 Greetings, everyone, and welcome to this SLIS colloquium presentation. I'm Dr. Bill Fisher, the coordinator for our colloquium series for this year, and it's my pleasure to introduce our speaker for today, Dr. Virginia Tucker, who will be talking about the expert searcher and threshold concepts. Dr. Tucker is a full-time lecturer with SLIS and has been teaching for us for the last few years, and has done a phenomenal job in that regard, and in 2011 was recognized as our outstanding lecturer with an award presented at the end of that academic year. Prior to working for us, she does have some roots in the Bay Area. She worked at Stanford for a while and was head of their physics library at one time, and left Stanford to go to work for what was in a fairly new company called Dialogue, which was one of the two major online searching providers for the information world at that time. Dialogue was bought out a number of years ago by a company called ProQuest, which may be how many of you have heard of Dialogue, and so Dr. Tucker's interest in expert searching got started when she was working for Dialogue, and in some of the teaching that she's done for us, because she does teach our online searching course. This interest got further reinforced when she started the San Jose State Queensland University of Technology Gateway PhD program, and this in fact was her PhD topic that she will be discussing with us today, and of course, since she now has the DR in front of her name, she graduated last year and is one of our first graduates from the program. So Dr. Tucker, let me turn things over to you. Great. Thank you very much, and hello everyone. Thank you for being here for this colloquium. This is Virginia Tucker, and today I'm very excited to talk to you about my research into transformative learning experiences and how these shape the acquisition of search expertise. We'll be looking specifically at critical concepts called threshold concepts, more on that in a minute, involved as search expertise is acquired. I'll talk about what it means to be a search expert in today's information environment, and what expertise is in a 2.0 world. We'll discuss how the role of the expert's searcher is evolving too, and then look at the implications of this research for theoretical models of the search experience, search interface design, and how we teach advanced search methods and concepts. In the current online environment, expert's searchers engage with information as independent information brokers, researchers, reference librarians, information architects, faculty who teach advanced search, and in a variety of other information intensive professions. Their experiences are characterized by a profound understanding of information concepts and search techniques, and they also have an agile ability to apply this knowledge to interacting with and having an impact on the information environment. In addition, they play a role in the information experiences of others. The focus of my research has been to explore the learning experiences of searchers in order to better understand how search expertise is acquired. But before I move on, here's a sketch of what we'll cover. I'll describe the design of the study, the participants, and the research methodology used. I will cover the research context, briefly describing the literature that informed the study. This will introduce the whole idea of threshold concepts, a relatively new framework, and how this can guide our way of thinking about transformative learning experiences. I'll speak about these concepts as representing learning portals. Most of the talk I will spend on the next parts, the research results, the findings, and their implications. So let's begin at the beginning and consider the information environment in which we live and work and search as library and information professionals. There's a belief that everyone can search to good effect in today's information environment, and there's certainly much truth to this. You simply sit down at the keyboard or perhaps you crawl to it and start keying in words. The kind of searching my study explored, however, is the type that demands far more than quote good enough googling unquote to use the words of Plosker. Indeed, as the intelligence of search engines continues to increase, a valid question to ask is what does the expert searcher bring to the search interaction and to the search outcome and results? We might also ask what makes a searcher an expert searcher? So exploring this was examined in my research as well as how search expertise is acquired. I decided to explore conceptual knowledge involved in acquiring search expertise, looking at critical concepts that could transcend two things. First, the particulars of an individual search engine, and second, the subject domain. And I focused on the liminality between the highly proficient novice and the search expert. The term liminality in this context refers to the conceptual space and the learning experiences through which a person passes the transformed perception of a given subject. Some people describe these in simpler terms as aha moments or times when the light clicks on. I designed the research plan so that insights were gained from searchers on both sides of this liminal or conceptual space and from those who may be actively journeying through it and in a liminal state. So first I'll describe the research design briefly. The methodology I used was grounded theory and I've included artwork here from NC Escher as I believe it graphically represents the iterative nature of the method. And also the iterative patterns in most successful searches. I'm not going to go in detail here today into grounded theory. Let's look first at the participants. The study included 20 participants drawn from two groups. There were 11 professional searchers with an average of 32 years of experience. This included LIS faculty who teach online searching, information brokers, and search engine developers. There were also nine participants who were MLIS