Do (how) digital natives adopt a new technology differently than digital immigrants? A longitudinal study

Highlights

• Nonage-based criteria were used to study digital natives and digital immigrants.

• An integrated framework of technology use has been tested across DNs and DIs.

• Three postadoption mechanisms have been used to study the continue use behavior.

• The postadoption mechanisms are sequential belief updating, feedback, and habit.

• Results of multigroup analysis explain the postadoption behavioral differences of DNs and DIs.

Abstract

Although the information systems (IS) literature has revealed a variety of mechanisms involved in technology adoption and postadoption use, the literature lacks insights about how individuals with different usage characteristics process the information related to new IS and how their belief judgments and use behavior unfold over time. This study fills this void in the literature by conceptualizing and testing a comprehensive model to investigate the impact of user orientation toward technology use by digital natives (DNs) and digital immigrants (DIs) on technology continuous use behavior. The effect of DNs and DIs is currently gaining the attention of researchers. This study investigated the postadoption and use behavior of these groups using a three-wave panel model and with decomposed theory of planned behavior (DTPB) as the initial adoption model. The longitudinal model is a unified framework that sheds light on four different mechanisms underlying postadoption phenomena: (1) the belief judgment evaluation processes suggested by the DTPB model, (2) sequential updating mechanism, (3) feedback mechanism, and (4) habit mechanism. Based on multigroup analysis, we show that a clear pattern of differences in effect exists between DNs and DIs with respect to the sequential belief updating mechanism and that these results are relatively stable over time.

Introduction

Every generation is shaped by the major developments and movements that take place during its time. The current generation is shaped by digital devices and ubiquitous technology. In today’s digitalized times, everything from tablets to health applications for parents to children’s games on computers has been affected by new era technologies. Major part of our daily activities, such as communication, meetings, reading, gaming, and entertainment, have now been digitalized ([1], 2016). Observing these behavioral preferences for the use of digital gadgets or technology, Prensky [2] has coined the terms “digital natives” and “digital immigrants” for the current tech-savvy generation. Digital immigrants (DIs) represent individuals born before the 1980s, or before the existence of digital technology; digital natives (DNs) refer to those born after the 1980s and exposed to these digital technologies at a very early stage of their lives. DIs too uses digital devices but only as adults; however, unlike DNs, their exposure to digital technology came at a much later stage of their lives. DNs, with early access to digital technologies, usually engrossed themselves in a networked world and were able to learn and use digital technology in a manner better than their counterpart DIs. Although many DIs have transformed themselves into expert users of digital technology, their attitudes toward technology differ from those of the DNs [3]. These behavioral differences highlight a wide generation gap. Most studies have treated DNs and DIs as mutually exclusive [[4], [5], [6]].¹ Such studies place younger users born after the 1980s on one side of the gap (i.e., DNs) and adults born prior to the 1980s on the other side (i.e., DIs). Such a distinction, based on age, between these two types of users is problematic because the literature has focused primarily on the young.

Researchers also have criticized this age-based dichotomous classification, advising instead that these two types of digital users (DNs and DIs) should be studied along a continuum depending upon their level of preference to use digital technologies [7,8,3]. These researchers argue that although DNs are assumed to be more tech-savvy, they share the same digital world with DIs, in which many DIs, older in age, not only command a superior knowledge but also are more proficient users of digital devices than are young people. Additionally, Herring [9] said that neglected in most of this discourse about the internet generation and its transformative potential is the continued presence and influence of adults in the larger digital landscape inhabited by young people. She particularly emphasized the characteristics of young people and their online and communication behaviors. While agreeing to the revolutionary impact of the digital disruptions, she recommends a paradigm shift from focusing only on the young generation to considering all generations. Thus, the present research has adopted the behavioral-based explanation of DN and DI classification. Studies have shown that apart from age factor, these behavioral differences are due to digital environment and individual’s experience with and the breadth of technology used [10,11]. Sharafi et al. [12] has found that personality orientations in conjunction with socio-demographic variables can result in differential engagement modes and flow experiences. Therefore, even several older generation but socially advantaged people are found to be more tech-savvy than younger but socially disadvantaged people [13].

In this research, computer engagement has been used as an individual difference variable to study DNs and DIs. Because there is a lack of any accepted measurement scale for delineating DNs/DIs, an attempt was made to identify the closest variables that could be utilized to delineate both the groups and to match with the definitional criteria of DNs/DIs. The rationale behind adopting the “computer engagement” scale to identify DNs/DIs is as follows: (1) the research attempts to delineate the digital users based on their behavioral differences rather than age. As there is an absence of an established measurement instrument (i.e., scale) to measure DNs/DIs, the study looked into literature to find a suitable (proxy) scale to capture the behavioral differences (particularly with regards to the usage of digital devices such as computers and smartphones) between DNs and DIs. Additionally, as the study context was the implementation of a learning management system, which was at that time only accessible through computer (desktop/laptop compatible), the use of “computer engagement” scale was deemed as suitable; and (2) though age may provide opportunities to gain exposure to and familiarity with technologies and thereby lessen anxiety and promote interest and usage, it is engagement that is the key driver for them being comfortable with these technologies and thereby leading to greater usage. Thus, more than age, the engagement level identifies whether an individual is classified as DNs or DIs [3,8].

