� Mapping� Modeling the structure of information objects

(using information architecture to build a

conceptual model)� Modeling expert thinking� Using a pathfinder� Using meta-tools (encyclopedias,

etc.)

CommunicatingUnderstanding curiosity

� Using talk as a metacognitive strategy� Asking, Who do I talk to?

� Using curiosity to drive the search� Regulating curiosity when it conflicts with the

search� Awareness of interests and

motivations

Connecting

Understanding memory

� Clarifying� Connecting information� Using part to whole thinking� Using a ‘‘narrow to broad’’ search

strategy� Using a ‘‘breaking apart and rebuild’’

strategy

Knowing that you do not know

� Identifying gaps in knowledge� Identifying prior knowledge

� Awareness of one’s own powers of memory� Highlighting text as a strategy to aid memory� Note taking to aid memory� Using ‘‘Post-it’’ notes to aid memory

Knowing your strengths and weaknesses

Understanding time and effort

� Identifying personal characteristics (how I

learn)� Identifying prior knowledge� Identifying personal characteristics

(general knowledge of self)� Identifying strengths/weaknesses

� Awareness of how an investment in time and

effort pays off cognitively� Awareness of the role of patience and

persistence� Understanding the role of effort� Using a cognitive short cut� Understanding the effort required to

summarize

106 Leanne Bowler


you do not make a choice, the search will be stalled. Balancing is also aboutweighing the options and making compromises. For example, choosingbetween feeding curiosity and finishing the assignment; choosing betweenprecision and recall; choosing one of two interesting topics; finding abalance between ‘‘good’’ information and information that is ‘‘goodenough.’’

Balancing has a strong evaluative component to it, and although therewas evidence that the students were aware of the cognitive complexities ofthe task of choosing and that they had an ability to make critical choices,many of the students actually found this aspect difficult. The problem forthem, however, was not just that there was too much information –– what iscommonly called ‘‘information overload’’ –– but that there was a choicebetween a few equally credible but contradictory sources. Perhaps for thefirst time in their lives, the students faced an information problem whoseanswer lay in shades of gray, rather than in black and white.

5.5.1.2. Building a Base The ISP is sometimes perceived in metaphoricalterms as a process of construction, where one makes sense of informationstep-by-step, brick by brick (Dervin, 1999; Kuhlthau, 1991, 2004). As withall construction, knowledge that emerges from the ISP must be built upon astrong foundation that can support the structure that rests upon it. Therealization of this metaphor was fully expressed in the actions of theparticipants in this study. The metacognitive knowledge related to theseactions was labeled building a base and it refers to the strategic use ofexploratory tactics to help build foundational domain knowledge. As oneparticipant explained, the Web proved to be a good environment in which todo this:

I just looked through really broad sites. I wasn’t going fordetails yet and I sort of had an idea of what I was interested in–– the topic –– so I was just looking for information that wasout therey I didn’t look at articles. I just looked at webportals, some general information.

5.5.1.3. Changing Course Changing course is metacognitive knowledge thatis used to guide the choice of new tactics and strategies when the search isstalled or less fruitful than expected. Most of the changes in tactics andstrategies made by the participants were related to IR or the selecting andgathering of resources. But the participants also made adjustments to theirtopic selection, a tactic designed to make it easier to find and useinformation later in the search process.

What makes this metacognitive? Changing course is dependent on theability to assess one’s status during the search process. Information seekers

108 Leanne Bowler

who decide to change course realize that tactics and strategies previouslyapplied either did not work or only worked up to a point. Changing courseis also representative of the ability to plan and predict an outcome becauseit reflects evidence of thinking ahead to the next step. Without this type ofmetacognitive knowledge, information seekers will not know whichchanges to implement and will continue along an unsuccessful pathtoward an uncertain future. Implicit in this attribute of metacognitiveknowledge is the assumption that information seekers are actually on aself-directed course of action, and not just making random decisions. Thestudy found that while the participants had the metacognitive means toadopt new tactics and strategies, these changes were applied to immediatecognitive difficulties, and not necessarily to the long-term problem. Sowhile the participants were able to predict an outcome, for many of themit was the outcome for the next step in the process, and not the finaloutcome of the school assignment.

5.5.1.4. Communicating This attribute of adolescent metacognitiveknowledge involves the use of people as information mediators andinformation sources during the search process. Quite simply, it is knowingthat talking to people is a useful cognitive strategy. Talking to peopleserves many cognitive purposes during the search process. ‘‘Talk’’ can helpto clarify points of confusion about conflicting information or it can helpto unite information into a cohesive unit. ‘‘Talk’’ can also be a quicksource of information, helping to build a knowledge base, or can provide aroad map for the next steps in the process. Information seeking can beseen as a solitary process –– if one has an information problem onesearches, for example, the library’s catalogue, the book shelves, online databases, or Web portals, alone, for a solution. And yet the participants inthis study used information mediators frequently and in a deliberatemanner to help them solve their information problems. They turned topeople in their network of relationships because they knew it was a goodstrategy for helping them to make sense of the problem. Knowing that thestrategy exists, when to implement it and why, is representative ofmetacognitive knowledge.

5.5.1.5. Connecting Participants who used this type of knowledge sawknowledge building as a process of construction and they understood that inorder to make sense of the disparate pieces of information they hadgathered, they now had to implement strategies for making connections.This type of metacognitive knowledge was demonstrated by one participantwho, when looking at all the information he had gathered, asked himself‘‘How I can make the links more coherent or obvious?’’ He now understoodthat his conceptualization of the information was more important than


having to find more. Although related to Building a base and Scaffolding(because it is about finding ways to model and construct knowledge), thiscategory of metacognitive knowledge is distinct because it is about thespecific task of creating links between pieces of information. Much of thisactivity occurred in the final stages of the ISP.