students. Now these students were ones who had completed coursework in information retrieval and in online searching and who had demonstrated exceptional ability. They were pre-selected by their different instructors. So using these different participants made it possible to collect data from those who search at a very high level as well as those who may be actively developing expertise. The study collected data through semi-structured interviews and search tasks. Each participant session consisted of an interview followed by searches during which the participant narrated his or her thoughts called the think-aloud protocol. After the tasks, I continued the interview with additional questions and talk after time. There were two search tasks, both of which were designed to require no specific subject domain knowledge. Again, I intentionally minimized this so the focus could be on conceptual knowledge that did not call for any domain specific understanding or vocabulary. Two searches were performed. The first one was a known item search where it's easy to know when you're done. Typically, it requires little more than key words or natural language terms. The second search was a subject search, harder to know for this one when you're done. More likely also to benefit from more advanced techniques such as the use of controlled vocabulary, more browsing, and more iteration. The participant's narrative, the think-aloud, captured the decisions they were making during the search and the considerations and thoughts that occurred to them. After the searches, I had more interview questions and also revisited some of the earlier questions. The searches provided two things. First, it was a window into the participant's search experience as it was going on in real time. And second, it helped in eliciting memories of their learning to search experiences because they were actively engaged in the process. This was particularly evident in the interview that followed the search time. This study was situated in multiple areas of existing research literature. And I'm going to mention just a few. First, we had a search expertise itself that comprised three main components. There's the professional searcher who's been studied rather extensively, as well as web searching behaviors, also studied rather extensively. Then there's the nature of expertise itself, which draws from a broad backdrop of novice expert research. Then there's the area of research into learning experiences. Lots of learning theory out there and lots of research. But the main theoretical framework that I looked into was threshold concepts. And I'm going to come back to this theory and explain it more fully in just a moment. So the novice expert literature functioned as a kind of backdrop, as I mentioned. These are some of the key areas I looked at. I was looking into how experts structure knowledge. And they do this much better than novices do for performance purposes. Experts process ambiguous information much more effectively. Novices can be misled by ambiguity. When experts are solving problems, they are able to encode information and also to use relevant representations of the information. And experts learn in different ways. One thing they do is use reflection in their learning processes. So let's go on and talk about threshold concept theory, which is just fascinating. And it was first described by Eric Meyer and Ray Land in 2003. So it's actually a fairly recent theoretical framework. They wrote that a threshold concept is a core concept that once understood transforms perception of a given subject. A threshold concept has five characteristics. And there are some other dimensions having to do with discourse and some other areas have been put forth. But these are still the main five characteristics. First, it is transformative, causing a major shift in perception of the subject. It is also irreversible, unlikely to be unlearned. Next is integrative, meaning that it exposes some kind of connectedness that was not previously understood. And troublesome, this is an interesting one, meaning that learning the concept is initially uncomfortable, counterintuitive, and usually there's some wrestling involved in trying to learn it. The bounded characteristic is not as often present as the other four. This refers to a concept that has frontiers or boundaries bordering on other thresholds into other conceptual areas. So I'm going to discuss each of these characteristics in a little bit more detail so that it's very clear what they constitute. The transformative characteristic is kind of a life will never be the same moment, a little bit like getting your driver's license. It involves a major shift in attitude, values, or understanding. And it's accompanied by profound ontological shift. Gwyneth Cousin describes the shift this way. New understandings are assimilated into our biography, becoming part of who we are, how we see, and how we feel. Irreversible means, no matter what happens, you cannot unlearn it. It's like riding a bike is a bit of a simplistic way to describe this characteristic. So in one example, in economics research, market equilibrium is considered to be an irreversible threshold concept. The integrative characteristic involves exposing something previously hidden or where the connectedness was not understood. And this means that separate ideas become unified in a person's understanding and the connections are made. Now, this one is often the characteristic that's easiest for us to recall when we look back on key learning experiences that involved a threshold concept. It means that it must be wrestled with in order to be grasped. It can be very counterintuitive or uncomfortable. And this discomfort may be due to some preconceptions that need to be shed, things that are incorrect. One example in the research is some novices learning to solve physics problems. I started out as a physics major, so I could relate to this one. And they suffered from some misconceptions and contradictions. And in wrestling with