Search task features in work tasks of varying types and complexity

Information searching in practice seldom is an end in itself. In work, work task (WT) performance forms the context, which information searching should serve. Therefore information retrieval (IR) systems development/evaluation should take the WT context into account. The present paper analyzes how WT features: task complexity and task types, affect information searching in authentic work: the types of information needs, search processes and search media. We collected data on 22 information professionals in authentic work situations in three organization types: city administration, universities and companies. The data comprise 286 WTs and 420 search tasks (STs). The data include transaction logs, video recordings, daily questionnaires, interviews and observation. The data were analyzed quantitatively. Even if the participants used a range of search media, most STs were simple throughout the data, and up to 42 % of WTs did not include searching. WT’s effects on STs are not straightforward: different WT types react differently to WT complexity. Due to the simplicity of authentic searching, the WT/ST types in interactive IR experiments should be reconsidered. Introduction Much of modern work is information-intensive and supported by diverse information systems. Information supply and searching are key activities in information-intensive work. Work tasks (WTs) are the building blocks of work. A job description includes abstract WTs that manifest as WT actions in daily work (Byström & Hansen, 2005). In working life, WT performance forms the context, which information supply and searching should serve. In order to design information search systems to properly serve WT requirements, it is necessary to study how the WTs as actions are connected to searching. Therefore information (retrieval) systems development and evaluation should not take place in isolation but take the work context into account and find out for what purposes and how the systems are used. Failing to do this may result in developing suboptimal systems for expected but biased search needs. In the present paper we aim to find out, how searching is carried out in work contexts and what affects it. Several researchers highlight the importance of task-based information searching (TBIS) as a research area (Järvelin et al., 2015; Kelly, Arguello, & Capra, 2013; Vakkari, 2003). However, naturalistic field studies of TBIS are rare. Other types of empirical approaches to TBIS seem more common, including log analyses (e.g., Kotov, Bennett, White, Dumais, & Teevan, 2011), simulated WTs (e.g., Borlund, 2003), self-report methods (e.g., Li, 2009), and seminaturalistic studies (e.g., Pharo, 2004). The present study belongs to the naturalistic field studies, where information searching is studied in the context of authentic WTs, at the work place, collecting multiple types of data in real-time. There are some predecessors to the present study, e.g., Hansen’s (2011) study on TBIS in the patent domain; Saastamoinen, Kumpulainen, and Järvelin’s (2012) study in the public administration domain; and Kumpulainen’s (2014) study in Molecular Medicine. Also Kellar, Watters, and Shepherd (2006) and Kelly (2006) used naturalistic methods but their work was more concerned with leisure time activities than WTs. We consider major search task (ST) features that have received little attention in past research, under the influence of WTs. First, information searching is typically studied in a limited context, where the searcher has only one given search tool to use. In reality, there are several options the searcher can choose from. Second, in experiments, the searcher cannot choose to use more than one search tool though this is often possible in real-life. Third, information needs are typically studied narrowly: Either the need is almost directly given to searchers (in the case of simulated WTs), or the searchers are asked to describe their needs (in self-report studies), or researchers attempt to reconstruct information