One could conceive of connecting as a linking process –– the act ofdefining the relationships between nodes in a mental map. If each piece ofinformation found in a search represents a node, then the relationships thatbind them together represent the links. During the course of informationseeking, many nodes may be found but, unless linked together, theinformation seeker will not understand how all the nodes fit together intoa whole. Connecting is dependent on knowing that you do not know, anattribute of metacognitive knowledge that triggers the steps involved inlinking information. Information seekers who know that they do not knowsay to themselves, I have all these pieces of information but now I need toconnect them so that they make sense to me. What steps can I take to tie thisinformation together?

5.5.1.6. Knowing that You Do not Know Knowing that you do not know is atype of self-knowledge and is related therefore to knowing your strengths andweaknesses. There are, however, differences between knowing that you do notknow and knowing your strengths and weaknesses. While knowing yourstrengths and weaknesses is dependent upon one’s ability to put a name towhat you know and then taking advantage of it, knowing that you do notknow is a state in which you can identify a gap in your knowledge base butyou cannot always say what that gap is. In other words, you cannot put aname to it. In the context of information science theory, knowing that you donot know is associated with Belkin’s (1980) anomalous states of knowledge(the ASK) and Dervin’s (1999) sense-making framework. Both modelssuggest that information seekers are prompted by an awareness of a gapbetween order and chaos. As simple as it may seem, knowing that you do notknow is a critical piece of metacognitive knowledge during the ISP because italerts the information seeker that there is a problem and it is time to reviseor make adjustments to the search strategy. However, in order to know thatyou do not know, you still need some knowledge of the territory, a paradoxthat prompts one to ask how much do you need to know in order to knowthat you do not know?

5.5.1.7. Knowing Your Strengths and Weaknesses In order to use yourstrengths you must first know what they are. Sometimes your strengths canonly be revealed when you analyze your weaknesses. Self-knowledge isdirectly related to self-assessment, the ability to critique one’s own cognitiveand affective states being a commonly accepted attribute of metacognition.

110 Leanne Bowler

With this type of metacognitive knowledge, the participants were able to puta name to what they knew and then take advantage of it. They were alsoable to pinpoint what they did not know. With this self-awareness, theparticipants were able to assess the usability of the information vis a vis theirown information needs. Often, there seemed to be a mismatch between theirown knowledge of history and the way that history was represented to themin the information resources they use. It seemed to them that they, thestudents, were the wrong audience. As one participant explained:

Articles from journals are really quite useless at my level. Theyexpect you to have a fountain of knowledgeyThey go onthis itsy-bitsy topic they’ve been studying for three weeks or, Idon’t know, months.

5.5.1.8. Parallel Thinking The participants’ thinking in this study was notalways ‘‘in-the-moment’’; while doing one thing (and thinking about it), theywere also thinking about something else. They frequently reviewed the pastand predicted the future –– even as they acted in the present. Some of thisthinking represented deliberate planning –– thinking about the next steps totake or forecasting one or two stages down the road. At other times, thethinking was more of an envisioning of the outcome or merely a vaguecuriosity about how the information search would unfold. To capture theimage of two streams of simultaneous thought, the term parallel thinkingwas applied to this attribute of metacognitive knowledge. Parallel thinking isvisualized as crossing over all other categories of metacognitive knowledgeduring the ISP.

Parallel thinking was most closely linked to the latter stages in the ISPeven though, in terms of actions, the participants were still located in earlystages of the search. In other words, the participants were anticipatingoutcomes even as they worked their way through the first tasks in thesearch process. For example, one participant, wondering about therelevance of information she was currently gathering, asked herself, ‘‘Isthis pertinent to my topic?’’ But, thinking ahead to when she would haveto construct the essay, she then asked, ‘‘Can I actually use this info? CouldI integrate it?’’

5.5.1.9. Pulling Back and Reflecting Taking the time to back away from aproblem and think about it proved helpful to some of the participants in thisstudy. As one participant advised, when confused, it helps to ‘‘leave for anhour. I find when I’m working on it too long it gets frustrating and I stopfocusing.’’ As little more than half of the students showed an awareness ofthe metacognitive benefits of reflecting, reviewing, and just stepping awayfrom the problem, this category presents itself as much as a gap in


knowledge as it is strength. Perhaps, this is more a reflection of the tightdeadlines and heavy workload carried by the students in this study than it isof any strength or weakness in metacognitive knowledge.

5.5.1.10. Scaffolding Knowledge of metacognitive strategies, knowingwhen to apply them and being able to anticipate their ‘‘pay off,’’ is acritical piece in the metacognitive knowledge toolkit. The students in thisstudy, perhaps knowing full well that they ‘‘did not know,’’ activelyimplemented a range of metacognitive strategies to help them see whatthey should know, one of which was scaffolding. Scaffolding refers to theact of searching for and using a cognitive support, or reinforcement, tohelp map out a conceptualization of the information environment. Theparticipants looked for support structures to help them map out theinformation environment, often in the very information they weresearching. For the students in this study, it did not matter where thestructure came from –– a person, a book, or even Wikipedia –– theimportant thing was its availability at the right time, providing somesignposts to help guide them through the ISP.

The structure of books helped several students. Asked to identifysomething that helped her focus in on her topic, one participant explainedhow the organization of a book helped her:

Well there was one book in particular. I really liked the way itwas organized. It had two chapters that were related to mytopic and then I looked at how they explored that. I looked atthe introduction and it explained how they could deduce allthis from the artifacts and then it went on to the factsy and itcame back to queenship in general.

Tools contained within books, such as bibliographies and indexes, actedas pathfinders –– a kind of travel guide to the information environment ––and several students actively used these tools as a scaffold. While theattribute of scaffolding is certainly an attempt to build knowledge, it isdifferent from building a base in that scaffolding is a strategic use of apreexisting structure to help map the information environment, whereas‘‘building a base’’ is a more open-ended exploration of the environment,akin to browsing.

