UBC Faculty Research and Publications

A Task-based Approach to Mobile Information Interactions Absar, Rafa; O'Brien, Heather, 1977- Oct 31, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Absar_Obrien_Task_Approach_Mobile.pdf [ 233.57kB ]
JSON: 52383-1.0107447.json
JSON-LD: 52383-1.0107447-ld.json
RDF/XML (Pretty): 52383-1.0107447-rdf.xml
RDF/JSON: 52383-1.0107447-rdf.json
Turtle: 52383-1.0107447-turtle.txt
N-Triples: 52383-1.0107447-rdf-ntriples.txt
Original Record: 52383-1.0107447-source.json
Full Text

Full Text

A Task-based Approach to Mobile Information Interactions Rafa Absar iSchool, University of British Columbia Vancouver, B.C. r.absar@ubc.ca Heather O'Brien iSchool, University of British Columbia Vancouver, B.C.   h.obrien@ubc.ca  ABSTRACT In this paper we examine current representations and classifications of mobile information interactions.  We propose a task-based approach to unify and guide future work of mobile information retrieval.    Keywords Mobile information retrieval (IR), user studies, diary study, task classification INTRODUCTION Between 2006 and 2013, ownership of desktop computers declined, while more Americans purchased portable devices including cell phones, laptop computers, e-Book readers, and tablets (Smith, 2012).   Recently, Pew Research found that 91% of survey respondents owned a cell phone. 65% of these individuals had smartphones and 55% were using their phones to access the Internet (Smith, 2012). This proliferation of mobile devices has implications for information seeking and retrieval.  Mobile devices provide constant connectivity for social purposes, and for locating and using information.  Despite the fact that mobile searching has been shown to be less sophisticated and less successful than desktop searching (Kamvar, Kellar, Patel & Xu, 2009; Sohn, Li, Griswold & Hollan, 2008), people are using mobile devices to perform complex information behaviors (i.e., browsing, downloading and sharing content, etc.) (Smith, 2012).  In fact, Nylander, Lundquist, Br?nnstr?m and Karlson (2009) found that people chose to access the Internet through their mobile devices, even while at home and within range of a computer.  A picture of mobile information behavior has been emerging within the last decade, primarily through log and diary studies of mobile use.  The strength of this literature is its description of mobile users? physical environments and social contexts, and the utilitarian and hedonic goals of mobile information seeking and use. However, attempts to classify mobile information interactions are not systematic, and often derived from the data generated by a particular study.  This makes comparisons across studies challenging and prevents the work from being theoretically grounded.  In this paper, we review how mobile information interactions have been classified and propose a task-based approach to understanding the nature of mobile information seeking and use.  We describe how we are applying the task-based approach in a mobile interview and diary study.  MOBILE INFORMATION INTERACTIONS Mobile information interactions include searching, browsing, encountering and checking (e.g. status-checking) behaviours. A number of studies have specifically the factors that affect search using log and diary studies.  Compared to desktop searchers, mobile users pose simpler queries and refine them less often (Church & Oliver, 2011; Kamvar et al., 2009); make shorter, less interactive visits to websites; view less content when visiting websites; (Nicholas, Clark, Rowlands & Jamali, 2013); and are less likely to follow up on or follow through with search results (Kamvar et al., 2009).  Mobile users also tend to use search engines only if there is not an appropriate app for their query (Church & Oliver, 2011; Tossell et al., 2012). Users exhibit different behaviors with mobiles than desktops due to device affordances, i.e., screen size, font, and menu design (Kim, Jacko, & Salvendy, 2011).  However, there are also contextual variables operating on the mobile user. Tojib and Tsarenko (2012) found that mobile activities are conducted to ?kill time? (e.g. during a commute) or under time constraints (e.g. subway arriving on the platform). Hence, physical context may lead mobile users to avoid complex information tasks. Further, Rahmati and Zhong (2010) reported that mobile devices are viewed as less obtrusive or anti-social compared to computers.  