In the era of mobile Internet,many users can access the Internet via their mobile terminals. With the increasing number and variety of personal IT devices, users often search and use information across multiple devices, such as smart phones, tablet PCs, e-readers, PCs, televisions, wearable devices, etc. In such case, the unique personal multiscreen ecosystem has been constructed. Users are no longer confined to only one device, but will constantly switch and transfer among their various devices. Cross screen behavior(CSB) is becoming a frontier research field in human-information behavior (HIB) and attracts more and more attention from both academia and industry. The existing research of CSB involves the exploration of user characteristics, the motivations, and the related implementation technologies. However, the current research stream is very fragmented, and lacks the systematic focus and integrative perspective on the underlying mechanism of CSB.

    
As an exploratory study, this paper aims at uncovering the black box of CSB from a macro view. A qualitative approach is conducted by grounded theory method to build an integrated framework of user's cross screen behavior in mobile internet environment with two layers (i.e., external performance and internal motivation) and four dimensions (i.e., user, task, information, technology, UTIT). The study adopts Strauss three-stage analysis with a combination of in-depth interviews and a focus group. The data collection work was launched from June 2016 to October 2016. In order to ensure the consistency of data collection and analysis, we keep three principles: 1)All sample subjects are interviewed, recorded, transcribed and re-read by the same researcher. 2) Two researchers use NVivo to code and write memos independently. 3) Weekly research team sessions debate the process of the interviews and extracted codes, achieving the opening and convergence of opinions. Through the open coding, the axial coding, the selective encoding, and theoretical saturation test, we propose the UTIT framework for CSB including four dimensions and two layers. We also find four types of associations among the factors identified in the coding process, namely manifestation, causality, mediation and moderation. In an ensemble view of CSB, the UTIT framework covers 9 main categories of the ability characteristics, emotional perception, task context, task-driven force, information scent demand, media type, information organization and presentation, technology fit and technology application, involving 39 categories and 117 initial concepts.
The UTIT framework breaks the limitation of the dichotomy and segregation on motivation and behavioral performance in traditional research. It integrates the cross-screeners performance and motivation into a single analytical framework, and presents a holistic view for the CSB research. Four types of research propositions and directions of CSB are revealed from the UTIT framework, including: 1)The characteristics and experience of daily CSB. With the task context as the starting point and the “container”, the daily CSB characteristics and experience can be sorted out and show a rich and ecological performance trajectory. 2) The influence factors of CSB. This proposition belongs to the motivation level research, and needs to be discussed in the combination with user, task, information and technology. Meanwhile, through the integration of Self-efficacy Theory, Technology Acceptance Model, Task-Technology Fit Model, Information Scent Theory, Information Literacy Theory and other related theories, the integrated model of CSB can be built to explore the all-round CSB motivation and mechanism. 3) The mediating effect of information technology on CSB. Specifically, how the task-driven force, information organization and presentation can be together to seek the information technology fit, thus affecting the mechanism of CSB is worth further excavating. 4) The moderating effect of user capability characteristics on task-driven force and CSB. The influence of task-driven force on CSB will be stronger or weaker with the characteristics of user's ability, and the user's ability will adjust the relationship between task-driven force and CSB in a positive or negative direction. To this end, the CSB can be more in-depth interpreted by analyzing the moderating effect of user's capability characteristics. In sum, the UTIT framework becomes a panoramic map of CSB, and provides some implications for the future theoretical and empirical investigations. 2 figs. 5 tabs. 27 refs.