5.5.1.11. Understanding Curiosity The participants in this study oftenfound themselves forced to make stark choices between their need todiscover versus their need to fulfill the requirements of the schoolassignment. The regulation of this conflict represents a special type ofmetacognitive knowledge which has been labeled as understanding curiosity.

112 Leanne Bowler

It involves a risk/benefit analysis that hinges on understanding how farcuriosity can take you before it becomes a liability, rather than a benefit, forthe task at hand. A few of the participants did not grasp the relationshipbetween curiosity and control and, as a consequence, gathered too muchinformation and then found it difficult to compress it into a neat package.

5.5.1.12. Understanding Memory Exploring an information-richenvironment, such as a large academic library, an electronic database, orjust surfing the Web, is like deep sea fishing: The information seeker casts awide net into a big ocean and draws in all manner of information sources.Not everything in the net is useful, but sometimes this is not known untilmany of the sources have been sorted. By this time, some informationseekers forget where the critical pieces of information are located. If theyhave not used techniques to help them find their way back to theinformation, it could be lost to them forever. Remembering whereinformation is located is an important part of the ISP. Understanding therole of memory in information seeking, knowing that it is difficult toremember everything, knowing how one’s own memory works, and knowinghow and when to use specific strategies in order to help one remember whereinformation is located so that it can be retrieved later are all importantmetacognitive aspects of the ISP. Remembering is no doubt assisted by astrong conceptual understanding of the information environment, but sincenovice information seekers do not always have this, they must depend on thelittle tricks that help them remember the pathway back to relevantinformation.

5.5.1.13. Understanding Time and Effort Successful outcomes are often theresult of sustained effort, attention to detail, and a consistent level ofpersistence. Woven throughout the data is evidence that the participantsunderstood the connection between effort and results at a general level butthat this understanding did not always translate into action. In other words,‘‘knowing’’ did not always relate to ‘‘doing.’’ The problem was perhapsrelated to the participants’ understanding of the task at hand –– completinga research paper for college. While the participants understood that effortgenerally pays off in life, many of them simply did not see the specific task ofsearching for information as something that required effort. These points toa lack of metacognitive awareness related to the cognitive demands of theinformation search.

Another facet of effort is the understanding associated with knowingwhen not to invest effort in a task. While effort is often necessary inorder to complete cognitively demanding tasks, sometimes the wisestaction to take is to do nothing (or at least very little). Taking a shortcut,or doing something that simplifies the process, may certainly require less


effort and if by doing less, one avoids wasting cognitive effort then it is auseful strategy. Managing time is a crucial skill for students. Withobligations and deadlines to consider, how much time one devotes toeach task, as well as when to work on it, becomes a critical decision.School projects that last the term present a particular ‘‘time manage-ment’’ problem for many students. As one participant said, long-termprojects are difficult because ‘‘it’s always the last on my list of priorities.Everything else comes first.’’

5.6. Mapping Adolescent Metacognitive Knowledge to the

Affective and Behavioral Dimensions of the Information

Search Process

5.6.1. Uncertainty and Adolescent Metacognitive Knowledge

As a group, the participants in this study did use metacognitive approachesto deal with their uncertainty, but the approaches were triggered in themoment, as an emergency response rather than as a well-formed plan. Thisis perhaps a reflection of the gap in knowledge related to both proceduraland content-area knowledge. The participants were working with a blankcanvas. Not knowing what they were getting into, the only approach left tothem was to micro-manage uncertainty via the application of specificcognitive strategies. Perhaps, this is the only approach for any informationseeker who is dealing with new areas of information.

The gap in content-area knowledge sometimes meant that the bestmetacognitive strategy for dealing with uncertainty was to turn to peoplewho did have content-area knowledge (communication). P1, for example,read much of the material she had gathered but was still confused about themeaning of the information, still ‘‘trying to grasp’’ the subject and ‘‘findgeneralities.’’ Prompted by her own question ‘‘what does this mean?’’, shetelephoned three experts in the topic of musicology –– professors at a localuniversity –– who she hoped would help to clarify her understanding. Thisaction was launched by a sense of cognitive uncertainty. Other approachesthe students used were to deliberately make connections (connecting), tofilter and weigh the value of information (balancing), to consider the timeand effort required to complete specific tasks in the search process(understanding time and effort), and to think about ways to remembermasses of information (understanding memory).

Uncertainty for these students did not always come in the shape of a‘‘lack of understanding,’’ as defined by Kuhlthau (2004, p. 92). The gap incontent-area knowledge sometimes led participants to attribute their

114 Leanne Bowler

cognitive confusion to factors external to their own thinking, such as theundiscovered territory of the information environment. The problem wasnot that they were confused; it was that they were uninformed –– theysimply did not know what was out there. Was the information environmentlarge, they wondered? Was it small? Would there be enough information toresearch the topic? P7, for example, was only comfortable with her topiconce she had confirmed that there was enough information to complete theessay. She marked an X on her timeline to indicate that this was the pointat which she was finally able to focus on her topic. Uncertainty turned tocertainty for P6 when she decided there was ‘‘enough reliable info tosupport & to prove it? [the hypothesis]’’ This decision made P6 ‘‘lessanxious.’’

For others, uncertainty was driven by the formulation of a usablehypothesis (as represented by a research question) and not by their ability tomake sense of information. The hypothesis drove the search, so a hypothesisthat made no sense translated into a search that made no sense. Asked if shehad chosen a title for her essay (a reflection of a focus on a specific topic), P2answered, ‘‘Not really. I’m not sure about the hypothesis.’’ For this student,eliminating uncertainty was about defining the research question via atightly formulated hypothesis. As P2 said, it was about ‘‘trying to refocus.What was my hypothesis before, wasn’t really a hypothesis. I’d like tonarrow it down.’’ But narrowing the hypothesis meant first exploring theinformation. For P2, reducing uncertainty about the hypothesis meant shefirst had to build up her knowledge base (building a base) and only thenweigh the options (balancing) because, as she explained, ‘‘First I need toexpand it before I narrow it down.’’

