Understanding adaptive information seeking in the context of microblogging from the cognitive switching perspective

Abstract

Drawing on adaptive structuration theory and cognitive switching theory, this study develops a research model exploring the effects of cognitive switching stimuli on adaptive information seeking and the moderating effects of information need and personal innovativeness in information technologies. Data collected from microblogging users were used to test the model. The findings suggest that other people’s use, discrepancies, and deliberate initiatives each have significant positive effects on trying new features to seek information. Other people’s use which essentially reflects the nature of learning from observing other people is the most important determinant. Meanwhile, information need and personal innovativeness in IT each positively moderate the effect of other people’s use on trying new features to seek information. This study contributes to theory by examining adaptive information seeking in the context of microblogging which has been largely overlooked by prior literature. The findings and more implications for theory and practice are discussed.

Keywords

Adaptive information seeking adaptive structuration theory cognitive switching theory information need personal innovativeness in IT microblogging

Introduction

Microblogging as a term was initially used in the early 2000s, then becoming more popular with the introduction of Twitter which continues to be the most widely used platform by far (Fransen-Taylor and Narayan, 2018; Williams et al., 2013). Microblogging builds on the ideological and technological foundations of Web 2.0, having attracted millions of active users (Kaplan and Haenlein, 2010). Characterized by content creation and information intermediaries, microblogging has become an important tool and channel for information seeking (Kim, 2014; Magoi et al., 2019). In China, the term Weibo is used as a synonym for microblogging and Sina Weibo is the most popular microblogging platform. In the current study, we focus on the context of Weibo in China.

The rapid development of Weibo is constantly creating new social structures within Weibo, resulting in higher requirements for and bigger challenges to Weibo use. Adaptive structuration theory (AST) which provides a framework for understanding how people integrate information technologies (IT) into their work practice described the interplay between the social structures contained within IT and the social structures which emerge in IT use (DeSanctis and Poole, 1994). Structuration and appropriation are the core ideas of AST. Following AST, prior research has extensively examined diversified behaviors related to adaptive IT use, such as feature use and feature extension (Jasperson et al., 2005), appropriation (Fuller and Dennis, 2009), innovativeness with IT (Wang et al., 2013), extended use of enterprise systems (Peng et al., 2018), workarounds of a global enterprise system (Malaurent and Karanasios, 2019). However, prior research mainly examined adaptive IT use at the levels of technology, system or feature. To the best of our knowledge, the existing literature contains few studies examining adaptive IT use at the level of information which we suggest could help further understand deeper social structures of IT use. To fill this gap, the current study examines adaptive information seeking in the context of Weibo, which has been largely overlooked by prior literature.

Information seeking refers to “the purposive seeking for information as a consequence of a need to satisfy some goal” (Wilson, 2000: 49). Cognitive style was proved to be related to information-seeking behavior (Ford et al., 2002). With the development of the Web and social media, information seeking has long been a focus of research. Huang et al. (2007) examined Web users’ online information-seeking behavior. It was found that width, length and depth of online information-seeking behavior are highly correlated. Spezi (2016) examined information seeking of doctoral students, suggesting that this follows a steady trend, with some subtle changes, particularly in the use of social media. Fang and Xie (2019) examined Chinese everyday life information seeking in terms of a comparison between the Web and Weibo. It was found that much information could be obtained through the two channels. Wellings and Casselden (2019) examined information seeking of engineers and scientists. It was suggested that online search engines, specialist databases and scholarly search engines were the most popular resources used by both engineers and scientists. Regarding the effectiveness of information seeking, it was suggested that the adoption of IT does not always result in effective information seeking (Xu et al., 2010). Consequently, adaptive information seeking becomes critical. Adaptive information seeking which is developed from adaptive IT use and information seeking, is the combination of adaptive IT use and information seeking and subsequently refers to adaptive use of IT with information seeking as the purpose.

Active thinking and conscious cognitive processing were suggested to be critical conditions for adaptive IT use behaviors (Jasperson et al., 2005). In general, individuals’ cognitive processing involves automatic and conscious cognitive modes. Informed by prior research which focused on the dynamics within a certain cognitive mode, the cognitive switching theory examined movement from automatic to conscious cognition, identifying three kinds of cognitive switching stimuli, namely, novelty, discrepancy and deliberate initiative (Louis and Sutton, 1991). Drawing on AST and cognitive switching theory, the current study explores the effects of cognitive switching stimuli on adaptive information seeking. Moreover, information need reflects a cognitive gap which motivates individuals to seek information to satisfy their information need (Lu and Yuan, 2011; Marchionini, 1995). Innovativeness which was described as willingness to change is much related to cognitive style (Hurt et al., 1977; Sun, 2012). So, the current study further explores the moderating effects of information need and innovativeness on the impacts of cognitive switching stimuli on adaptive information seeking.

Adaptive IT use is exactly captured by appropriation which is used to describe how users actively choose to use the structures within IT (DeSanctis and Poole, 1994). Similarly, adaptive information seeking can be described as how users actively choose to use the structures within IT for information seeking. Adaptive information seeking becomes critical for people to satisfy their information need given higher requirements and bigger challenges of Weibo use. We suggest this study presents a new lens for adaptive IT use research and practice alike. Following this introduction, the theoretical background is introduced. Specifically, we pay attention to AST and cognitive switching theory. Meanwhile, the behavior related to adaptive IT use and the main findings of prior research are reviewed. Then, we develop the research model and hypotheses. After that, we describe the method and data collection. Finally, the results of the research are presented, followed by the discussion and implications.

Theoretical background and hypotheses

Cognitive switching theory

People’s cognitive processing generally involves automatic and conscious cognitive modes, each of which could function well in different social environments. Sometimes automatic cognitive activities with “habits of mind” are sufficient, but sometimes situations call for conscious cognitive activities with active thinking (Louis and Sutton, 1991). Informed by prior research which focused on the dynamics within a certain cognitive mode, Louis and Sutton focused on movement or connection between modes. Automatic and conscious cognitive activities were suggested to form a cycle in terms of automatic mode, sensing condition for switching, switching to conscious cognitive mode, conscious mode, sensing condition for switching, switching to automatic cognitive mode. Louis and Sutton (1991) did not discuss the second switch from conscious to automatic but focused on the first switch especially switching cognitive gears from automatic mode to conscious mode. Three kinds of situations were identified and discussed that would stimulate people to switch from automatic to conscious cognition, namely, novelty, discrepancy and deliberate initiative. Novelty refers to the situations when people experience events that are extraordinary, unique, previously unknown or unfamiliar; discrepancy refers to the situations when there is an unexpected failure, or there are setbacks and difficulties, or there is a significant difference between reality and expectations; deliberate initiative refers to the conditions when people choose to try something new in response to internal request for active thinking, or when people are asked to think actively or explicitly questioned by external request (Louis and Sutton, 1991).

Adaptive structuration theory

Giddens (1979, 1984) proposed structuration theory (ST) in the 1970s and 1980s. Then, DeSanctis and Poole (1994) proposed AST based on ST. After that, Stones (2005) proposed strong structuration theory (SST). Drawing on the experience in IT research, Pozzobon and Pinnsonneault (2005) improved the application of ST in empirical work. Jack and Kholeif (2007) answered critics’ questions about ST, and further emphasized that ST and its later derivations can be used in substantive empirical research rather than as analytical tools. Broady-Preston (2009) indicated that ST, AST and SST are formal social theories, which concerned the evolution of organizations and groups.