Users value mobile information interactions for the way they add to their other activities, allowing for ?lots of 1 minute internet interactions around real life? (p. 71, Church & Oliver, 2011). Thus, due to time and situational factors, ideal mobile interactions may be brief in nature, easily halted if need be, and effortlessly integrated into everyday life and social interactions.   CLASSIFICATIONS OF MOBILE IIR A number of studies have attempted to classify mobile information interactions according to the topic or domain of the search, the motivation for the information need, or the type of activity in which the mobile searcher was engaged. Domain-based classifications The most common way mobile information interactions have been described is by domain or topic.  Kamvar and Baluja (2006), who conducted one of the first large scale analyses of mobile search behavior, examined the logs of one million page requests from Google?s mobile search sites. They categorized queries into 23 categories, including entertainment, local services, lifestyle and community, food  Copyright held by the authors  Proceedings of HCIR 2013, Vancouver, BC, Canada, Oct 3-4, 2013  and drink, current events, etc., but did not elaborate on how these categories were selected. Later, Kamvar et al. (2009) carried out another logs-based comparison of search patterns across computers, iPhones, and conventional mobile phones with the Google search interface. They used the same categorization as the previous study, but introduced 7 additional categories, such as Animals, Photo, etc. In addition to log studies, diary studies have contributed topical categories based on participants? articulated information needs and queries. Some of these categories include business hours; blog/forum; directions; downloads; information site; phone number; news/weather; media; recipes; shopping, timetable/transit; transactions; travel; trivia; etc. (Church & Smyth, 2009; Komaki, Hara & Nishio, 2012; Nylander et al., 2009; Sohn et al., 2008).  Collectively, there is a lack of consistency in the categories used in domain-based approaches.  Compare, for example, ?news/weather? (Komaki et al., 2012), ?sports/news/stocks? (Sohn et al. (2008) and ?news and current events? (Kamvar et al., 2006; 2009).  Here news appears in each category, but is paired with different items.  Generally speaking, some categories appear in only one study while others co-occur, and the number of categories ranges widely. These issues make meta-analyses challenging.  Further, the scope of each classification scheme is problematic.  In some cases we see a type of activity, e.g. ?shopping,? or a domain, e.g. ?food and drink,? while in other cases the information object (e.g. ?recipe?) or format (e.g. ?media,? ?photo?) is referred to in the classification scheme.  B?hmer (2011) classified mobile information behavior based on the domain of the application being utilized, rather than the actual information need or query. Apps were classified in a similar manner to domain-based classifications, with categories such as Finance, Travel, Communication, Shopping, Entertainment, Games, etc.   Motivation or intent-based classifications Church and Smyth (2009) built upon Broder?s (2002) taxonomy of general web search that proposed web queries could be classified according to their navigational, informational, or transactional intent. Church and Smyth (2009) focused on mobile information needs and the impact of contextual variables, such as location and time, on these needs. They extended Broder?s taxonomy by defining informational, geographical, and personal information management needs for mobile information needs. Kim, Kim, Lee, Chae and Choi (2002) created two broad-based categories for mobile users? context of use: hedonic (for pleasure, to kill time) and utilitarian (to accomplish a specific task expediently). Taylor, Anicello, Somohano, Samuels, Whitaker and Ramey (2008) furthered this work in their study of mobile motivations and behaviors. Taylor et al. proposed six motivational factors: awareness (to keep informed and current); time management (to be timely and efficient); curiosity (to pursue an unfamiliar topic); diversion (to kill time); social connection (to socially interact with others); and social avoidance (to separate oneself from others).  They found that awareness was the most frequently observed motivation, while social avoidance was the least. Church and Oliver (2011) also used these six key motivators from Taylor et al. (2008) to analyze diary entries according to motivations for information needs.  