Uncertainty, according to the ISP model, is associated with under-standing the meaning of information, the tipping point between uncertaintyand certainty happening during focus formulation. According to the ISPmodel, during the latter stages of the search –– collection and presentation ––it is assumed that uncertainty has been dealt with and the information seekercan move on. In this study, however, uncertainty relating to the final stagesin the search process loomed large. The students, thinking ahead to whenthe information would need to be synthesized and packaged into a product,worried that they did not have the skills and knowledge needed to create acohesive unit, this being for many their first experience at writing a detailedresearch essay. More specifically, they were unsure about how to constructan outline that would reflect the compare/contrast nature of the assignmentand they were confused about the rules for citing sources. Many studentsacknowledged this sense of confusion and took steps to alleviate it, either byvisiting the teacher for clarification, talking to parents about the next stepsin the process, or by looking for information about how to write a researchpaper.


Uncertainty, as it emerged in this study, took on many faces and wasabout more than an ability to construct a personal understanding of themeaning of information. As a result, there is no single pattern ofmetacognitive approach for dealing with uncertainty. Rather, the partici-pants chose approaches that fitted the problem of the moment and dealtwith uncertainty in individualistic ways.

5.6.2. Tasks in the Information Search Process and AdolescentMetacognitive Knowledge

Although this study did not set out to test or redesign the ISP model, theexperience of using it to code the data provided the researcher with someinsight into its nature. Framing the study around the six tasks in the ISPmodel actually proved quite difficult at times and it is worthwhilecommenting on the reasons for this. The participants in practice movedback and forth between prefocus exploration and information collection in ahighly iterative manner. The nonlinear nature of the search process has beenconfirmed in other studies. Kracker and Wang, for example, suggested thatan ‘‘iterative’’ variable be added to the model in order to address its spiralnature (Kracker, 2002; Kracker & Wang, 2002). Chung and Neuman (2007),in their study of high school students’ information seeking, also found thatthe tasks of gathering and selecting (related terms for exploration andcollection) repeated themselves in a dynamic, nonlinear way. One reasonaccounting for the nonlinear nature of the search process may have been thecourse requirement to produce a critical bibliography early in the semester,which forced the students to explore and make hard decisions aboutinformation almost simultaneously.

Another issue to emerge from the coding was the difficulty inidentifying exactly when focus formulation was achieved. The participantsidentified on a timeline the point at which they felt they had achieved afocus on their topic. For some, focus formulation seemed to happen at theend of the process, as they delved into their sources and started to lay outa plan for the essay. This was the case for P3. Marking his focusformulation with an ‘‘X’’ on the timeline, he then noted ‘‘Copying notes/quotes, Organization of my text.’’ Three participants said they formed afocus more than once, P6 reporting three times. But their topics did notchange substantially, suggesting a gradual move toward understandingrather than a radical shift in topic. Looking at the frequencies of thecoding for ISP tasks, one sees that focus formulation is dwarfed incomparison to the frequency of coding for its neighbors, exploration andcollection. One reason for this may be that the students did not achieve a

116 Leanne Bowler

focus on their topic –– they simply jumped ahead to the next stage. Butanother reason might be that focus formulation is cumulative rather than aspecific point in time. It is therefore difficult to identify it as a discretestage in the search process. Focus formulation seems to emerge well intothe latter stages of the search process (as witnessed by the students whofelt they had a focused understanding of their topic only after they startedpreparing an outline, or for some, even after they began writing), so onlythe teacher reading the final essay could be in a position to assess whethera focus had been achieved.

Another finding was that there were different genres of search embeddedwithin the larger search process. The ISP model as presented in theliterature seems to present a one-to-one relationship between the searchprocess and the information need. In other words, the beginning, middle,and end of the process represent one unit driven by one purpose. Thisstudy found that embedded within the search process were several genres ofsearch, each one representing a qualitatively different information need (theidea of genre is used here to mean a type of search). Although each of thesesearches served the larger purpose of the school assignment, each wasdriven by distinct information need (see Bowler, 2009, for more detailsabout the genres of search in this study). It is hoped that the insightsoffered here into the nature of the ISP model will help to expand ourunderstanding of the model. While new elements additional to the modelwere observed, and the pathway of the search process was more iterativethan expected, the ISP model did, nevertheless, predict the information-seeking tasks fulfilled by the students. At some point, each of the studentsdid go through task initiation, topic selection, prefocus exploration, focusformulation, and information collection and presentation, but not in theexpected or prescribed order.

5.7. Discussion

5.7.1. Challenges Facing the Study

At the outset of this study, it became apparent that there were numerousmethodological challenges which would shape its design. In a nutshell, howdoes one investigate the deep layers of thinking in an authentic way, amonga population that is difficult to access, and in such a manner that the entiretyof a process that lasts several months is captured? As one of the goals of thisbook is to explore novel approaches to solving methodological problems inthe study of information behavior, this section explores the challenges facedby this study further.