In order to study the role of IT in organization change, DeSanctis and Poole (1994) proposed AST in the context of Group Decision Support System (GDSS). AST provides a framework for understanding how people integrate IT into their work practice and describing the interplay between IT, interpersonal interaction and social structures. On the one hand, IT contains social structures, on the other hand, the use of IT also contains social structures. In a recursive manner, two kinds of structures promote the formation of each other (DeSanctis and Poole, 1994). Huang et al. (2019) drew on AST to investigate the social structure of ResearchGate (RG), suggesting that the social structure emerging in using RG differed from that embedded in RG.

Structuration and appropriation are the core ideas of AST. Structuration is considered as a social process of interaction between human and social structures, and the current structures are the result of previous activities (Orlikowski, 1992). Appropriation is used to describe how users actively choose to use the structures within IT. They will choose a variety of ways to adapt to the structural characteristics, such as direct appropriation, substitution, combination, enlargement, contrast, constraint, affirmation and negation (DeSanctis and Poole, 1994). AST uses two ways to describe social structures provided by IT: “the structural features of the given technology and the spirit of this feature set” (DeSanctis and Poole, 1994: 126). Structural features are used to refer to the specific types of resources, capabilities, or rules; spirit refers to the values and goals underlying the structural features set (DeSanctis and Poole, 1994).

Adaptive IT use

DeSanctis and Poole (1994) suggested that AST can be applied to many kinds of other IT contexts and other organizational or social contexts. Following AST, many studies were carried out by researchers on the adaptive use of IT. Jasperson et al. (2005) examined post-adoptive IT use behaviors after recognizing that the functional potential of installed IT applications in organizations is underutilized. They proposed a conceptual model of post-adoptive behavior which was defined as the myriad feature adoption decisions, feature use behaviors and feature extension behaviors. It was suggested that reflective cognitive processing in the individual cognition model may modify individuals’ (already existent) post-adoptive intentions, which then lead to future post-adoptive behaviors. They also direct future research on factors influencing users to (continuously) exploit and extend the feature built into IT applications. Fuller and Dennis (2009) built on the Fit Appropriation Model to examine how fit and appropriation influence performance over time. It was found that fit could predict team performance soon after technology adoption. However, over a short time period, for the teams using better fitting technology, their performance remained constant. But for the teams using poor-fitting technology, they innovated and adapted, thus improving performance. Schmitz et al. (2016) examined the complexity of malleable IT use by developing a theoretical perspective of adaptation behaviors including exploitive technology adaptation, exploratory technology adaptation, exploitive task adaptation and exploratory task adaptation. The results revealed the compounding effects of four distinct adaptation behaviors, suggesting that the effect of technology adaptation on individual performance is mediated by task adaptation. Peng et al. (2018) examined employees’ extended use of enterprise systems. It was found that system self-efficacy and modularity have positive effects on extended use. Meanwhile, leader-member exchange positively moderates the effects of system self-efficacy and modularity on extended use. System modularity positively moderates the effect of system self-efficacy on extended use. Lee et al. (2018) examined how users adapt to the use of email to handle conflict. Four key conflict-triggered adaptation strategies were identified, namely, interaction avoidance, blame-protection, disempowering and image-sheltering. Barrett (2018) examined technological appropriations and health information technology workarounds. It was found that appropriating electronic health record technology in a way other than it is designed to be used across an organization results in enhanced perceptions of the technology and its implementation. Peng and Guo (2019) examined antecedents of employees’ exploration of enterprise systems. It was found that both system modularity and task variety have direct effects on employees’ system exploration. Meanwhile, task variety positively moderates the effect of system modularity on system exploration. Malaurent and Karanasios (2019) examined how a global enterprise system led to the emergence of unexpected practices (workarounds) based on a four-year longitudinal case study. It was found that users collectively construct and implement workaround practices rather than simply accept or reject an information system.

Prior research mainly examined adaptive IT use behavior at the level of technology, system or feature. There are few studies which examined adaptive IT use behavior at the level of information, such as innovative information seeking (Zha et al., 2019a) and adaptive information sharing (Zha et al., 2019b). To the best of our knowledge, no research examined adaptive information seeking in the context of Weibo. In the current study, we examine adaptive information seeking in the context of Weibo, which we suggest could help further understand deeper social structures of Weibo.

Research model and hypotheses development

Weibo functions for their users not only as social settings but also as “information neighborhoods” (Burnett, 2000; Pang, 2018). Information seeking is nonlinear, dynamic and flowing (Foster, 2004). Due to the electronic information revolution, people’s information-seeking behavior is constantly changing (Ellis and Oldman, 2005). With the rapid development of Weibo, new features are constantly creating new social structures within Weibo. According to AST, it is reasonable to suggest that when users seek information in Weibo, they interact with the social structures within Weibo. In other words, users interact with the structural features of Weibo and the spirit of these features (DeSanctis and Poole, 1994). Adaptive IT use examined by prior research covers various behaviors at the level of technology, system or feature. Following the research by Sun (2012) who examined trying new features as one type of adaptive system use behavior, we examine trying new features to seek information as one type of adaptive information-seeking behavior. Specifically, we examine the effects of cognitive switching stimuli on trying new features to seek information and the moderating effects of information need and personal innovativeness in IT in the context of Weibo. AST and cognitive switching theory provide sound theoretical support for the current study. The research model is presented in Figure 1.

Figure 1.

Research model.

Trying new features refers to adding new features to one’s features in use and thus expanding the scope of the features in use (Sun, 2012). In the current study, trying new features to seek information is defined as trying new features in Weibo to purposively seek information as a consequence of a need to satisfy some goal (Sun, 2012; Wilson, 2000). We suggest that trying new features to seek information is essentially adaptive IT use at the information level. Adaptive IT use which is exactly reflected by the core idea “appropriation” in AST is used to describe how users actively choose to use the structures within IT and how they choose a variety of ways to adapt to the structural characteristics. Active thinking and reflective cognitive processing were found to be critical conditions for adaptive IT use behaviors (Jasperson et al., 2005). Following cognitive switching theory which identified three kinds of situations that would stimulate people to switch from automatic to active and conscious cognition (Louis and Sutton, 1991), we examine the effects of three kinds of stimuli on trying new features to seek information.

Observations and perceptions are equally important in active and conscious cognitive processing. If people are looking elsewhere, then the novel situations may remain unseen and there will be no perceptions on it (Starbuck and Milliken, 1988). In the current study, other people’s use is defined as users’ observation of other people’s use of new features in Weibo (Sun, 2012). Other people’s use is related to observational learning which suggests that people can learn some complex behavior simply by observing the behavior of others and their results (Bandura, 1986). For example, observational learning plays a significant role in determining users’ willingness to self-disclose information on social network sites (Ashuri et al., 2018). In the context of Weibo, different users often develop distinct patterns of Weibo use, namely, people use different Weibo features to seek information for different information needs and tasks (Burton-Jones and Gallivan, 2007). Through observing such different uses of Weibo features, individuals may learn from each other. Specifically, users might learn from power users (Boudreau and Robey, 2005) and sound information seekers (Zha et al., 2015). For a particular user, observations of different Weibo feature use by other people will be perceived as new by her/him. In other words, compared to previous use of Weibo, when a user observed other users were using some features they did not use before, she/he would perceive a novel situation. Indeed, observations of other people’s use of new features may usefully serve as novel situations for the user (Sun, 2012). Consequently, other people’s use reflects the novel condition identified by cognitive switching theory that stimulates individuals to switch from automatic to conscious cognition.