these, they were then able to reach new levels of understanding. The researchers in this study referred to them as confluences. As mentioned, the bounded characteristic is less common than the other four. It represents a kind of not-in-Kansas anymore threshold that has terminal frontiers or boundaries, and they may border other thresholds for other conceptual areas. For example, specialized terminology may take on new meaning in the subject area. So, as you can see, the bar is set rather high for a concept to be considered a threshold concept. And there are also many nuances to identifying threshold concepts. As an example, researchers have studied the academic concept of the learning experience as a factor in exploring evidence of threshold concepts. In addition, the aspect of epistemological perspectives needs to be fully acknowledged. Gwyneth Cousin provides this illustration. A Keynesian economist and a Marxist one may propose different threshold concepts for the economics they respectively teach, because they have quite different views about what is central to their subject. So, having taken this closer look at the characteristics of threshold concepts, let's put it all together and look at an example most of us can relate to. Here's our elderly gentleman, again, on his bike and a toddler having his first transformative experience on one. I've highlighted three of the characteristics that are prominent in the experience of learning to ride a bike. This is not to say, however, that my research established riding a bike as a threshold concept. Initially, the experience of learning about it can be troublesome. That one jumps out quite easily. There's typically a lot about learning to ride a bike that involves balance and gravity and often some troublesome skinned elbows and knees. Definitely trouble here and there. But it's also transformative. When you transition from being a walker to being a cyclor, your world becomes a bigger place. It is simply transformed. And it's irreversible. Once you've got it, you've always got it. No matter how old you get, again, the old adage about learning to ride a bike really does hold up. However, and this is a big however, it's important to underscore that in my research, I was looking at the threshold between the proficient novice and the expert, specifically how the research can add to our understanding of acquiring expertise in searching. So, I'd like to pause for a moment here. Having described the research design, the data collected and the context in which the study was situated. And this leads us on now to look at the results. In the next section, I'll cover the categories in the coding and analysis of the data rather briefly. And then I'll talk about the thematic results, which were the basis of the theoretical model that emerged for search expertise. So, in speaking of grounded theory, this quote refers to how using straightforward categories can shine with bright meanings through our analytical renderings, kind of a nutshell version of how grounded theory can be applied. And these are the categories that ultimately represented the data. There was one for broad view, which had to do with the big picture of the search experience. There was one for subject domain, which was when the participants described an aspect of searching that was tied to subject knowledge, such as medical or legal information, where subject knowledge is critical. Third category had to do with the nature of learning. This had to do specifically with learning itself in a broad way, not even necessarily tied to searching. The fourth one had to do with qualities and approaches of the expert searcher. And this gathered in a lot of data, which I'll cover in more depth in just a moment. Not so much having to do with the concepts, again, but with qualities and approaches. Then there was a category for tools and search knowledge. This included multiple code clusters, I termed them, for search commands, skills, and techniques. Things like knowing how to use proximity operators fluidly. We'll see in a minute how some of these categories morphed and shifted into somatic results as I looked at the conceptual knowledge represented in the tools being used. And last, there was a category for work-related experiences. This was a small group that also evolved during the analysis process of examining these experiences to see what it was about them that had impacted the participant's learning experience. Before speaking about the themes from the research, I want to say a few words about differentiating between a theme and a category. This touches on the nature of research itself. In some studies, these words are used almost interchangeably. But I believe that differentiating between categories and themes is a key effort when working with data, and it's definitely part of the grounded theory approach that I used. This musical analogy from Janice Morshed's Light on the Difference. A category is a collection of similar data sorted into the same place. A theme, on the other hand, is a meaningful essence that runs through the data. Just as the theme and opera occurs over and over again, sometimes in the foreground, sometimes in the background, and sometimes co-occurring with other tunes. So does the theme in our research. So let's look at the themes next with that in mind. The somatic results fell into three areas. First was critical concepts that both novice and highly experienced searchers determined were essential in their learning-to-search experiences. The second theme I called praxis, because this included the approaches, qualities, strategies, tactics, and grain practices and tools used regularly by the expert's searcher. Sometimes these were being used during the search tasks. Sometimes these were reported in the interviews as being important. And third, there were the traits, attitudes, and qualities of the expert's searcher. I think I said qualities in the second group, but I did not mean to. The final stage of analysis utilized the lens of threshold concept