Looking at metacognitive knowledge through the lens of Kuhlthau’s ISPmodel (in other words, as a series of successive search tasks which togetherhelp to construct knowledge, rather than as one discrete incident of IR)meant that the study was necessarily longitudinal and therefore methodshad to be devised that would capture change over time. Furthermore, inkeeping with the purpose of this study, data was collected in a naturalisticsetting, rather than in a controlled laboratory environment, in order tocapture the breadth and scope of variables associated with metacognitiveknowledge during the ISP. Since it was not known when the participantswould choose to conduct their information search (as soon as theassignment is given by the teacher or the night before it is due) nor wherethey would choose to search (in the school library or at home, from theircomputer), direct observation was not possible. Finally, the participants inthis study were active and busy young people, many with conflictingdemands on their time. The problem of how to encourage their long-termengagement in the study (almost four months) was a serious consideration.Given these constraining factors, it was critical to devise a methodology thatwas portable, user friendly, interesting, and noninvasive.

Layered over these issues are questions related to the study of thinkingprocesses. Some studies in library and information science that sought touncover the metacognitive processes of thinking during the ISP have doneso from a deductive point of view (e.g., Wolf et al. (2003) tested a preexistinginstructional model and Gorrell et al. (2009) study developed a taxonomy ofmetacognitive skills for Web searching based on a review of the literature).A deductive is valuable, as it performs the necessary task of testinghypotheses derived from previous studies and synthesizing research findingsinto a meaningful whole. The study described in this chapter, however, usedan inductive approach identified here as metacognitive ethnography.

The starting point for this study was the assumption that informationsearch is a distinct phenomenon and as such needs its own set of categories,concepts, frameworks, models, and theory. Second, the study assumed thatthere are developmental differences between the way that adults andadolescents search for information. Rather than extrapolate from studiesthat have emerged from other disciplines and with adult populations, thisstudy set out to start from a clean slate, as it were, and build categoriesthat speak to the phenomena of adolescent information seeking. To dothis, the study took a naturalistic, inductive approach in order to deriveconcepts that were grounded in the real lives and authentic informationtasks of the participants. The resulting taxonomy, although potentiallyrelevant to many aspects of learning, is intentionally specific to the ISP ofadolescents.

How do you look at metacognitive knowledge in naturalistic ways?Metacognitive knowledge, by its very nature, lies under the surface and is

118 Leanne Bowler

often not even apparent to the thinker herself. How is it possible to lift thelid on the box called ‘‘metacognition’’ when it may actually be hidden fromeven the participant’s conscious awareness? Trying to investigate thisphenomenon is akin to using shadows in order to determine the sun’sposition in the sky –– rather than look directly at the sun, one must useinference and interpretation to draw conclusions. One cannot simply ask‘‘what do you know about your own thinking?’’ and assume that theanswers represent reality. Verbal reports are, after all, just data –– ‘‘nothingmore, nothing less’’ –– and should not always be taken at face value (Genest &Turk, 1981, p. 244). Why is this? First of all, participants may not bepracticed in externalizing their thoughts and may simply lack the skillsneeded to make their knowledge about their own thinking explicit to others.Second, there is no guarantee that the answer is authentic –– is it what theparticipant really thinks or is it what they want the researcher to know?Third, the participant’s memory of the event may be hazy, especially if thereport is not in close proximity to the event.

A common practice in studies that rely on verbal reports istriangulation or crosschecking of information to corroborate the evidence.In this study, triangulation during data collection occurred on three levels.First of all, four types of data collection instruments were designed andused. Second, the instruments were applied at different times throughoutthe study in order to contrast ‘‘think-aloud’’ data (information that isreported at the time of the event) and ‘‘think-after’’ data (information thatis reported at a later point in time) (Anders Erisson & Simon, 1980;Branch, 2000). In addition, data collection questions were crafted in sucha way as to ensure that the participants’ thoughts were juxtaposed next totheir actions. This was done to highlight patterns or produce incon-sistencies among findings about the same phenomenon and to provideevidence of competency. Irrespective of how well data is triangulated, thequestion remains –– does it open a window on thinking? This problemwas tackled specifically through open-ended data collection questions thatasked the participants to answer why they did what they did (in otherwords, what was their rationale) and what types of self-prompts guidedtheir thinking.

Those who conduct research with youth know that ‘‘do-ability’’ is alwaysa major concern. Devising a research method that would, on the one hand,reveal metacognitive thinking related to information seeking in a natural,nonintrusive way but would, on the other hand, provide credible evidence oftheir metacognitive thoughts was critical to the success of this study. Inshort, a balance had to be found between methods that could beimplemented in nonintrusive ways and methods that were actually useful ––a difficult task indeed. One solution to this conundrum was to adapt amethod from cognitive science. Called cognitive ethnography, it is a way to


study thinking from a sociocultural perspective –– cognition in its naturalhabitat or thinking ‘‘in the wild’’ (Hutchins, 1995). This study took a similarapproach, although it focused on metacognition –– the thinking behind thethinking –– rather than the broader perspective of cognition. For thisreason, the method used has been called a metacognitive ethnography.

Hutchins coined the term cognitive ethnography when he studiedcognition in non-laboratory settings (e.g., navigator’s practices on thebridge of a ship in the US Navy). Calling it the ‘‘alternative history ofcognitive science,’’ Hutchins argued that cognition is part of a socioculturalprocess. From this perspective, cognition is something that can best becaptured longitudinally, over the course of time, and in context. Cognitiveethnography is applied ethnography and does not necessarily fit into theprototypical template of ethnography as defined by anthropology and socialsciences. For one thing, traditional ethnography starts from a clean slateand is not theory driven, while this study, as an example of appliedethnography, was framed by the theories of Flavel and Kuhlthau.Traditional ethnography is concerned with the meanings that members ofa cultural group create, but ‘‘cognitive ethnography is concerned with howmembers create those meanings’’ (1995, p. 838).