Prior research examined how individuals’ adaptive IT use was impacted by learning from other people. Boudreau and Robey (2005) took a new ERP system as an example to explain why and how users’ enactments of information technologies change from inertia (initially choosing to avoid using it) to reinvention (working around system constraints in unintended ways). It was found that improvised learning accounted for changing. Users’ learning process were motivated by influences from other people such as project leaders who force them to use, power users who provide unofficial trainers for them and peers who encourage them to use. Ryu et al. (2005) found that members can achieve specialized knowledge about their own tasks through learning-from-others in an enterprise information portal (EIP) environment. Sun (2012) suggested other people’s use, new task and changes in system environments serve as three dimensions of novel situations which were found to have significant and positive influence on users’ adaptive system use behaviors in the context of Microsoft Office. In the context of Weibo, it is reasonable to suggest that if users observe others’ use of new features in Weibo, they would be likely to try these features to seek information. Thus, we hypothesize:

H1. Other people’s use has a positive effect on trying new features to seek information.

In general, discrepancies refer to situations where a disruption, an unexpected failure or a significant difference exists between expectations and the reality (Armstrong and Hardgrave, 2007). Researchers were interested in assessing discrepancies between expected and received technology performance in the information systems context (Chin et al., 2014). Discrepancies appear when an individual’s habit of mind does not match the existing cognitive schemas (Wong and Weiner, 1981). Discrepancies between expectations and the reality can prompt individuals to switch from habit of mind to active thinking, and then change their behaviors to adapt to the cognitive schemas in a specific environment (Hastie, 1984; Louis and Sutton, 1991; Wong and Weiner, 1981).

Prior research investigated the influence of discrepancies and some similar constructs on individual’s adaptive IT use behaviors. Bhattacherjee and Premkumar (2004: 23) defined disconfirmation as “the extent to which subjects’ pre-usage expectation of technology usage is contravened during actual usage experience”. They focused on an online retailing site and found that disconfirmation drove the change of users’ beliefs and attitudes about the use of IT. Jasperson et al. (2005) suggested that work system outcome expectation gap refers to the differences of work system outcomes between desired and perceived. To resolve the gap, individuals need to apply unused features, to use the already-used features at a higher level, or to try new features into IT application. Sun (2012: 461) defined discrepancies as situations where “the outcomes of system use are different from what were expected”, and found discrepancies have a significant and positive influence on adaptive system use. In the current study, discrepancies are defined as situations where the outcomes of Weibo use are different from what were expected (Sun, 2012). It is reasonable to suggest that if users encounter discrepancies during information seeking in Weibo, they would be likely to try new features to seek information to cope with the discrepancies. Thus, we hypothesize:

H2. Discrepancies have a positive effect on trying new features to seek information.

Generally, deliberate initiatives refer to “the initiatives one takes in response to a request for an increased level of attention, when asked to think, or while being explicitly questioned” (Sun, 2012: 459). Individuals may change their cognitive modes and behaviors in response to other people’s requests (Louis and Sutton, 1991; Sun, 2012). Schön (1983) argued that when individuals are confronted with demands, they may perform active thinking and behavior. Langer (1989) also made a similar argument that, when individuals are explicitly questioned, they may become more mindful.

Deliberate initiatives are suggested to be similar to mandatory use (Sun, 2012). Prior research argued that individuals’ IT use can be explicitly questioned. In other words, individuals may use IT in a mandatory manner (Brown et al., 2002; Venkatesh and Davis, 2000). In the mandatory use environment, individuals are often required to use new systems or new features in a system (Hartwick and 4, 1994; Hsieh et al., 2012; Sun, 2012). Guo and Zhang (2010) pointed out that mandatory IT adoption is very common in China. They suggested that in the mandatory adoption context, users will apparently adopt the new system and use it frequently. Sun (2012) suggested that deliberate initiatives have a positive effect on discrepancies which further impacts adaptive system use. In the current study, deliberate initiatives are defined as situations where Weibo users are asked to use certain features in Weibo (Sun, 2012). It is reasonable to suggest that if users are asked to use certain features in Weibo, they would be likely to obey the requests and try to use new features in Weibo to seek information. Thus, we hypothesize:

H3. Deliberate initiatives have a positive effect on trying new features to seek information.

Following Lu and Yuan (2011), information need is defined as the gap between individuals’ current information about the task in their study/research and an information sufficiency threshold where they feel that they have already obtained adequate information and thus have no further need for information. Information need is much related to individuals’ cognitive processing and information seeking. Indeed, information need reflects a cognitive gap which prevents individuals from making sense of a particular task. In order to fill the cognitive gap, they are likely to try many ways to seek information to satisfy their information need (Lu and Yuan, 2011; Marchionini, 1995).

Information need is an important motivation variable and can potentially stimulate individuals to process information and cognition carefully (Zha et al., 2016). Zha et al. (2016) examined the positive moderating effect of information need in the basic processes underlying the effectiveness of persuasion to use digital libraries for getting information, finding that information quality of digital libraries (central route) has a greater effect on information usefulness for individuals with high information need. In the context of Weibo, observing other people’s use of new features may serve as novel situations for users (Sun, 2012).When Weibo users have a high information need, they would carefully process information conveyed by novel situations and learn more from observing other people’s use like the use of power users (Boudreau and Robey, 2005) and the use of sound information seekers (Zha et al., 2015). It was found that when individuals have a high information need, they are more likely to be influenced by others and then perform risk information-seeking behavior (Huurne and Gutteling, 2008). It is thus reasonable to suggest that other people’s use has a greater effect on trying new features to seek information for users with high information need. In the context of Weibo, discrepancies are hard to avoid since the features in Weibo do not always function as expected (Abel et al., 2011). When Weibo users have a high information need, they would carefully process information conveyed by discrepancies and reflect more deeply on the difference between the outcomes of Weibo use and what were expected. It is thus reasonable to suggest that discrepancies have a greater effect on trying new features to seek information for users with high information need. We thus make the hypotheses below:

H4. The effect of other people’s use on trying new features to seek information will be positively moderated by information need such that this effect will be stronger for individuals with high information need.

H5. The effect of discrepancies on trying new features to seek information will be positively moderated by information need such that this effect will be stronger for individuals with high information need.

Zha et al. (2016) examined the negative moderating effect of information need in the basic processes underlying the effectiveness of persuasion to use digital libraries for getting information, finding that reputation of digital libraries (peripheral route) has a greater effect on information usefulness for individuals with low information need. In the context of Weibo, the nature of deliberate initiatives mean that users are asked and instructed to use certain Weibo features, which is similar with mandatory use in IT and information systems research (Sun, 2012). In the mandatory use environment, users are likely to have negative emotion and pressure caused by controlling behaviors (Ryan and Grolnick, 1986; Venkatesh et al., 2003). In the current study, when users have high information need, they have more motivation to carefully process information when asked to use certain Weibo features by others. In this situation, the persuasion effect by others’ asking and questioning is likely to become weaker. To the contrary, when users have low information need, they have less motivation to carefully process information when asked to use certain Weibo features by others. In this situation, the persuasion effect by others’ asking and questioning is likely to become stronger. It is thus reasonable to suggest that information need has a negative moderating effect on the influence of deliberate initiatives on trying new features to seek information. We thus make the hypothesis below:

H6. The effect of deliberate initiatives on trying new features to seek information will be negatively moderated by information need such that this effect will be weaker for individuals with high information need.

Innovativeness is described as willingness to change (Hurt et al., 1977). Personal innovativeness or similar concepts such as cognitive style (a person’s preferred way of gathering, processing and evaluating information) has long been the focus of innovation research (Sun, 2012). Personal innovativeness in IT (PIIT) is defined as “the willingness of an individual to try out any new information technology” (Agarwal and Prasad, 1998: 206). PIIT reflects personal innovativeness in the specific domain of IT and one’s disposition to engage in innovative behaviors (Agarwal and Karahanna, 2000; Agarwal and Prasad, 1999). Adaptive IT use which describes how users actively choose to use the structures within IT is by nature innovative, making PIIT a closely related internal contextual factor (Sun, 2012).