They found that mobile users motivations were less goal-directed and urgent, and primarily leisure-based and social in nature. Activity-based classifications Taylor et al. (2008) also classified mobile users? behaviors (status checking, browsing, information gathering, fact checking, in-the-moment, planning, transaction and communication) based on Kellar, Watters and Inkpen?s (2007) interview study of web monitoring behavior.   They found that some behaviors tended to co-occur with specific motivations.  For example, awareness motivations were often paired with status checking.  Other motivations, such as curiosity and diversion, were satisfied with a range of behaviors. Overall, they noted that mobile behaviors were short, opportunistic, and occurred alongside other activities. In summary, mobile information interactions have been examined from multiple perspectives, including motivations and needs, activities, and topic or domain.  Few studies have combined these approaches.  As noted previously, the domain-based approach is particularly problematic in terms of the number, scope, and labeling of categories.  It would be beneficial to take a more systematic approach to the classification of mobile information needs, seeking, and use in order to advance research in this area.  One way in which to do so would be to apply a task-based approach.   A TASK-BASED APPROACH TO MOBILE INFORMATION INTERACTION Vakkari (2003) defined a task as ?an activity to be performed in order to accomplish a goal.? Bystr?m and Hansen (2005) emphasized the dynamic nature of tasks and the physical, affective, and/or cognitive aspects of executing a task. They distinguished work tasks, information seeking tasks (embedded in information-intensive work tasks) and search/information retrieval tasks. Mobile search tasks, which have received little attention to date, represent an emerging area of research, but can be grounded in previous task research more generally.  Hert and Marchionini (1998) developed a query-based taxonomy, based on statistical websites, identifying three dimensions of task: pragmatic, semantic, and syntactic.  The pragmatic dimension refers to the context or situation in which the task is being completed.  It involves goal attributes (learning, verifying, evaluating, etc.); constraints, (time, amount of information needed, etc.); and system (fit between the search tool and task, characteristics of the access point, etc.).  The semantic dimension is topical, and pertains to the level of abstraction, specification, and facetedness of the information need.  Lastly, the syntactic dimension includes the expression of the query, goal type, (closed versus open, etc.), and the users? interaction with the system in a given situation.  Li and Belkin (2008) suggest that the above three dimensions may not be sufficient to cover all aspects of tasks, and extended this further using a faceted approach. Task facets were defined as different aspects or properties of tasks, such as task source, performer, time, product, process and goal. Li and Belkin?s (2008) scheme provided a more fine-grained classification of the characteristics of search tasks. The contribution of prior research in information retrieval is that it takes into account multiple dimensions of tasks, rather than looking only at the domain, activity or motivation. With a wealth of task research in information retrieval to draw upon, we selected a task-based taxonomy to study mobile information interactions.  Our selection was based upon the results of mobile user experience studies.  For example, we noted that users are more likely to use apps than search engines (Church & Oliver, 2011; Tossell et al., 2012), and that mobile searchers goals may be more leisure-based (Kim et al., 2002; Taylor et al., 2008).  Thus query-based and goal-oriented approaches to mobile information interactions may not be appropriate.  We admired the work of Li and Belkin (2008) for its emphasis on the task, the task performer, and the outcome, and for its rigorous summation of task approaches in information retrieval.  However, we reasoned that mobile information search and retrieval is a relatively new area of research and, as such, task-based approaches may need to be more fluid at this stage.  For this reason, we adopted the framework used by Toms, O?Brien, Mackenzie, Jordan, Freund, Toze, Dawe, and MacNutt (2008), which was based on Li and Belkin?s work, but was more flexible for exploratory research.   ? The term motivation used by Toms et al. (2008) was derived from Li and Belkin?s  (2008) source of task, defined as being motivated by the self, collaboration or external factors.  