Let us take a closer look at this approach. First of all, the phenomenoninvestigated in this study was situated in a sociocultural context and thatcontext was the everyday lives of the participants which revolved aroundhome, school, and community. The study asked the participants to recordtheir thoughts, feelings, and actions as they searched for informationanytime, anyplace, using any kind of informational resource. As a result, theparticipants’ ISPs were not restricted to library use. Their informationpractices went beyond the walls of the school, to include the people in theirlives and informational resources across the city. In this sense, the study waslooking at information ecology, and not a particular library system. Whilethe study did not look at information problem solving from a groupperspective (the 10 participants were not solving a group problem), theywere all operating within the sociocultural context of home, school, and thecommunity of Montreal, Canada.

5.7.2. From Taxonomy to Practice: Next Steps

This study generated a new model of metacognitive knowledge in the formof an original and distinct taxonomy of ISP metacognitive knowledge. Inaddition, the findings of this study will help to expand general models ofmetacognition. It is perhaps not a surprise that there was some evidence ofmetacognitive knowledge in this study, given that the participants were

120 Leanne Bowler

academic achievers and nearing the tail end of their adolescence. We doknow that children as young as five years old have been able to demonstrateself-regulatory, self-monitoring behaviors, so information behavior among16- to 18-year-olds that is completely devoid of metacognitive practiceswould be highly suspect (see, e.g., the work of Bodrova and Leong (2007)and their Tools of the Mind curriculum for preschoolers for research in thearea of young children and self-regulation).

More interesting is the surprisingly diverse and contextual quality of thisknowledge. The idiosyncratic nature of the participant’s metacognitiveknowledge raises some questions. Why did some participants favor oneattribute over another? Do the highly individual approaches seen in this studyrepresent individual personality differences, the participants’ attempt tocustomize their response to match the information topic, or a lack of trainingin metacognitive ways of thinking related to information seeking? The shorthorizon of the participant’s metacognitive knowledge is also of note.Was thisshort horizon due to their undeveloped metacognitive thinking or is thissimply the nature of information seeking in general? Further study in theseareas is necessary in order to explore these questions more fully.

At the beginning of this research project, a model that would mapmetacognitive knowledge directly on to Kuhlthau’s ISP model wasenvisioned, such that one could see how one type of metacognitiveknowledge was perfectly aligned with one specific task in the ISP. Thiswas not to be the case. Most types of metacognitive knowledge identified inthis study were used across the ISP model, from beginning to end. Whilesome categories of metacognitive knowledge had stronger tendencies towardone task in the ISP over others, it cannot be said that a direct one-to-onepattern between the 6 stages in the ISP model and the 13 categories of ISPmetacognitive knowledge exists. As has been suggested earlier, this isprobably due to the global nature of metacognitive knowledge.

For this reason, the 13 types of metacognitive knowledge presented heredo not represent a grand model of behavior but rather taxonomy of therange of possibilities that exist. Metaphorically speaking, the taxonomyrepresents a toolkit of intellectual skills, to be used as need requires. Whatthe taxonomy does is to define and describe the general characteristics ofeach type of ISP metacognitive knowledge. Using this as a baseline, furtherresearch can expand the definitions to include intensity of application andcontextual attributes (e.g., What does metacognitive knowledge look likebeyond the school-based inquiry project? Are some categories ofmetacognitive knowledge more relevant in different settings?). Somethingelse to consider is that adolescence has a developmental arc and themetacognitive thinking modeled in this study represents that of the olderadolescent. What does it look like in the 13-year-olds? The picture may bequite different.


The taxonomy of adolescent metacognitive knowledge may contribute tothe design of more intelligent and ultimately more effective IR tools for thisuser group by providing a road map to the strengths and gaps in adolescentmetacognitive thinking. An IR system that models effective metacognitivedecision making and addresses gaps in metacognitive thinking might helpyoung people find meaning in information by providing the kind of supportneeded to help them monitor their own progress through the information-seeking process. But the purpose of designing an intelligent IR system that ismodeled on metacognitive knowledge would not be to create a tool thatsimply thinks for young people. IR systems that provide the user withopportunities to experience metacognition may lead to the transfer ofmetacognitive knowledge from the system to user, leading perhaps tosmarter users.

Finally, this study may provide the basis for the design of library servicesappropriate for adolescents, such as information literacy instruction. Theterm ‘‘metacognitive knowledge,’’ rather than the more general term‘‘metacognition,’’ was used quite deliberately in this study in order toemphasize the idea that metacognition is not just something people do butsomething that people know to do. Indeed, Anderson and Krathwohl (2001),in their revision of Bloom’s taxonomy, list metacognitive knowledge as oneof four types of knowledge, the others being factual, conceptual, andprocedural. In the context of Library and Information Science, metacog-nitive knowledge, as knowledge, is one part of the package that teachers,librarians, and designers of interactive information systems must deliver toinformation seekers. The implications of this on how young people aretaught to search for, choose, and use information are tremendous. As acritical piece of the information literacy puzzle, metacognitive knowledge iscontent that needs to be taught, and if metacognitive knowledge is to betaught, then methods for teaching and assessing it must be developed.

5.8. Conclusion

Metacognitive knowledge lies at the deepest layers of thinking and can bedifficult to capture, so exploring it in the context of information seeking –– aprocess that can happen anyplace, anytime, and anywhere –– was achallenging proposition. To do so with adolescents, a population that istraditionally hard to access, added yet another layer of difficulty. Thischapter presented the results of a study that tackled this difficult problemand which resulted in taxonomy of adolescent metacognitive knowledgeduring the ISP. The chapter also laid out a method for meeting the seriousmethodological challenges faced by this study, demonstrating that it is

122 Leanne Bowler

indeed possible to uncover the thoughts and feelings of adolescentinformation seekers in the context of their everyday lives.