Prior research has examined the positive moderating effect of PIIT in different IT context. Agarwal and Prasad (1998) examined intentions to use a new IT in the context of the World Wide Web, finding that PIIT positively moderates the impact of compatibility on intentions. Sun (2012) examined adaptive system use in the context of Microsoft Office, finding that PIIT positively moderates the impact of novel situations on adaptive system use. Li et al. (2013) examined innovative use in the context of business intelligence systems, finding that PIIT positively moderated the impacts of intrinsic motivation toward accomplishment, intrinsic motivation to know and intrinsic motivation to experience stimulation on innovative use. In the context of Weibo, for users with high PIIT, they are likely to be more sensitive to cognitive switching stimuli given that a stimulus does not necessarily stand out as a stimulus unless individuals could recognize it (Louis and Sutton, 1991). Innovative users who have willingness to change are more likely to receive new information or ideas that are needed for innovative behaviors (Sun, 2012). Consequently, other people’s use and discrepancies are more salient for Weibo users with high PIIT than for Weibo users with low PIIT. Users with high PIIT are more likely to sense other people’s use and discrepancies and subsequently engage in adaptive information-seeking behaviors. It is thus reasonable to suggest that other people’s use and discrepancies each have a greater effect on trying new features to seek information for users with high PIIT. We thus make the hypotheses below:

H7. The effect of other people’s use on trying new features to seek information will be positively moderated by PIIT such that this effect will be stronger for individuals with high PIIT.

H8. The effect of discrepancies on trying new features to seek information will be positively moderated by PIIT such that this effect will be stronger for individuals with high PIIT.

Sun (2012) examined the negative moderating effect of PIIT in the context of Microsoft Office, finding that deliberate initiatives have a weaker effect on adaptive system use for individuals with high PIIT. In the context of Weibo, deliberate initiatives are essentially similar with mandatory use of Weibo. Users with high PIIT are less responsive to demands from others with adaptive IT use behaviors than users with low PIIT, demonstrating more confidence in their capabilities to use new IT (Sun, 2012). Indeed, innovative people who are often characterized by high autonomy, flexibility and self-confidence are less likely to follow the orders or behavioral guidance and standards (Feist and Gorman 1998). They often pursue autonomy and resist controlling behaviors as to some extent captured by deliberate initiatives. Consequently, controlling behaviors cause people to feel pressure which might be detrimental for innovative people (Feist and Gorman, 1998; Oldham and Cummings, 1996). It is thus reasonable to suggest that PIIT has a negative moderating effect on the influence of deliberate initiatives on trying new features to seek information. We thus make the hypothesis below:

H9. The effect of deliberate initiatives on trying new features to seek information will be negatively moderated by PIIT such that this effect will be weaker for individuals with high PIIT.

Measures development and data collection

Measures development

To ensure content validity, all the constructs and their corresponding measure items were adapted from existing literature to fit the context of this study (Straub et al., 2004). Specifically, the items measuring other people’s use, discrepancies, deliberate initiatives and trying new features to seek information were adapted from Sun (2012); the items measuring personal innovativeness in IT were adapted from Agarwal and Karahanna (2000); the items measuring information need were adapted from Huurne and Gutteling (2008) as well as Zha et al. (2016). Twenty-two graduate students were invited to take part in a pilot survey. According to their feedback and comments, wordings were adjusted in several items to improve readability and clarity. The complete instrument is presented in Appendix A. All the items were measured with a 7-point disagree-agree Likert scale (1 represents “strongly disagree” while 7 represents “strongly agree”).

Data collection

The large-scale survey data collection was conducted in one comprehensive university located in central China. In the survey questionnaire, adaptive information behaviors were first described as various information activities such as information seeking and information sharing by using more features of IT or changing feature goals in innovative ways based on information need. Also, the development of Weibo was briefly introduced with Sina Weibo listed as the most popular Weibo platform in China. It was illustrated that Weibo is one type of important social media, which have become important tools and channels for people to seek, share and exchange information.

This study targeted undergraduate and graduate students of this university. After the questionnaire was published online, potential respondents were invited to visit the online questionnaire where the purpose of our study was explained, and their participation solicited. First, some classes were selected and emails were sent to potential respondents of these classes who were invited to visit the online questionnaire by clicking the link. Second, some classes were selected and approached, where potential respondents were invited to use their smart phones to scan the two-dimension code of the online questionnaire. Third, potential respondents who entered the university library were invited to scan the two-dimension code of the online questionnaire. Fourth, invitations were sent to potential respondents through other ways such as certain QQ groups and Baidu Post Bar of this university. Data collection was undertaken on a voluntary basis and lasted for about six weeks. One question regarding whether respondents were presently studying in this university was used to delete the responses not coming from this university. Consequently, data collected from 420 respondents were used for data analysis. Table 1 documents the demographic information of these 420 respondents.

Table 1.

Demographic information of survey respondents.

Category	Items	Frequency	%
Gender	Male	183	43.57
Gender	Female	237	56.43
Age	<18	6	1.43
	18‒25	335	84.52
	26‒35	52	12.38
	36‒45	6	1.43
	>46	1	0.24
Position	Undergraduate	273	65.00
	Master’s student	92	21.9
	Doctoral student	55	13.10
Field	Natural sciences	91	21.67
	Social sciences	228	54.29
	Arts and humanities	55	13.10
	Others	46	10.95
Experience with mobile Internet (years)	<1	4	0.95
	1‒2	16	3.81
	2‒3	35	8.33
	3-4	51	12.14
	>4	314	74.76
Experience with Weibo (years)	<1	105	25.00
	1‒2	80	19.05
	2‒3	81	19.29
	3‒4	59	14.05
	>4	95	22.62

Data analysis and results

In the current study, we employed partial least squares (PLS) structural equation modeling to verify our measurement and research models. Unlike covariance-based structural equation modeling methods such as LISREL and AMOS, PLS algorithm is a component-based structural equation modeling technique, “allowing each indicator to vary in how much it contributes to the composite score of the latent variable,” thus being “preferable to other techniques” (Chin et al., 2003: 197). PLS is in essence exploratory (Gefen et al., 2011) and can accommodate the exploratory nature and the presence of moderating factors of a research model (Goodhue et al., 2012; Ringle et al., 2012). In this sense, PLS is appropriate for this research given the exploratory nature of the research model and the moderating effects of information need and personal innovativeness in IT. Specifically, we employed SmartPLS 2.0 (Ringle et al., 2005) to verify the measurement and research models.

Measurement model validation

Generally, before testing the hypothesized relationships, the measurement validity needs to be assessed in terms of content validity, convergent validity and discriminant validity (Straub et al., 2004). With respect to content validity, given that all constructs and items in the current study were based on the previous literature, and the wordings were adjusted and improved after the pilot survey, we believe that all the constructs and items each have clear and correct meaning.

Table 2 shows the average variance extracted (AVE), composite reliability (CR) and Cronbach’s Alpha. The values of AVE, CR, Cronbach’s Alpha can be used to assess convergent validity and reliability (Straub et al., 2004). As shown in Table 3, the smallest value of CR is 0.909 and the smallest value of Cronbach’s Alpha is 0.830, both exceeding the recommendation of 0.7. The smallest value of AVE is 0.769, exceeding the recommendation of 0.5. So, all the constructs have a high degree of convergent validity and reliability (Straub et al., 2004).