theory, and further articulation of the themes was possible. What I asked was the question, which of these themes demonstrated the characteristics of threshold concepts? With this in mind, let's move on to some of the key findings from the research. First up are the threshold concepts that were identified. The data provided evidence of four threshold concepts for search expertise. The first was information environment. This being a profound discernment of the total information setting and the ability to apply this understanding to the search experience. Now, if this threshold concept emerged, I realized that an apt and vivid metaphor to explicate it would be to build on the well-known berry picking model of information-seeking and search behavior of Marcia Bates. An essential part of the nature of berry picking is that searchers adapt the strategy to their particular need at the moment. For the expert searcher, this would mean extending the model to explain that she understands how the berries came to grow in the bush, why they grew where they did, where there might be clusters of berries hidden away under foliage, and even who planted the bush, tended it, amended the soil, and how this impacted its growth and harvest. This profound level of knowledge is integrative and impacts the searcher's decisions and activities during a search. I will share a quote from a participant that exemplifies this concept. This is from a professional searcher who does intellectual property research and described an expert as being taller, someone who can see and notice things others can't. He said, so if let's say there's someone who invented something, they filed a patent, they did extensive search, they don't find anything, then the patent office, uses the same sources, a lot of people use exactly the same sources, they're trained by the same people, they don't find anything either, and they allow the patent. So, there's this short piece of prior art out there, that's a fact, but no one has found it. Maybe that's okay. In the same way, the dog just walked down the trail and didn't even know that the deer was there. Maybe that's okay. It doesn't mean that a good searcher, of course, is a little bit taller searcher, might not find it. The next one is information structures. This is where ideas such as database structures, document structures, index structures, and underlying retrieval algorithms are understood and applied to the search. Understanding structures is critical in the profession of information, as it commonly is in other professions, and it is present at different structural levels. Document structures, for example, could include the components within an individual item, record, or object within a database, such as fields, segments, subfields, metadata, XML markup, or other tagging. There might also be weighting of these fields, or segments, or other value added features applied by indexers or by automated processing. Understanding how to manipulate these in order to optimize search results is at the core of this threshold concept. One of the novice student searchers in the study was describing database structures and told of this experience. Most of my aha moments are where I gain an understanding of the behind-the-scenes part of the search. I guess it's that I gain more clarity on how to formulate it. Another novice said this. It gave me an idea of really how databases are structured, things about indexing in particular, understanding that an index is not necessarily something that you can see on the surface. You have to know how a database is put together so you can search it properly. The next one is information vocabularies. This encompassed several things. It's a fluency in search activities related to language, including natural language, keywords, and controlled vocabulary, such as descriptors and subject codes, as well as finesse using proximity operators, truncation, and other language-based tools. One of the highly experienced searchers who teaches in an MLIS program said this. Learning how to use synonyms. That was a real aha moment for me in dealing with students. I thought they would automatically know that they should be searching on synonyms. Well, guess what? They don't. Then moving on to the fourth threshold concept, which I have called concept fusion. This is the ability to integrate or fuse the other three threshold concepts, and it is further defined by three properties. The first property is visioning or anticipating next moves. The second property are called dancing or being light on one's search feet. And third, a profound ontological shift, deeply identifying that the concept of search professional, not just I search. Now identity shift, as mentioned earlier, accompanies the grasping of a threshold concept, but here I emphasize that there was a qualitatively deeper change. Cousin describes the experience of someone learning to speak French, who shifts an identity from, I can speak French, to being a frank file. There's a parallel here. Here's a quote from a novice participant. I think it's kind of like the more expert searcher looks up and sees how deep the hole is, like, hmm, maybe I can try digging somewhere else, and then just come back to this one later if I can't find anything. It's actually a weird but instinct kind of thing. It's like developing patterns in your brain for what you've seen in the past and applying it to what you see now, but also being aware that anything you try could work, so you've got to try anything. Definitely willingness to step out of your paradigm, but also be aware of the pattern, sort of this weird dual process. And this from one of the highly experienced searchers. I think that's when you become an expert in anything. You get to the point of where you're playing music. You aren't looking at the music. It just comes to you. I'm not a musician, so maybe that's not the best analogy, but it gets to the point where you don't have to think about the mechanics anymore. I play golf now, so I guess that's a good analogy. When