The young people who participated in the study were active partners inresearch and those who committed to the study completed it. Many of theparticipants were intrigued by the study and hoped to read about it one day.The fact that none of the 10 participants withdrew from the study suggeststhat the methods did not threaten or intrude on the participants’ busy lives.The participants were generally open about their thoughts and feelingsabout information seeking and had strong opinions about the accessibilityof information (or lack thereof) as well as their own abilities to find it. Thenotion of actually stopping to think about how to find, choose, and useinformation was a novelty for most, and many found the idea intriguing.Despite the methodological challenges faced by this study, the researchdesign allowed for triangulation provided a framework for the expression ofthe participants’ thoughts, feelings, and actions and resulted in a rich andvaried data set.

The attributes of metacognitive knowledge identified in this study mayprovide a road map for the development of a metacognitive toolkit that, iftaught to young people, may help them search for information. The 10young people who participated in this study helped to lay out this map byshowing how they were helped or hindered by the geography of their ownmetacognitive knowledge. It is hoped that the map they set out will be usedto assist other adolescents who will be stepping into the adult world andnavigating through new, uncharted territory.

Acknowledgments

The author gratefully acknowledges the vital contribution of the youngpeople who participated in this study. The research was funded in part byresearch grants from McGill University (the Herschel and Christine VictorFellowship in Education) and the Fond quebecois pour la recherche sur lasociete et la culture (FQRSC).

References

Agosto, D. E. (2002). A model of young people’s decision-making in using the Web.

Library and Information Science Research, 24(4), 311–341.

American Association of School Librarians. Standards for the 21st century learner.

Chicago, IL: ALA. Retrieved from http://www.ala.org/ala/mgrps/divs/aasl/guide

linesandstandards/learningstandards/standards.cfm. Accessed on November 14,

2010.

http://www.ala.org/ala/mgrps/divs/aasl/guidelinesandstandards/learningstandards/standards.cfm

http://www.ala.org/ala/mgrps/divs/aasl/guidelinesandstandards/learningstandards/standards.cfm


Anders Erisson, K., & Simon, H. A. (1980). Verbal reports as data. Psychological

Review, 87(3), 215–251.

Anderson, L. W., & Krathwohl, D. R. (Eds). (2001). A taxonomy for learning,

teaching, and assessing: A revision of bloom’s taxonomy of educational objectives

(Abridged ed.). New York, NY: Addison Wesley Longman.

Beheshti, J., Bowler, L., Large, A., & Nesset, V. (2005). Towards an alternative

information retrieval system for children. Chapter 8. In: A. Spink & C. Cole (Eds),

New directions in cognitive information retrieval. Dordrecht: Springer.

Belkin, N. (1980). Anomalous states of knowledge as a basis for information

retrieval. The Canadian Journal of Information Science, 5, 133–143.

Bertland, L. (1986). An overview of research in metacognition: Implications for

information skills instruction. School Library Media Quarterly, 15(Winter), 96–99.

Bodrova, E., & Leong, D. (2007). Tools of the mind: The Vygotskian approach to

early childhood education (2nd ed.). Boston, MA: Allyn & Bacon.

Bowler, L. (2009). Genres of search: A concept for understanding successive

search behaviour. Canadian Journal of Library and Information Science, 33(3/4),

119–140.

Bowler, L. (2010a). The self-regulation of curiosity and interest during the

information search process of adolescent students. Journal of the American

Society for Information Science and Technology, 61(7), 1332–1344.

Bowler, L. (2010b). A taxonomy of adolescent metacognitive knowledge during the

information search process. Library and Information Science Research, 32(1),

27–42.

Branch, J. L. (2000). Investigating the information-seeking processes of adolescents:

The value of using think alouds and think afters. Library and Information Science

Research, 22(4), 371–392.

Branch, J. L. (2003). Instructional Intervention is the key: Supporting adolescent

information seeking. School Libraries Worldwide, 9(2), 47–61.

Byron, S., & Young, J. (2001). Information seeking in a virtual learning

environment. Research Strategies, 17, 257–267.

Centre for Information Behaviour and the Evaluation of Research (CIBER). (2008).

Information behaviour of the researcher of the future: A ciber briefing paper.

Retrieved from http://www.ucl.ac.uk/slais/research/ciber/downloads/. Accessed

on November 14, 2010.

Chen, S-H. L. (2003). Searching the online catalog and the World Wide Web. Journal

of Educational Media and Library Sciences, 41(1), 29–43.

Chung, J. S., & Neuman, D. (2007). High school students information seeking and

use for class projects. Journal of the American Society for Information Science and

Technology, 58(10), 1503–1517.

Dervin, B. (1999). Chaos, order, and sense-making: A proposed theory for

information design. In: R. Jacobson (Ed.), Information design (pp. 35–57).

Cambridge, MA: MIT Press.

Dervin, B., & Nilan, M. (1986). Information needs and uses. Annual Review of

Information Science and Technology, 21, 3–33.

Dresang, E. (2005). The information-seeking behaviour of youth in the digital

environment. Library Trends, 54(2), 187–196.

http://www.ucl.ac.uk/slais/research/ciber/downloads/

124 Leanne Bowler

Fallows, D. (2005). Search engine users: Internet searchers are confident, satisfied

and trusting –– but they are also unaware and naive. Pew Internet and American

Life Project. Retrieved from www.pewinternet.org/pdfs/PIP_Searchengine_

users.pdf

Fidel, R., Davies, R. K., Douglass, M. H., Holder, J. K., Hopkins, C. J., Kushner,

E. J., et al. (1999). A visit to the information mall: Web searching behaviour of

the high school students. Journal of the American Society for Information Science,

50(1), 24–37.

Fitzgerald, M. A. (1999). Evaluating information: An information literacy challenge.

School Library Media Research, 2. Retrieved on November 14, 2010, from http://

www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume2

1999/vol2fitzgerald.cfm#implications

Flavell, J. (1979). Metacognition and cognitive monitoring: A new area of cognitive-

developmental inquiry. American Psychologist, 34(10), 906–911.