We elect to use the term motivation based on prior work by Taylor et al. (2008) that has proven valuable for capturing motivations unique to mobile contexts.   ? The requirements and constraints category encompassed Li and Belkin?s task doer and time facets, but is broader in scope.  We adopted Toms et al.?s (2008) category to account for the physical, geographical, and temporal aspects of mobile search. ? Li and Belkin (2008) describe process as a task that occurs once or over multiple sessions.  We broaden this category to include the step-by-step task procedure of carrying out a task. ? Li and Belkin (2008) refer to the quality or quantity of goal attributes. Toms et al.?s (2008) description of common search task goals of information gathering, fact checking or decision making are more relevant in the mobile context and take into account the activity based approaches described in the previous section. ? Li and Belkin (2008) do not include task domain and topic as a facet, arguing that it is not possible to assign limited values to topic. However, since previous mobile research has used topical classifications, we propose to include it. A finite list of broad topical categories may be generated from previous research and from our own data. ? We use the term outcome (Toms et al., 2008) instead of product (Li & Belkin, 2008), for its fit with mobile interactions, which may be undertaken for the experience rather than a tangible end product.   We are employing a version of Li and Belkin?s task taxonomy, modified by Toms et al. (2008), which we feel is well-suited to studying mobile information interactions.  Having identified our approach, we provide operational definitions of these task categories below based on prior mobile studies that we are applying in current research on mobile user engagement with search.    Motivation is the impetus to undertake a task. We propose Taylor et al.?s (2008) six motivational factors as values for this facet: awareness, time management, curiosity, diversion, social connection and social avoidance.  In the mobile environment, motivations may be more experiential in nature and not necessarily be for the purposes of satisfying an information need; information search may be a by-product of a social or contextual situation. Requirements and constraints are the conditions or limitations on the task. In the mobile environment, the temporal, locational and physical context imposes constraints on mobile users and search tasks (Rahmati and Zhong, 2010; Teevan et al., 2011; Tojib and Tsarenko, 2012) and is considered here. Context is important in mobile use and has not been looked at independent of other factors in previous studies (for example, with motivation).  Goal is the purpose or aim of the task. These include hedonic or utilitarian goals (Kim et al., 2002) and depend on the type of activity mobile searchers engage in to satisfy their information needs, such as: information gathering, fact checking, or decision making (Toms et al., 2008). Domain and topic constitute the subject matter of the search task. This considers domain-based approaches, such as News, Health, Directions, Social Media, etc. (Kamvar et al., 2009; Sohn et al., 2008; etc.).  However, we noted challenges inherent in how mobile studies have classified domain and, as such, we will look to domain-based classifications to guide topical categories.  Process is the sequenced set of actions and events involved in the task. This includes how the search was undertaken, what steps were involved, and what app or search engine was used to conduct the mobile search. Outcome: the results of the task performance. This includes whether the information need was satisfied, as well as experiential factors, e.g. engagement, fun.  We are applying this task-based approach to an interview and diary study with nineteen mobile users. Participants completed an initial interview, diary entries, and a second  interview over a one-week period. Diary entries were created in situ and consisted of text descriptions of search tasks and photos of local contexts. For our purposes photos are particularly salient for understanding requirements and constraints. In the second interview, we explored a portion of the diary entries in more depth by randomly selecting five events and asking participants to select one notable event from each set. We are qualitatively analyzing the diary entries and interview data using the pre-defined facets of motivation, requirements and constraints, goal, etc.  