Table 2.

Overview of measurement model.

Constructs	Items	AVE	CR	Cronbach’s Alpha
Deliberate initiatives (DI)	2	0.854	0.921	0.830
Discrepancies (DISC)	2	0.860	0.925	0.838
Personal innovativeness in IT (PIIT)	3	0.792	0.920	0.869
Information need (INEED)	3	0.900	0.964	0.945
Other people’s use (OPU)	3	0.769	0.909	0.849
Trying new features to seek information (TRYSEEK)	4	0.809	0.944	0.921

Table 3.

Correlations between constructs and square roots of AVE.

	DI	DISC	PIIT	INEED	OPU	TRYSEEK
DI	0.924
DISC	0.164	0.927
PIIT	0.181	0.201	0.890
INEED	0.027	0.237	0.256	0.949
OPU	0.357	0.300	0.266	0.149	0.877
TRYSEEK	0.305	0.273	0.371	0.220	0.465	0.899

Note: Diagonal elements are the square root of each construct’s AVE.

Table 3 shows the correlations between constructs and square roots of AVE. Since the square root of each construct’s AVE (bold values) is far greater than the correlations between constructs, sufficient discriminant validity can be established (Straub et al., 2004).

From Table 4, it can be seen that each item loads (bold values) much higher on its assigned construct than on other constructs, which further suggests sufficient convergent and discriminant validity for all the constructs (Straub et al., 2004).

Table 4.

Loadings and cross-loadings.

	DI	DISC	PIIT	INEED	OPU	TRYSEEK
DI1	0.938	0.172	0.206	0.037	0.374	0.305
DI2	0.909	0.127	0.121	0.011	0.277	0.254
DISC1	0.150	0.934	0.175	0.208	0.271	0.263
DISC2	0.154	0.921	0.199	0.233	0.287	0.242
PIIT1	0.111	0.188	0.906	0.272	0.230	0.324
PIIT2	0.269	0.159	0.884	0.155	0.295	0.354
PIIT3	0.090	0.192	0.880	0.266	0.176	0.309
INEED1	0.035	0.205	0.250	0.947	0.128	0.207
INEED2	0.027	0.215	0.224	0.958	0.130	0.184
INEED3	0.016	0.251	0.252	0.942	0.161	0.231
OPU1	0.265	0.268	0.229	0.163	0.802	0.347
OPU2	0.311	0.268	0.202	0.106	0.918	0.411
OPU3	0.355	0.258	0.267	0.128	0.906	0.455
TRYSEEK1	0.241	0.278	0.349	0.233	0.358	0.902
TRYSEEK2	0.240	0.231	0.357	0.222	0.443	0.928
TRYSEEK3	0.297	0.264	0.321	0.221	0.411	0.894
TRYSEEK4	0.315	0.210	0.308	0.121	0.454	0.872

Common method bias

In the current study, the data collected are self-reported, which may lead to a possibility that common method bias (CMB) arises. Harmon’s single-factor test is the most commonly used approach for assessing common method variance (CMV) (Podsakoff et al., 2003). Thus, we performed Harmon’s single-factor test by conducting a principal components analysis in SPSS. The factor solution resulted in six factors with eigenvalues greater than 1.0, accounting for 83.279% of variance. Also, the first factor accounted for 19.054 of the variance, which suggests that this factor does not account for the majority of the variance (Podsakoff et al., 2003).

Furthermore, following Liang et al. (2007) and Podsakoff et al. (2003), a latent method factor is included in the PLS model which is reflected by all the constructs’ items in the research model. Then, we examined the coefficients of each indicator’s two incoming paths from its substantive construct and the method factor. The results are shown in Appendix B. It can be seen that most method path coefficients are not significant. Meanwhile, the path coefficients of substantive constructs are apparently greater than their method path coefficients, explaining apparently greater variance of items than the method factor. Thus, it is reasonable to contend that CMB is not a concern in the current study.

Structural model with results

The structural model with results is presented in Figure 2 where p is based on two-tailed t value. According to the recommendation that the sample size should be larger than 500 (Wetzels et al., 2009), we conducted the bootstrap resampling procedure with 1000 samples to assess the path significances. The explained variance of trying new features to seek information is 0.369, showing a good predictive validity of the model (Straub et al., 2004).

Figure 2.

Research model with results.

Regarding the cognitive switching stimuli, the effects of other people’s use, discrepancies and deliberate initiatives on trying new features to seek information are all significant with the magnitude being 0.300, 0.103 and 0.141, respectively, thus, H1, H2 and H3 are all supported. Regarding the moderating effect of information need, it positively moderates the impact of other people’s use on trying new features to seek information with the magnitude being 0.139, thus H4 is supported. The moderating effects of information need on the impacts of discrepancies and deliberate initiatives on trying new features to seek information are not significant, thus H5 and H6 are not supported. Regarding the moderating effect of PIIT, it positively moderates the impact of other people’s use on trying new features to seek information with the magnitude being 0.190, thus H7 is supported. The moderating effects of PIIT on the impacts of discrepancies and deliberate initiatives on trying new features to seek information are not significant, thus H8 and H9 are not supported.

Discussion and implications

Moderating effects

The findings of the current study show that information need and PIIT each positively moderate the effect of other people’s use on trying new features to seek information. To further illustrate the significant interaction effects, we conducted separate regression analyses for subgroups of the sample. Following Stewart (2006), the sample was first split and formed high information need subgroup (at least one standard deviation above the mean) and low information need subgroup (at least one standard deviation below the mean). And then other people’s use as independent variable was regressed on trying new features to seek information to form two regression equations which were plotted in Figure 3. Meanwhile, the same procedure was used to illustrate the moderating effect of PIIT on the relationships between other people’s use and trying new features to seek information (see Figure 4).

Figure 3.

Moderating effect of information need.

Figure 4.

Moderating effect of PIIT.

As presented in Figure 3 and Figure 4, for the high information need or high PIIT subgroup, the effect of other people’s use on trying new features to seek information becomes stronger with a steeper positive slope than without the moderating effect. Adversely, for the low information need or low PIIT subgroup, the effect of other people’s use on trying new features to seek information becomes weaker with a shallower positive slope than without the moderating effect. The practical implications of the moderating effects of information need and PIIT are discussed below.

Implications for theory

Structuration and appropriation are two notions of AST. Structuration reflects the social process of putting rules and resources of IT into practice while appropriation reflects how users actively choose to use the structures within IT. Various behaviors related to adaptive IT use has been examined, such as working around an ERP system’s constraints in unintended ways (Boudreau and Robey, 2005), trying new features of Microsoft Office (Sun, 2012), applying business intelligence technologies in novel ways (Wang et al., 2013). However, prior research examining adaptive IT use behavior mainly stayed at the level of technology, system or feature. In the current study, we examined adaptive information seeking in the context of Weibo, which reflects the information level. When users try new features in Weibo to purposively seek information to satisfy information need, they are essentially interacting with the social structures within Weibo. With structuration and appropriation, the structures contained within Weibo and the emergent structures from Weibo use are intertwined, recursively promoting the formation of each other. We suggest that examining adaptive information seeking at the information level could uncover deeper social structures of Weibo.

Examining adaptive information seeking in the context of Weibo is one important theoretical contribution of the current study. Specifically, we examine the effects of cognitive switching stimuli on trying new features to seek information. Meanwhile, recognizing that information need reflects a cognitive gap which prevents individuals from making sense of a particular task and that personal innovativeness reflects individuals’ cognitive style, we examine the moderating effects of information need and personal innovativeness in IT on the effects of cognitive switching stimuli on trying new features to seek information. We suggest that the research model developed in the current study contributes to the theoretical extension of cognitive switching theory and AST in the specific context of Weibo and beyond more generally.