you play golf, you have to be conscious and think about where your arms are at the top of the swing and the angle of your spine and all these other details. But when you're swinging in a golf club, you can't actually think of all that stuff in the time that you are actually swinging. If you practice enough, it becomes automatic. That's true for searching, too. It eventually got to the point where I was good enough at understanding the structure of the databases, enough that, and the whole concept of if you're looking for something, they might be close to each other. So I understood the fundamentals, so I could focus on modifying my search in real time to get closer to the thing that you want. An illustration of the profound ontological shift is that it's often comprehended through the recognition of others. One of the most experienced searchers in the study said this, I began to be invited to be a speaker at these conferences, then I began to realize that if I talk about this, I must be an expert. Now, in addition to the findings about threshold concepts, the data were informative about the nature of search experts. I set out to explore learning experiences, and in describing these experiences, the participants contributed their perspectives on attitudes, practices, and characteristics of expert searchers. This adds to existing research, such as studies by Marcia Bates and Raya Fidel. So let's look at some of these practices and traits. First, some of their approaches include collaboration, emphasis on reference interview, and planning ahead. The strategy, sometimes called search styles, included pro-growing, one-mowing, or successive fractions, and lots of iterating and refining. And some of the tactics are listed here as well, such as analyzing, evaluating, prioritizing. The traits, attitudes, and qualities are listed here. Some of the most dominant ones included an attitude of adventuring and exploring, enjoying the hunt, perseverance, confidence, extreme curiosity, open-mindedness, patience. And another one here, seemingly balancing perseverance, was knowing when to stop. Considering all of this together, I put forth an integrated model of search expertise. At the center is Concept Fusion, the compounded threshold concept that integrates the three threshold concepts of information environment, information structures, and information vocabularies. Surrounding this are the traits, practices, qualities, attitudes, and the like. These strengthen the model and also build on existing research literature regarding characteristics. So now we move on to consider some of the implications of this research. First, for LIS professionals and practitioners, generally, certainly all LIS professionals search, and very often for others, and in the instruction of others in some setting. And this way of looking at learning experiences can deepen our understanding of conceptual knowledge that transcends the search engine. The research also helps update our understanding of the traits and practices of expert searchers in today's information environment. This brings into the present day the research going back into the early 1980s. And it also provides a theoretical model for further research into understanding the fusion of concepts that define or are involved in search expertise. As we look at implications for LIS educators and researchers, it makes possible different ways to think about exploring threshold concepts for non-LIS and non-expert searchers. I've recently written an extended article with three of my colleagues that will be out in Jealous the 1st of the year, which sets out a research agenda to consider these and other questions. Other areas to explore for threshold concepts are what concepts might be present in learning portals for other areas within LIS? Are there learning portals in collection management, for example, or information literacy? Quite interestingly, ACRL has appointed a task force to revise its information literacy standards, and it was announced in October that two new elements will be incorporated, threshold concepts, and meta-literacy. It also raises the question, can we determine better ways to teach advanced search to LIS professionals? I've been working actively with these concepts in the courses I teach and hope to report in the future on those efforts. Then also, what does this tell us about the implications for search interface design? It seems the pendulum for search interface design has reached an apex of the, quote, done-down interface, and search engines have been incorporating more and more advanced search features into the basic search mode. The exalide search engine is one example where in-context help leads the searcher directly into advanced features and for an experience with more finely tuned control over the results. Another example, researchers at Google have been studying search behaviors when search becomes more difficult. Another indication that interface design, interface designers are exploring more challenging search scenarios. So an understanding of threshold concepts in searching could provide direction to the development of new kinds of interfaces for more expert-like searching. So that concludes my talk. Thank you very much, Dr. Tucker. That was great. As Dr. Tucker indicated, she does have an article coming out based on this research that should be out sometime next year, and the journal is, she gave the acronym Jealous, which is the Journal for Education in Library and Information Science, and we do have that as part of the San Jose State collection. If anybody is interested in that, and it should be out at some point in early 2014, Jealous I think is a quarterly, so it will either be the first or second issue in that regard. Let me thank Dr. Tucker. Thank you all for being here. I wish everybody a happy Thanksgiving later this week, and for those of you who may be traveling over the holiday, I hope you, everybody gets to where they want to go safely and with as little adventure as possible. Thank you very much.