In the presentation of this short paper, we will focus on the 242 diary entries and explore aspects of requirements and constraints, goal, domain, and outcome in greater detail.     CONCLUSION Although task-based approaches are a staple in traditional information retrieval, they have yet to be applied in mobile information retrieval. We propose that applying a task-based approach to mobile information interactions will provide a more holistic view of, as well as establish a much-needed unified approach to, mobile search tasks.  REFERENCES B?hmer, M., Hecht, B., Sch?ning, J., Kr?ger, A., & Bauer, G. (2011). Falling asleep with Angry Birds, Facebook and Kindle. In Proceedings of the International Conference on Human Computer Interaction with Mobile Devices and Services (pp. 47-56).  Broder, A. (2002). A taxonomy of web search. ACM SIGIR Forum, 36(2), pp. 3-10. Bystr?m, K., & Hansen, P. (2005). Conceptual framework for tasks in information studies.  JASIST, 56(10), pp. 1050-1061. Church, K., & Oliver, N. (2011). Understanding mobile web and mobile search use in today's dynamic mobile landscape. In Proceedings of the International Conference on Human Computer Interaction with Mobile Devices and Services (pp. 67-76).  Church, K., & Smyth, B. (2009). Understanding the intent behind mobile information needs. In Proceedings of the 14th international conference on Intelligent user interfaces (pp. 247-256). Kamvar, M. & Baluja, S. (2006). A large scale study of wireless search behavior: Google mobile search. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 701-709).  Kamvar, M., Kellar, M., Patel, R. & Xu, Y. (2009). Computers and iPhones and mobile phones, oh my!: A logs-based comparison of search users on different devices.? In Proceedings of the 18th International Conference on World Wide Web (pp. 801-810).  Kellar, M., Watters, C., & Inkpen, K.M. (2007). An exploration of web-based monitoring: implications for design. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 377-386).  Komaki, D., Hara, T., & Nishio, S. (2012). How does mobile context affect people's web search behavior? In Proceedings of the Advanced Information Networking and Applications Conference (pp. 245-252). Kim, H., Kim, J., Lee, Y., Chae, M., & Choi, Y. (2002). An empirical study of the use contexts and usability problems in mobile internet. In Proceedings of Hawaii International Conference on System Sciences (pp. 1-10). Kim, K., Jacko, J., & Salvendy, G. (2011).  Menu design for computers and cell phones: Review and reappraisal.  INT J HUM-COMPUT INT, 27(4), 383-404. Li, Y. & Belkin, N. J. (2008). A faceted approach to conceptualizing tasks in information seeking. IPM, 44(6), 1822-1837. Nicholas, D., Clark, D., Rowlands, I., & Jamali, H. R. Information on the go: A case study of Europeana mobile users. JASIST, 64(7), 1309-1310. Nylander, S., Lundquist, T., Br?nnstr?m, A., & Karlson, B. (2009). ?It?s just easier with the phone??A diary study of internet access from cell phones. PERVASIVE COMPUT, 5538, 354-371. Rahmati, A., & Zhong, L. (2012). Studying Smartphone Usage: Lessons from a Four-Month Field Study. IEEE TRANS MOBILE COMPUT, 31 May 2012.  Smith, A. (2012). The best (and worst) of mobile connectivity. Pew Internet & American Life Project, Retrieved from http://pewinternet.org/Reports/2012/Best-Worst-Mobile/Key-Findings.aspx.   Sohn, T., Li, K.A., Griswold, W.G., & Hollan, J.D. (2008). A diary study of mobile information needs. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 433-442).  Taylor, C. A., Anicello, O., Somohano, S., Samuels, N., Whitaker, L., & Ramey, J.A. (2008). A framework for understanding mobile internet motivations and behaviors. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 2679-2684).  Teevan, J., Karlson, A., Amini, S., Brush, B., & Krumm, J. Understanding the importance of location, time, and people in mobile local search behavior. In Proceedings of the 13th International Conference on Human Computer Interaction with Mobile Devices and Services MobileHCI '11, pp. 77-80. New York: ACM. Toms, E.G., O?Brien, H., Mackenzie, T., Jordan, C., Freund, L., Toze, S., Dawe, E. and MacNutt, A. (2008). Task Effects on Interactive Search: The Query Factor. N. Fuhr et al. (Eds.): INEX 2007, LNCS 4862, 359?372. Tojib, D., & Tsarenko, Y. (2012).  Post-adoption modeling of advanced mobile service use.  J BUSI RES, 65(7), 922-928. Vakkari, P. (2003). Task-based information searching. ARIST, 13(1), 413-464.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items