Implications for practice

The rapid development of IT is constantly creating new social structures within IT, resulting in higher requirements for and bigger challenges to IT use. Adaptive information seeking reflects adaptive IT use to purposively seek information as a consequence of a need to satisfy some goal. In recent years, Weibo has become one type of important social media application building on the ideological and technological foundations of Web 2.0 (Kaplan and Haenlein, 2010). Given the autonomy (D’heer and Verdegem, 2014) and unstructured information (He et al., 2013) of Weibo context, adaptive information seeking becomes critical for people to satisfy their information need. This study examines adaptive information seeking in terms of trying new features to seek information. We suggest that the findings have important implications for practice.

From Figure 2, it can be seen that other people’s use, discrepancies and deliberate initiatives each have significant positive effects on trying new features to seek information, which is consistent with the research by Louis and Sutton (1991) who suggested that novelty, discrepancy and deliberate initiative are three kinds of situations prompting people to switch from habit of mind to active thinking, and with the research by Jasperson et al. (2005) who found that active thinking and reflective cognitive processing were critical conditions for adaptive IT use behaviors. Among three kinds of cognitive switching stimuli, it can also be seen that other people’s use is the most important determinant. This finding highlights the importance of learning from observing other people’s use on adaptive information seeking in the specific context of Weibo. Indeed, improvised learning accounts for changing (Boudreau and Robey, 2005); specialized knowledge can be achieved through learning from others (Ryu et al., 2005); novel situations which cover other people’s use, new task and changes in system environments have a significant positive impact on users’ adaptive system use (Sun, 2012). We thus recommend that individuals should be much encouraged to observe other people’s use of new features in Weibo. Observation could possibly occur in many ways such as among friends and classmates, or in laboratories and other places. For developers of Weibo, they should use various effective ways to make Weibo features be easily observed by users. For example, they could encourage users to use live broadcasting (a new feature appearing in Sina Weibo) to show how they try new features of Weibo. They could encourage users to show Weibo features in blogs and discuss Weibo features with other users. They could encourage users to search for guidelines regarding how other users try Weibo features within Weibo or on the Internet. They could encourage users to ask for help about how to use Weibo features in online question and answering communities such as Baidu Know.

From Figure 2 and Figure 3, it can be seen that information need positively moderates the effect of other people’s use on trying new features to seek information, concordant with the research by Huurne and Gutteling (2008) who suggested that individuals with higher information need are more likely to be influenced by others and then perform risk information-seeking behavior. This finding illustrates that the effect of other people’s use on trying new features to seek information will be stronger for users with higher information need. In contrast, the effect of other people’s use on trying new features to seek information will be weaker for users with lower information need. Information need which reflects a cognitive gap motivates individuals to seek information to satisfy their information need (Lu and Yuan, 2011; Marchionini, 1995). In the context of Weibo, it is reasonable to suggest that users who frequently use Weibo and spend a lot of time using Weibo are more likely to have high information need. We thus recommend that developers of Weibo could identify the users with high information need in terms of usage frequency and time and recommend to them blogs or live broadcasting regarding how other users try Weibo features. According to the finding above, when users with high information need have more opportunities to observe other people’s use of Weibo features, they are more likely to try new features of Weibo to seek information.

From Figure 2 and Figure 4, it can be seen that PIIT (personal innovativeness in IT) positively moderates the effect of other people’s use on trying new features to seek information, concordant with the research by Sun (2012) who found that PIIT positively moderates the effect of novel situations which include other people’s use on users’ adaptive system use. This finding illustrates that the effect of other people’s use on trying new features to seek information will be stronger for users with high PIIT. In contrast, the effect of other people’s use on trying new features to seek information will be weaker for users with lower PIIT. Innovativeness reflects individuals’ willingness to change (Hurt et al., 1977) and users with high PIIT are more willing to try new information technologies (Agarwal and Karahanna, 2000). In the context of Weibo, it is reasonable to suggest that users who have used more features of Weibo have higher innovativeness. We thus recommend that developers of Weibo could identify the users with high PIIT in terms of usage percent of all the Weibo features and recommend to them blogs or live broadcasting regarding how other users try Weibo features. According to the finding above, when users with high PIIT have more opportunities to observe other people’s use of Weibo features, they are more likely to try new features of Weibo to seek information.

The current study focused on the context of Weibo, finding that the effect of other people’s use on adaptive information seeking is salient, especially for the users with high information need and high PIIT. Other people’s use which refers to users’ observation of others’ use of new features essentially reflects the nature of learning from observing other people. For developers of Weibo, how to attract users to use more features is a big challenge. Indeed, there is interplay between the types of structures that are inherent in Weibo and the structures that emerge when users use Weibo (DeSanctis and Poole, 1994). The interplay between two structures raises higher requirements for and bigger challenges to Weibo use. We recommend that developers of Weibo should create a learning atmosphere within Weibo. It is reasonable to suggest that once a learning atmosphere is created, more and more users would constantly learn from observing other people and further try new features of Weibo to seek information. In this situation, a learning atmosphere would hopefully lubricate potential conflicts between the social structures contained within Weibo and the social structures which emerge in Weibo use, thus facilitating the two social structures to be jointly optimized. Only in this way, can the Weibo ecosystem be guaranteed, with the result that both developers and users of Weibo can reap significant rewards.

Limitations and future research

This study has limitations. First, some hypotheses regarding moderating effects are not supported. Future research could use other methods such as cognitive neuroscience and focus group interview to collect richer data to give more explanations. Second, given a variety of ways of appropriation as suggested by DeSanctis and Poole (1994), future study could explore other adaptive information behaviors, which we believe would complement the research presented here.

Conclusion

According to AST, when users seek information in Weibo, they are essentially interacting with the social structures within Weibo (DeSanctis and Poole, 1994). Drawing on AST and cognitive switching theory, this study examines the effects of cognitive switching stimuli on trying new features to seek information and the moderating effects of information need and personal innovativeness in IT. This study contributes to adaptive IT use theory by examining adaptive information seeking in the context of Weibo at the information level. Empirical data illustrate that other people’s use, discrepancies and deliberate initiatives each have significant positive effects on trying new features to seek information, which is consistent with prior research. Especially, other people’s use which essentially reflects the nature of learning from observing other people is the most important determinant. Meanwhile, information need and PIIT each positively moderate the effect of other people’s use on trying new features to seek information. We suggest the current study could usefully help understand deeper social structures of Weibo.

Footnotes

Appendix

Appendix B.

Common method bias analysis in PLS.

Constructs	Items	Substantive path coefficients	Method path coefficients
DI	DI1	0.899***	0.057*
DI	DI2	0.950***	−0.058*
DISC	DISC1	0.930***	−0.005
DISC	DISC2	0.925***	0.005
PIIT	PIIT1	0.916***	−0.010
	PIIT2	0.840***	0.052
	PIIT3	0.914***	−0.042
INEED	INEED1	0.951***	−0.004
	INEED2	0.978***	−0.030
	INEED3	0.917***	0.035
OPU	OPU1	0.818***	−0.003
	OPU2	0.958***	−0.056*
	OPU3	0.852***	0.059
TRYSEEK	TRYSEEK 1	0.925***	−0.022
	TRYSEEK 2	0.928***	0.003
	TRYSEEK 3	0.871***	0.027
	TRYSEEK 4	0.872***	−0.007

Notes: *p<0.05; ***p<0.001.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is supported by the National Natural Science Foundation of China [Grant Numbers 71573195, 71774126, 71874124, 91646206, 71921002].

ORCID iD

Yalan Yan

Author biographies

Xianjin Zha is full professor of Information Science at the School of Information Management, the Center for Studies of Information Resources, Wuhan University. His publications have appeared in Library & Information Science Research, Aslib Journal of Information Management, Online Information Review, Journal of Librarianship & Information Science, Computers in Human Behavior, Internet Research and Behaviour & Information Technology, among others.

Kunfeng Liu is a PhD student at the School of Information Management, Wuhan University. His publications have appeared in Online Information Review and Computers in Human Behavior, among others.

Yalan Yan is full professor at the Evergrande School of Management, the Center for Service Science and Engineering, Wuhan University of Science and Technology. Her publications have appeared in Journal of the American Society for Information Science and Technology, Aslib Journal of Information Management, Computers in Human Behavior, Online Information Review, Internet Research and Behaviour & Information Technology, among others.

Chengsong Huang is a PhD student at the School of Information Management, Wuhan University. His publications have appeared in Computers in Human Behavior and Aslib Journal of Information Management, among others.

References

Abel

Celik

Houben

, et al. (2011) Leveraging the semantics of tweets for adaptive faceted search on Twitter. In: Aroyo

, et al. (eds) The Semantic Web – ISWC 2011. Lecture Notes in Computer Science, Vol 7031. Berlin, Heidelberg: Springer, pp. 1–17.

Agarwal

Karahanna

(2000) Time flies when you’re having fun: Cognitive absorption and beliefs about information technology usage. MIS Quarterly 24(4): 665–694.

Agarwal

Prasad

(1998) A conceptual and operational definition of personal innovativeness in the domain of information technology. Information Systems Research 9(2): 204–215.

Agarwal

Prasad

(1999) Are individual differences germane to the acceptance of new information technologies? Decision Sciences 30(2): 361–391.

Armstrong

Hardgrave

(2007) Understanding mindshift learning: The transition to object-oriented development. MIS Quarterly 31(3): 453–474.

Ashuri

Dvir-Gvisman

Halperin

(2018) Watching me watching you: How observational learning affects self-disclosure on social network sites? Journal of Computer-Mediated Communication 23(1): 34–68.

Barki

Titah

Boffo

(2007) Information system use-related activity: An expanded behavioral conceptualization of individual-level information system use. Information Systems Research 18(2): 173–192.

Bandura

(1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Upper Saddle River, NJ: Prentice-Hall.

Barrett

(2018) Technological appropriations as workarounds: Integrating electronic health records and adaptive structuration theory research. Information Technology & People 31(2): 368–387.

10.

Bhattacherjee

(2001) Understanding information systems continuance: An expectation-confirmation model. MIS Quarterly 25(3): 351–370.

11.

Bhattacherjee

Premkumar

(2004) Understanding changes in belief and attitude toward information technology usage: A theoretical model and longitudinal test. MIS Quarterly 28(2): 229–254.

12.

Boudreau

Robey

(2005) Enacting integrated information technology: A human agency perspective. Organization Science 16(1): 3–18.

13.

Broady-Preston

(2009) Structuration and social identity theories: Qualitative methodologies for determining skills and competencies for the information profession in the 21st century. Performance Measurement and Metrics 10(3): 172–179.

14.

Brown

Massey

Montoya-Weiss

, et al. (2002) Do I really have to? User acceptance of mandated technology. European Journal of Information Systems 11(4): 283–295.

15.

Bruce

(1994) A cognitive view of the situational dynamism of user-centered relevance estimation. Journal of the American Society for Information Science and Technology 45(3): 142–148.

16.

Burnett

(2000) Information exchange in virtual communities: A typology. Information Research 5(4).

17.

Burton-Jones

Gallivan

(2007) Toward a deeper understanding of system usage in organizations: A multilevel perspective. MIS Quarterly 31(4): 657–679.

18.

Burton-Jones

Straub

(2006) Reconceptualizing system usage: An approach and empirical test. Information Systems Research 17(3): 228–246.

19.

Chin

Marcolin

Newsted

(2003) A partial least squares latent variable modeling approach for measuring interaction effects: Results from a Monte Carlo simulation study and an electronic-mail emotion/adoption study. Information Systems Research 14(2): 189–217.

20.

Chin

Junglas

Schwarz

, et al. (2014) Don’t mind the gap: A conceptual and psychometric analysis of the individual evaluation of discrepancies in the context of IS user service satisfaction. ACM SIGMIS Database 45(1): 9–28.

21.

D’heer

Verdegem

(2014) Conversations about the elections on Twitter: Towards a structural understanding of Twitter’s relation with the political and the media field. European Journal of Communication 29(6): 720–734.

22.

DeSanctis

Poole

(1994) Capturing the complexity in advanced technology use: Adaptive structuration theory. Organization Science 5(2): 121–147.

23.

Ellis

Oldman

(2005) The English literature researcher in the age of the Internet. Journal of Information Science 31(1): 29–36.

24.

Fang

Xie

(2019) A study on Chinese people’s everyday life information seeking: Taking a comparison between Web and Weibo. International Journal of Services Technology and Management 25(2): 179–196.

25.

Feist

G J

Gorman

(1998) Psychology of science: Review and integration of a nascent discipline. Review of General Psychology 2(1): 3–47.

26.

Ford

Wilson

Foster

, et al. (2002) Information seeking and mediated searching. Part 4. Cognitive styles in information seeking. Journal of the American Society for Information Science and Technology 53(9): 728–735.

27.

Foster

(2004) A nonlinear model of information-seeking behavior. Journal of the American Society for Information Science and Technology 55(3): 228–237.

28.

Fransen-Taylor

Narayan

(2018) Challenging prevailing narratives with Twitter: An #AustraliaDay case study of participation, representation and elimination of voice in an archive. Journal of Librarianship and Information Science 50(3): 310–321.

29.

Fuller

Dennis

(2009) Does fit matter? The impact of task–technology fit and appropriation on team performance in repeated tasks. Information Systems Research 20(1): 2–17.

30.

Gefen

Rigdon

Straub

(2011) An update and extension to SEM guidelines for administrative and social science research. MIS Quarterly 35(2): iii–xiv.

31.

Giddens

(1979) Central Problems in Social Theory. Berkeley, CA: University of California Press.

32.

Giddens

(1984) The Constitution of Society: Outline of the Theory of Structuration. Berkeley, CA: University of California Press.

33.

Goodhue

Lewis

Thompson

(2012) Comparing PLS to regression and LISREL: A response to Marcoulides, Chin and Saunders. MIS Quarterly 36(3): 703–716.

34.

Guo

Zhang

(2010) User attitude towards mandatory use of information systems: A Chinese cultural perspective. Journal of Global Information Management 18(4): 1–18.

35.

Hartwick

Barki

(1994) Explaining the role of user participation in information system use. Management Science 40(4): 440–465.

36.

Hastie

(1984) Causes and effects of causal attribution. Journal of Personality and Social Psychology 46(1): 44–56.

37.

Zha

(2013) Social media competitive analysis and text mining: A case study in the pizza industry. International Journal of Information Management 33(3): 464–472.

38.

Hsieh

Rai

Petter

, et al. (2012) Impact of user satisfaction with mandated CRM use on employee service quality. MIS Quarterly 36(4): 1065–1080.

39.

Huang

Zha

Yan

, et al. (2019) Understanding the social structure of academic social networking sites: The case of ResearchGate. LIBRI 69(3): 189–199.

40.

Huang

Shen

Chiang

, et al. (2007) Characterizing web users’ online information behavior. Journal of the American Society for Information Science and Technology 58(13): 1988–1997.

41.

Hurt

Joseph

Cooed

(1977) Scales for the measurement of innovativeness. Human Communication Research 4(1): 58–65.

42.

Huurne

Gutteling

(2008) Information needs and risk perception as predictors of risk information seeking. Journal of Risk Research 11(7): 847–862.

43.

Jack

Kholeif

(2007) Introducing strong structural theory for informing qualitative case studies in organization, management and accounting research. Qualitative Research in Organizations and Management: An International Journal 2(3): 208-225.

44.

Jasperson

Carter

Zmud

(2005) A comprehensive conceptualization of post-adoptive behaviors associated with information technology enabled work systems. MIS Quarterly 29(3): 525–557.

45.

Kaplan

Haenlein

(2010) Users of the world, unite! The challenges and opportunities of social media. Business Horizons 53(1): 59–68.

46.

Kim

(2014) Observation on copying and pasting behavior during the Tohoku earthquake: Retweet pattern changes. International Journal of Information Management 34(4): 546–555.

47.

Langer

(1989) Minding matters: The consequences of mindlessness-mindfulness. Advances in Experimental Social Psychology 22(12): 137–173.

48.

Lee

JYH

Panteli

Bulow

, et al. (2018) Email adaptation for conflict handling: A case study of cross-border inter-organisational partnership in East Asia. Information Systems Journal 28(2): 318–339.

49.

Hsieh

JJPA

Rai

(2013) Motivational differences across post-acceptance information system usage behaviors: An investigation in the business intelligence systems context. Information Systems Research 24(3): 659–682.

50.

Liang

Saraf

, et al. (2007) Assimilation of enterprise systems: The effect of institutional pressures and the mediating role of top management. MIS Quarterly 31(1): 59–87.

51.

Louis

Sutton

(1991) Switching cognitive gears: From habits of mind to active thinking. Human Relations 44(1): 55–76.

52.

Yuan

(2011) Shall I Google it or ask the competent villain down the hall? The moderating role of information need in information source selection. Journal of the American Society for Information Science and Technology 62(1): 133–145.

53.

Magoi

Aspura

MKYI

Abrizah

(2019) Social media engagement in developing countries: Boon or bane for academic libraries? Information Development 35(3): 374–387.

54.

Malaurent

Karanasios

(2019) Learning from workaround practices: The challenge of enterprise system implementations in multinational corporations. Information Systems Journal DOI: 10.1111/isj.12272

55.

Marchionini

(1995) Information Seeking in Electronic Environments. Cambridge: Cambridge University Press.

56.

Oldham

Cummings

(1996) Employee creativity: Personal and contextual factors at work. Academy of Management Journal 39(3): 607–634.

57.

Orlikowski

(1992) The duality of technology: Rethinking the concept of technology in organizations. Organization Science 3(3): 398–427.

58.

Pang

(2018) Can microblogs motivate involvement in civic and political life? Examining uses, gratifications and social outcomes among Chinese youth. Online Information Review 42(5): 663–680.

59.

Peng

Guo

(2019) A multilevel investigation on antecedents for employees’ exploration of enterprise systems. European Journal of Information Systems 28(4): 439–456.

60.

Peng

Sun

Guo

(2018) Antecedents of employees’ extended use of enterprise systems: An integrative view of person, environment, and technology. International Journal of Information Management 39: 104–120.

61.

Podsakoff

Mackenzie

Lee

, et al. (2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879–903.

62.

Pozzebon

Pinsonneault

(2005) Challenges in conducting empirical work using structuration theory: Learning from IT research. Organization Studies 26(9): 1353–1376.

63.

Ringle

Sarstedt

Straub

(2012) A critical look at the use of PLS-SEM in MIS Quarterly. MIS Quarterly 36(1): 3–20.

64.

Ringle

Wende

Will

(2005) SmartPLS 2.0. Available at: http://www.smartpls.de (accessed 20 May 2018).

65.

Ryan

Grolnick

(1986) Origins and pawns in the classroom: Self-report and projective assessments of individual differences in children’s perceptions. Journal of Personality and Social Psychology 50(3): 550–558.

66.

Ryu

Yong

Chaudhury

, et al. (2005) Knowledge acquisition via three learning processes in enterprise information portals: Learning-by-investment, learning-by-doing, and learning-from-others. MIS Quarterly 29(2): 245–278.

67.

Schmitz

Teng

JTC

Webb

(2016) Capturing the complexity of malleable IT use: Adaptive structuration theory for individuals. MIS Quarterly 40(3): 663–686.

68.

Schön

(1983) The Reflective Practitioner: How Professionals Think in Action. New York: Basic Books.

69.

Spezi

(2016) Is information-seeking behavior of doctoral students changing? A review of the literature (2010–2015). New Review of Academic Librarianship 22(1): 78–106.

70.

Starbuck

Milliken

(1988) Executives’ perceptual filters: What they notice and how they make sense. In: Hambrick

(ed.) The Executive Effect: Concepts and Methods for Studying Top Managers. Greenwich, CT: JAI Press, pp. 35–65.

71.

Stewart

(2006) How hypertext links influence consumer perceptions to build and degrade trust online. Journal of Management Information Systems 23(1): 183–210.

72.

Stones

(2005) Structuration Theory. London: Palgrave.

73.

Straub

Boudreau

Gefen

(2004) Validation guidelines for IS positivist research. Communications of the Association for Information Systems 13(1): 380–427.

74.

Sun

(2012) Understanding user revisions when using information system features: Adaptive system use and triggers. MIS Quarterly 36(2): 453–478.

75.

Venkatesh

Davis

(2000) A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science 46(2): 186–204.

76.

Venkatesh

Morris

Davis

, et al. (2003) User acceptance of information technology: Toward a unified view. MIS Quarterly 27(3): 425–478.

77.

Wang

Hsieh

JJPA

(2013) The contingent effect of personal IT innovativeness and IT self-efficacy on innovative use of complex IT. Behaviour & Information Technology 32(11): 1105–1124.

78.

Wellings Casselden (2019) An exploration into the information-seeking behaviours of engineers and scientists. Journal of Librarianship and Information Science 51(3): 789–800.

79.

Wetzels

Odekerken-Schröder

van Oppen

(2009) Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS Quarterly 33(1): 177–195.

80.

Williams

Terras

Warwick

(2013) What do people study when they study Twitter? Classifying Twitter related academic papers. Journal of Documentation 69(3): 384–410.

81.

Wilson

(2000) Human information behavior. Informing Science 3(2): 49–56.

82.

Wong

Weiner

(1981) Why people ask ‘why’ questions and the heuristics of attributional search. Journal of Personality and Social Psychology 40(4): 650–663.

83.

Kim

Kankanhalli

(2010) Task and social information seeking: Whom do we prefer and whom do we approach? Journal of Management Information Systems 27(3): 211–240.

84.

Zha

Cao

Yan

, et al. (2019a) Exploring innovative information seeking: The perspectives of cognitive switching and affinity with digital libraries. Journal of Academic Librarianship 45(5): 102045.

85.

Zha

Yan

, et al. (2016) Exploring digital library usage for getting information from the ELM perspective. Aslib Journal of Information Management 68(3): 286–305.

86.

Zha

Yang

Yan

, et al. (2019b) Exploring adaptive information sharing from the perspective of cognitive switching: The moderating effect of task self-efficacy. Aslib Journal of Information Management 71 (4): 535–557.

87.

Zha

Zhang

Yan

, et al. (2015) Sound information seeking in Web 2.0 virtual communities: The moderating effect of mindfulness. Behaviour & Information Technology 34(9): 920–935.