The academia and business worlds are also witnessing a similar paradigm shift in technology use behavior. Therefore, it becomes imperative to investigate whether DNs and DIs are accepting new information systems (IS) similarly, and, more importantly, to foster sustained use of the new IS, how their belief judgments and use behavior unfold over time. Each time users are exposed to a new technology, they develop beliefs concerning system use, such as ease of use and self-efficacy, and it is on these beliefs, they conduct their usage behavior. Although the initial adoption phenomenon has been studied through various initial adoption models, the eventual success of a new technology relies upon both its first-time adoption (initial use) and the subsequent usage behavior (continued usage) of the users [[13], [14], [15], [16], [17]]. Continued usage behavior by an individual depends upon four pillars: (a) initial adoption, i.e., the effect of current belief evaluation on current use behavior; (b) sequential belief updating mechanisms, i.e., the effect of current belief evaluations on subsequent belief evaluations [18]; (c) feedback mechanisms, i.e., the effect of current use behavior on subsequent belief evaluations [19]; and (d) habit, i.e., the effect of current use behavior on subsequent use behavior [20].

Four leading IT/IS journals were chosen for a content analysis. These included Information Systems Research, Journal of Management Information Systems, Management Information Systems Quarterly, and Management Science. Their content from 2005 to 2015 revealed that though initial adoption was studied in over a hundred articles, literature on continued usage is fairly limited. Even for this literature, only two research articles [16,17] study it longitudinally. These two works investigated the important predictors of continued usage behavior by integrating the three postadoption mechanisms (sequential belief updating, feedback, and habit) within an initial adoption theoretical framework of the technology acceptance model (TAM) [17] and the theory of planned behavior (TPB) [16]. Only Kim and Malhotra’s [17] study used primary data. This sole study, however, was limited by its use of a two-wave panel data model, where relationships found could be methodological artifacts. The sequential updating, feedback, and habit mechanisms appear only once in a two-wave panel model, thus not offering robust support to long-term continuous usage behavior. Kim and Malhotra [17] quote, “To gain deeper insight into continued IS usage, researchers are encouraged to conduct more than two waves of panel surveys” (p.753). The three-wave panel model should be preferred over the two-wave model because previous judgments could have long-term effects on the current judgment [16]. Even though Kim [16] used a three-wave panel model, the author employed secondary data represented by factor correlations. As the original studies did not consider issues of measurement errors when calculating factor correlations, the results of Kim [16] are likely to be erroneous. In addition to the above research gaps, assessment of common method variance and social desirability bias (SDB) were not performed in some of these studies.

Considering the above stated concerns, this study re-examines the postadoption mechanisms that influence continued usage behavior of DNs and DIs. Thus, the objective of this research is to study postadoption phenomena of information technology usage of DNs and DIs. This study fills a void in the literature by conceptualizing and testing a longitudinal adoption model based on the decomposed theory of planned behavior (DTPB) [21]. Although Kim and Malhotra [17] and Kim [16] have used the TAM and the TPB, respectively, as their initial adoption models, our study uses DTPB as the initial model. The DTPB model has advantages over the other models because it includes salient beliefs that affect IT usage [22]. DTPB is better than TPB because of its increased explanatory power; further, DTPB provides a fuller understanding of usage behavior than that with TAM [21]. Our study investigates the continuous use behavior of DNs and DIs using a three-wave panel model based on primary data, with DTPB as the base model. Following efforts in past research, this longitudinal model integrates the four postadoption phenomena/mechanisms. This study undertakes a multigroup analysis to examine the (non) invariance of the three-wave model between the two groups: DNs and DIs. In this research, the data are validated for common method bias (CMB) and SDB. Thus, our research plugs gaps identified in previous research.

To enhance organizational productivity, it is important for managers to help their workers develop a habit of using desirable technology features. The outcomes of this study may provide managers useful insights about how to ensure the optimum utilization of information technology keeping in mind the DI-DN classification of their workers. The model proposed in this study may further help in the deployment of new technologies. The organization of the remaining paper is as follows: Section 2 presents a review of the literature on DNs/DIs and postadoption mechanisms. Section 3 describes the proposed model and hypothesis development. Section 4 discusses our research methodology and data collection. Section 5 provides details about the data analysis. Next, section 6 provides the results obtained from the analysis. Section 7 offers a discussion based on the findings, and section 8 to section 11 present the theoretical contribution, managerial implications, limitations and scope for future research, and conclusion of the study, respectively.