Friel, L. D. L. (1995). The information research process with low-achieving freshman

using Kuhlthau’s six-stage model and the interventions that facilitate the process.

Unpublished doctoral dissertation. University of Massachusetts, Lowell, MA.

Genest, M., & Turk, D. C. (1981). Think-aloud approaches to cognitive assessment.

In: T. Merluzzi, C. Glass & M. Genest (Eds), Cognitive assessment (pp. 233–269).

New York, NY: New York University Press.

Gorrell, G., Eaglestone, B., Ford, N., Holdridge, P., & Madden, A. (2009). Towards

‘‘metacognitively aware’’ IR systems: An initial user study. Journal of

Documentation, 65(3), 446–449.

Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press.

Kracker, J. (2002). Research anxiety and students’ perceptions of research: An

experiment. Part I. Effect of teaching Kuhlthau’s ISP model. Journal of the

American Society for Information Science and Technology, 53(4), 282–294.

Kracker, J., & Wang, P. (2002). Research anxiety and students’ perceptions of

research: An experiment. Part II. Content analysis of their writings on two

experiences. Journal of the American Society for Information Science and

Technology, 53(4), 295–307.

Kuhlthau, C. C. (1985). Teaching the library research process: A step-by-step program

for secondary school students. West Nyack, NY: Center for Applied Research in

Education.

Kuhlthau, C. C. (1988). Longitudinal case studies of the information search process

of users in libraries. Library and Information Science Research, 10(3), 257–304.

Kuhlthau, C. C. (1989). The information search process of high-middle-low

achieving high school seniors. School Library Media Quarterly, 17(4), 224–228.

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. (2004). Seeking meaning: A process approach to library and

information services. Westport, CT: Libraries Unlimited.

Kuhlthau, C. C., Turock, B. J., George, M. W., & Belvin, R. J. (1990). Validating a

model of the search process: A comparison of academic, public and school library

users. Library and Information Science Research, 12, 5–31.

http://www.pewinternet.org/pdfs/PIP_Searchengine_users.pdf

http://www.pewinternet.org/pdfs/PIP_Searchengine_users.pdf

http://www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume21999/vol2fitzgerald.cfm#implications




Land, S., & Greene, B. (2000). Project-based learning with the World Wide Web: A

qualitative study of resource integration. Educational Technology, Research and

Development, 48(1), 45–68.

McGregor, J. (1994a). Cognitive processes and the use of information: A qualitative

study of higher-order thinking skills used in the research process by students in a

gifted program. In: C. C.Kuhlthau (Ed.),School librarymedia annual (pp. 124–133).

Englewood, CO: Libraries Unlimited.

McGregor, J. (1994b). Information seeking and use: Students’ thinking and their

mental models. Youth Services in Libraries, Fall, 69–76.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded

sourcebook. Thousand Oaks, CA: Sage Publications.

Moore, P. A. (1995). Information problem solving: A wider view of library skills.

Contemporary Educational Psychology, 20, 1–31.

Moore, P. A., & St. George, A. (1991). Children as information seekers: The

cognitive demands of books and library systems. School Library Media Quarterly,

161–168.

Neilsen, J. (2005). Usability of websites for teenagers. In Jakob Neilsen’s alert box,

January 31, 2005. Retrieved from http://www.useit.com/alertbox/teenagers.html.

Accessed on November 14, 2010.

Shamo, E. (2001). University students and the internet. Unpublished doctoral

dissertation. University of North Texas, Denton, TX.

Swain, D. E. (1996). Information search process model: How freshman begin

research. In S. Hardin (Ed.), ASIS96: Proceedings of the 59th ASIS annual

meeting, Baltimore, MD, October 21–24 (33, pp. 95–99). Medford, NJ:

Information Today, Inc.

Todd, R. (2003). Adolescents of the information age: Patterns of information seeking

and use, and implications for information professionals. School Libraries

Worldwide, 9(2), 27–46.

Valenza, J. K. (2006). They might be gurus. Teacher Librarian, 34(1), 18–26.

Watson, J. S. (1998). If you don’t have it, you can’t find it: A closer look at students’

perceptions of using technology. Journal of the American Society for Information

Science, 49(11), 1024–1036.

Williamson, K., McGregor, J., Archibald, A., & Sullivan, J. (2007). Information

seeking and use by secondary students: The link between good practice and the

avoidance of plagiarism. School Library Media Research, 10Retrieved on

November 14, 2010, from http://www.ala.org/ala/mgrps/divs/aasl/aaslpubsand-

journals/slmrb/slmrcontents/volume10/williamson_informationseeking.cfm

Wolf, S. (2007). Information literacy and self-regulation: A convergence of disciplines.

School Library Media Research, 10. Retrieved on November 14, 2010, from http://

www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume10/

wolf_informationliteracy.cfm

Wolf, S., Brush, T., & Saye, J. (2003). The big six information skills as a

metacognitive scaffold: A case study. School Library Media Research, 6. Retrieved

on November 14, 2010, from http://www.ala.org/ala/aasl/aaslpubsandjournals/

slmrb/slmrcontents/volume62003/bigsixinformation.htm

http://www.useit.com/alertbox/teenagers.html

http://www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume10/williamson_informationseeking.cfm

http://www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume10/williamson_informationseeking.cfm

http://www.ala.org/ala/mgrps/divs/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume10/wolf_informationliteracy.cfm



http://www.ala.org/ala/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume62003/bigsixinformation.htm

http://www.ala.org/ala/aasl/aaslpubsandjournals/slmrb/slmrcontents/volume62003/bigsixinformation.htm

Documents

[Library and Information Science] New Directions in Information Behaviour Volume 1 || Chapter 5 Into the Land of Adolescent Metacognitive Knowledge During the Information Search Process: