Exploring early adolescents’ stressful IT use experiences

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Exploring early adolescents’ stressful IT use experiences © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published version Mehtälä, Saana; Salo, Markus; Tikka, Sara; Pirkkalainen, Henri Mehtälä, S., Salo, M., Tikka, S., & Pirkkalainen, H. (2022). Exploring early adolescents’ stressful IT use experiences. Behaviour and information technology, Early online. https://doi.org/10.1080/0144929X.2022.2109991 2022 RESEARCH ARTICLE Exploring early adolescents’stressful IT use experiences Saana Mehtälä a , Markus Salo a , Sara Tikka a and Henri Pirkkalainen b a University of Jyvaskyla, Faculty of Information Technology, Jyvaskyla, Finland; b Tampere University, Unit of Information and Knowledge Management, Tampere, Finland ABSTRACT Early adolescents are accustomed to using information technology (IT) for different purposes as a part of their everyday life. While IT use can provide many benefits, it can also include negative experiences, such as feelings of stress (i.e. technostress). Considering that young IT users’ technostress perspectives have remained limited in the literature, this paper attempts to contrast their experiences with aspects from earlier research. A qualitative content analysis of semi-structured interviews with 31 primary and secondary school students reveals that early adolescents have encountered stressful IT use experiences that vary based on the degree of stress. The experiences relate to different dimensions of IT use, namely 1) online social (media) behaviour, 2) notification-driven behaviour, 3) social distancing and online education behaviour, and 4) gameplay and other aspects of IT use behaviour. Common denominators for these experiences include the attention or time required by IT use and coming across unwanted content or communications. In addition to reporting negative feelings and emotions, the participants mentioned physical strains, including headache and exhaustion. Although early adolescents can face experiences that can be contrasted with similar stressors and strains as adults, the distinct characteristics of their IT use should be considered in future research. ARTICLE HISTORY Received 15 October 2021 Accepted 1 August 2022 KEYWORDS Early adolescents; stressful experiences; IT use; technostress 1. Introduction Information technology (IT) has become an integral part of the activities that people engage in as part of their daily life. IT users consist of people who operate in various roles in different contexts (Lamb and Kling 2003); these include children and young people who tend to grow up with the latest IT and learn to use it for specific school and leisure purposes. There are various outcomes associated with IT use, including adverse and beneficial effects on well-being. One of these is technostress, which is a term used to characterise the stress associated with IT use (Tarafdar, Cooper, and Stich 2019). Previous research has largely focused on technostress experienced by adults, especially in the work context (see, e.g. Tarafdar, Cooper, and Stich 2019). Even though IT use among adolescents has been widely studied from various perspectives, there is a lack of research focusing on their technostress experiences. This is a valuable perspective because IT use can be quite different for adolescent and adult users. In addition to the obvious developmental differences, adolescents’IT use can be guided by the characteristics and boundaries set by their typical IT use environments (e.g. home, school and hobbies). Additionally, the role of certain actors (e.g. parents and teachers) that might not have as much influence on IT use among adults, should not be overlooked. The current article inspects the different stress-initiating situations that can be present in the everyday IT use of adolescents in home and school settings. Perspectives from the technostress literature (e.g. stressors, strains, and IT characteristics) and the literature on adolescents’IT use (e.g. uses of IT and health and wellbeing outcomes) are combined to build understanding of the different dimensions in adolescents’IT use context that could be relevant for the emergence of stressful experiences. As a whole, the study addresses a combination of influencing factors related to IT use behaviour and user activities, IT characteristics and perceptions of stress. The underlying data consist of 31 semi-structured interviews with Finnish primary and secondary school students. The paper is structured as follows: First, we describe the prior literature on IT use among adolescents and technostress, which forms the theoretical background of the paper. Then, we describe in detail how the interview data was collected and analyzed. The results section presents an overview of the stressful experiences © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. CONTACT Saana Mehtälä saana.s.s.mehtala@jyu.fiUniversity of Jyvaskyla, Faculty of Information Technology, P.O. Box 35, FI-40014, Finland BEHAVIOUR & INFORMATION TECHNOLOGY https://doi.org/10.1080/0144929X.2022.2109991 associated with IT use as reported by the participants, including the identification of significant themes and factors introduced together with examples from the data. The study contributes to research by discussing the characteristics of the early adolescents’stressful IT use experiences, as well as by contrasting the experiences with concepts identified in the technostress literature. Additionally, it is argued that the findings of the study may be relevant to various stakeholders seeking to gain a more profound understanding of early adolescents’IT use behaviour, including designers, health professionals, teachers, and parents. Finally, the limitations of this research and topics for future research are described and are accompanied by concluding remarks. 1.1. Research on IT use among adolescents Since young people have long been a significant user group of IT, previous research has explored IT use among adolescents. This includes identifying relevant demographics associated with IT use (see, e.g. Jackson et al. 2008), as well as the type and prevalence of IT use. In 2011, it was established that 95% of American children aged 12–17 use the internet, while up to 80% of them also used social media websites (Lenhart et al. 2011). Smartphones, in particular, are often used to alleviate feelings of boredom, but also to intentionally relax and connect with friends (Allaby and Shannon 2019). According to a recent Finnish survey (Kinnunen, Taskinen, and Mäyrä 2020), in the age group of 10–19-yearolds, nearly half of the participants claimed to play a digital entertainment game daily, while as many as 79% did this on a weekly basis or more often. In contrast, merely a bit less than 1% of the participants did not play these kinds of games at all (Kinnunen, Taskinen, and Mäyrä 2020). Thus, it can be stated that different online environments seem to form an important domain in the lives of adolescent users. A considerable amount of research exists within various disciplines inspecting the duration (i.e. screen time), type, timing (time of the day), and variety of IT use in connection to adverse physical, psychological, and physiological health and well-being outcomes (Lissak 2018). Screen time, in particular, has been linked with problems in the domains of sleep, metabolism, and mental health (Hale and Guan 2015; Hardy et al. 2010; Cao et al. 2011), including poor outcomes associated with compulsive smartphone use (Panda and Jain 2008). It has also been established that adolescents can experience stress when they cannot access the internet (Díaz-López, Maquilón-Sánchez, and Mirete-Ruiz 2020), while encountering harmful content and communications can even lead to symptoms of posttraumatic stress (McHugh et al. 2018). This shows that stress related to IT use is an existing phenomenon among adolescents. However, gaining a more profound understanding of the topic requires placing more attention towards the various IT use cases in which stress might occur. Although it is evident that the prior literature has traditionally focused on the adverse effects of IT use, more recent reviews undermine the perception of a strong connection between IT use among adolescents and negative wellbeing outcomes (Orben and Przybylski 2019; Dienlin and Johannes 2020). For example, more research might be needed to find support for the current IT use recommendations (Gottschalk 2019). It might be true that a fixated focus on poor well-being outcomes can be viewed as a quite narrow perspective when contrasted with the diversity of interactions and elements present in IT environments, such as the potential positive effects for socialisation (see, e.g. O’Keeffe and Clarke-Pearson 2011; Pujazon-Zazik and Park 2010). Thus, the benefits of IT use for young people should not be overlooked. However, because adolescence is portrayed by different physical, cognitive, and psychosocial developments, careful attention must be placed on the role of IT use in this process. 2. Research on technostress The concept of technostress refers to the stress that a person experiences because of their use of IT (Tarafdar, Cooper, and Stich 2019). Here, stress is viewed as a ‘relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her wellbeing’(Lazarus and Folkman 1984, 19). Altogether, the literature on stress highlights the complicated, dynamic interactions between a person and their environment (e.g. technological environment). For example, the characteristics and features of IT enable use situations that create various kinds of demands, which can result in technostress (Ayyagari, Grover, and Purvis 2011). Although (techno)stress can also be positive (e.g. Benlian 2020), we mainly focus on negative (techno)stress because of its harmful consequences. Technostress research has focused on concepts such as technostressors (i.e. the creators or creating conditions of technostress), strains (i.e. users’negative responses to stressors), and coping mechanisms (i.e. users’attempts to manage stressful situations) (Fischer and Riedl 2017; Tarafdar, Cooper, and Stich 2019). Technostress can emerge in various forms. Much of the prior technostress research has examined 2S. MEHTÄLÄ ET AL. technostress in work-related contexts, such as knowledge work (Fischer and Riedl 2017; Tarafdar, Cooper, and Stich 2019). These studies have identified common forms of technostress, including techno-invasion (e.g. constant overflow of IT use to other domains of life), techno-overload (e.g. too much IT), techno-complexity (e.g. IT feels too difficult to use), techno-uncertainty (e.g. IT changes are too quick), and techno-insecurity (e.g. other people know more about IT) (Tarafdar et al. 2007; Fischer and Riedl 2017). Strains and negative outcomes include poor work productivity, task performance, job dissatisfaction, decreased organisational commitment, turnover, and burnout (Califfet al. 2020; Maier, Laumer, and Eckhardt 2015a; Pawlowski et al. 2007; Pirkkalainen et al. 2019; Ragu-Nathan et al. 2008; Srivastava et al. 2015; Tams et al. 2018; Tarafdar et al. 2007). Recently, researchers have started to pay more attention to technostress emerging in the personal, nonworkrelated use of IT (Benlian 2020; Maier et al. 2015b; 2015c;2020; Salo, Pirkkalainen, and Koskelainen 2019; 2022). In contexts such as social media use, researchers have identified new stressors, including life/social comparison (e.g. contrasting one’s own life with content published by others), social overload (e.g. too many social requests online), and online conflicts (e.g. disputes and arguments on social networking services) (Fox and Moreland 2015; Maier et al. 2015b; Salo, Pirkkalainen, and Koskelainen 2019). With personal use, users often gradually create their own technostress as their use intensifies because of existing use schemes (i.e. influence/persuasion from the design of the IT and other users), as well as because of the fun and hedonic gratifications they experience (Salo et al. 2022). Intensified use creates increasing demands, which can then exceed one’s abilities for dealing with them (Salo et al. 2022). For example, receiving constant personal notifications can at first be considered fun and exciting but can later introduce stress experiences or even turn out to be entirely stressful. Technostress (e.g. from notifications and interruptions) can occur in various contexts, as well as spill over from study/work contexts to personal/leisure contexts and vice versa (Benlian 2020; Galluch, Grover, and Thatcher 2015). Strains and negative outcomes in personal, non-work-related use include exhaustion, use discontinuance, concentration problems, sleep problems, identity issues, and interpersonal conflicts (Benlian 2020; Maier et al. 2015b;2015c; Salo, Pirkkalainen, and Koskelainen 2019). Importantly, (techno)stress is subjective, and users have different responses to externally similarlooking events and situations based on their personal interpretations of the events and situations (Fox and Moreland 2015; Lazarus and Folkman 1984; Salo et al. 2018). Although prior research about work-related and non-work-related technostress provides the initial foundation for investigating technostress in various contexts, it offers few insights into how children and young people experience technostress. Children lead manifold lives that often include the use of IT for various purposes, making it purposeful to examine the characteristics of their potential stress experiences. While the school context, in turn, can be contrasted with work, the obvious differences between these two environments should not remain unexplored. Understanding children’s and young people’s perspectives is important because IT is a natural part of their everyday lives. In addition, IT is an integral part of many societies, making it probable that adolescents will become more exposed to different forms of IT in the future through their increasingly active societal agency. This, in turn, calls for specific attention to the possible effects that IT use might have on the health and wellbeing of young users. In conclusion, while technostress has been studied among the adult population and especially in the work context, research focusing on younger users is lacking. Additionally, studies exist on IT use among adolescents, with their focus more on aspects such as the duration and type of use, as well as their perceived effects on the physical, physiological, and psychological health and well-being of adolescents. However, as IT use can entail diverse types of interactions and various dimensions that might not always be intuitively identifiable from the outside, a focus on aspects such as the duration of use might not be successful in discovering the diversity of the actual stress experiences. In the following sections, the current research attempts to combine these two perspectives by exploring how the stressful IT use experiences faced by early adolescents can be contrasted with technostress. 3. Methodology We chose to conduct a qualitative field study using semi-structured interviews (SSIs) to provide descriptions and explanations for early adolescents’stressful IT use experiences. Thus, the research design has taken a phenomenographic stance. The approach to inquiry can be viewed as interpretive, which allows for focusing on language as a conveyer of phenomena and ideas, here taking into account the complex nature of human experience (see, e.g. Myers 1997). We feel that interviews can help gain an understanding of early adolescents’stressful experiences, while providing early adolescents with the possibility to discuss their BEHAVIOUR & INFORMATION TECHNOLOGY 3 experiences in their own words. In the present study, flexibility (Myers and Newman 2007) was seen as one of the greatest strengths of this data collection technique because of the emerging nature of the topic and the young age of the interviewees. As a whole, the data collection and analysis process could be described as a continuous interplay between the collected data and previous research. We present further details of this process below. 3.1. Data collection SSIs are a common method for collecting data in qualitative research in general (Kallio et al. 2016) as well as in information systems (Myers and Newman 2007). These kinds of interviews characteristically fall between the unstructured and strictly structured interview forms. Although SSIs usually circle around a set of predetermined themes, new aspects arising from the dialogue are an important aspect of the interview process (DiCicco-Bloom and Crabtree 2006). The interview scheme that we used was inspired by previous research because it has been established that the literature and previous work are good sources for building a thorough understanding of the subject matter (Rabionett 2011). This included the identification of themes that could be relevant to the adolescent population (e.g. social pressure and multitasking), followed by more careful wording of questions. Although prior literature guided the process, an attempt was made to ensure that the questions were constructed using a language that would feel natural and easily understandable to the participants. Additionally, the initial scheme was further developed based on the results received from the interviews, such as through the identification of emerging themes. The interview included mostly open-ended questions (e.g. Can you think of a situation where a notification has felt bothersome? What was the situation like? Do you feel that you should answer the messages that you receive as soon as possible? Why or why not?) to ensure the richness and depth of the resulting data, but the dialogue could sometimes be supported with closed questions as well (e.g. How often do you feel that those kinds of notifications can annoy you? Is it daily, weekly, or monthly? Do you feel that you spent more time using applications and devices in your leisure time as well during the online education period?). We conducted interviews with 31 students from primary (n = 20) and secondary schools (n = 11) in the spring of 2018 and fall of 2020. The age of the students ranged from 9–15 years, and 19 of the study participants were male (61%). While adolescence is often considered as the age period between 10 and 19 years of age, its definition can be extended to up to 24 years of age (Sawyer et al. 2018). Because of this, we feel that the age range of the participants is most fittingly described as early adolescents. However, the terms young people and adolescents are used interchangeably in the text to avoid tediousness. In addition, it should be noted that the ages of the students within the same class could differ from one another (e.g. students of a class consisted of 9and 10-year-olds) due to the placement of their birthday during the calendar year. However, this was not seen as a reason to exclude the perceivably younger students from the study, as it seemed irrational to assume that their experiences would noteworthy differ from those of their classmates because of the slight age difference. Even though the technostress experiences of different individuals can differ in intensity, it is important to note that the goal of the current study was to understand early adolescents’experiences through their customary, everyday IT use practices. We feel that this is a valuable perspective, especially because technostress research targeting early adolescents is still emerging. Purposive sampling was used in the sense that the sampling process was foremost guided by the age of the participants, as well as their willingness to participate in the study. Because the study attempted to capture the experiences of early adolescents, there was a need to choose a sufficient number of interviewees representing different age groups. Additionally, even though the participants were viewed as being old enough to express their views on the study topic, it was also true that older participants could often provide more detailed narratives. Thus, having a wider pool of participants was viewed as especially important for capturing the experiences of primary school children. The resulting set of interviews was seen as representing the different dimensions of IT use that can be relevant for early adolescents and, thus, fit the purposes of the current study. After conducting 31 interviews, we estimated that the benefit from conducting further interviews would be marginal. On average, an interview with a primary school student lasted half an hour. In the case of the secondary school students, in turn, the interviews ranged from half an hour to an hour. The interviews were transcribed for the relevant parts. Further information about the age and grade of the participants is provided in Table 1. Table 1. Age and education level of the participants. n Educational Stage Grade Age Range 20 primary school 4–6th 9–12 years 11 secondary school 7–9th 13–15 years 4S. MEHTÄLÄ ET AL. The study was conducted following the university’s ethical guidelines. Because the study involved minors, the permission to participate in the study was required from the children’s guardians. The participants for the interviews were recruited by making enquiries to collaboration networks, including one primary and one secondary school. Initially, the rectors were approached about their willingness to let their school participate in the study. Following a positive answer, a similar inquiry was sent for the teachers regarding the participation of their students. Finally, permission from a parent/guardian was required for the children to participate in the research activities. In addition, the interest of the adolescents themselves in participating was inquired through a permission slip because this was seen to support their agency in decisions affecting their own lives. In the schools, participation was encouraged by stating that a small prize would be raffled among the interview participants. In addition, every participant would receive a small, sweet treat as a token of gratitude for their participation. Because of the COVID-19 situation, permission to conduct the interviews in the fall of 2020 was applied for and received from the university management. Additionally, the COVID-19 situation was continuously monitored and discussed with the rector of the affiliated school, as well as the class teachers. During the interviews, precautionary measures, such as the use of face masks, thorough hand hygiene, and the use of an interview space with sufficient ventilation were taken to minimise the risks of the interview situation itself. 3.2. Data analysis The data analysis process included creating codes that would describe the underlying data and allow a meaningful categorisation for interpretation and analysis, which is customary for qualitative content analysis (see, e.g. Blair 2015). This approach was viewed to fit the data collected through SSIs, especially because children’s technostress experiences are still an emerging topic. Furthermore, a qualitative approach for analysis could help identify the various units of interest that might emerge and be visible on different levels of the data. The coding process was carried out using the NVivo qualitative data analysis software (version 1.0) by a member of the research team. The coding process was jointly designed and reviewed throughout the process among the different team members from various academic positions (e.g. associate professor, post-doc, PhD student). Even though qualitative research is rarely—and does not need to be—free of the interpretations made by the researcher, this was viewed as an important step to increase the quality of the inferences made from the data. The creation of codes and categories followed the identification of relevant topics and themes arising from the interviews (see, e.g. Williams and Moser 2019), here with a focus on capturing the different levels of IT use as closely as possible. For example, the code Notifications are annoying was placed in the category Feelings and events associated with notifications and included subcodes When you’re doing something else and When you get many of a certain kind.In this way, higher categories could be built based on the perceived similarities between the codes, without losing relevant specifications. However, it should be kept in mind that the interview frame was in fact informed by prior research and did affect the themes discussed during the interviews (e.g. notifications). Using this kind of interview frame would connect the study to earlier research and treat the participants in a similar manner, diminishing the effect of the interviewer on the results. However, because the interviewer has an active role in the interview process, these effects cannot be fully eliminated. The data analysis process could be viewed as a dialogue consisting of initial research-driven perceptions, insights arising from the interviews, and how these, in turn, can be contrasted with the views from the literature. In the first phase, only one interview was coded to ensure that this approach fit the data. This was followed by the coding of one interview representing each grade in the data to see how well the logic of the emerging categories and codes worked with the interviewees of different ages. At this point, the main structure of the coding scheme became quite established. Finally, the rest of the interviews were coded, here accompanied by the creation of any necessary subcategories. In the second phase, all the code categories (e.g. Notifications,Messages and comments) and related subcodes (e.g. Notifications are disturbing,Unpleasant messages and comments) were analyzed to identify potential areas of interest in the data that might indicate the presence of stressful experiences related to IT use. The code and underlying interview text were more closely scrutinised whenever the code seemed to portray a negative emotionally loaded experience (e.g. the wording used the phrases stressful/annoying/annoys or disturbing/disturbs) and was related to IT use. In this way, neutral captions characterising IT use (e.g. applications and devices used), as well as the captions representing positive experiences, could be left out of the analysis for this particular study. This approach to analysis was discussed and agreed upon together by the research team. BEHAVIOUR & INFORMATION TECHNOLOGY 5 In the third phase, these data were analyzed in relation to the perceived degree of stress associated with the event or experience. This analysis resulted in 161 quotations associated with some degree of stress. In addition, 29 quotations were identified as not stressful and, hence, were left out of further analyses. We sorted quotations into four categories: highly stressful, moderately stressful, mildly stressful, and potentially stressful. A quotation was categorised as highly stressful if it indicated a strong connection to stress (e.g. use of strongly negative words to describe the emotions at play during a certain event). A moderately stressful quotation, in turn, referred to stress that was clearly stated but included wordings that made the experience seem less stressful (e.g. occasional disturbance). Similarly, descriptions of a mildly stressful experience used words that referred to a very low degree of stress (e.g. slight annoyance). Finally, a quotation was characterised as potentially stressful when it did not include a clear relation to stress but, at the same time, could not be unambiguously categorized as not stressful because of more implicit properties (e.g. describing an event that can be viewed as negatively loaded but does not include explicit references to personal significance). The categorizations were reviewed multiple times together with the research team and refined accordingly. Furthermore, any quotations that could not be unambiguously categorised were discussed to find the most suitable category. Thus, the final data set consists of quotations arranged based on the perceived level of stress experienced by the participants. Additionally, the data were further categorised based on the aspects of IT use that seemed to be relevant to the stress-creating conditions. 4. Results During the interviews, the participants were asked about their customary ways of using IT. The participants were accustomed to using IT for different purposes, in leisure and school contexts. All 31 adolescents were regular mobile phone users and a majority mentioned using a computer. They were also asked about the applications and services that they used on their devices. Communication, video, and streaming services and applications were widely used among the participants, which is not surprising because of their popularity among different user groups. Most of the participants also used their devices to access social media and communities, news and books, games, and music. A typical participant used their phone as the primary device to access different kinds of contents and applications depending on their own interests. Additionally, it was common especially for the boys to use computers or consoles for game playing and related communications. Social media (e.g. Lenhart et al. 2011) and games (e.g. Kinnunen, Taskinen, and Mäyrä 2020) are often used by these age groups, and prior research also suggests that parental models remain an important factor in relation to news consumption, even in the age of mobile devices (Edgerly et al. 2018). In addition, music has traditionally been an important area of interest for adolescents, trending toward an increasing significance in late adolescence (Miranda 2013; Roberts, Henriksen, and Foehr 2009). During the interviews, the participants were asked about their different IT use experiences in diverse situations, such as in relation to interactions with friends, family members, or other internet users. A closer analysis revealed that up to 161 experiences identified in the data could be associated with a certain level of stress (see the data analysis for more information). The results suggest that all 31 participants had faced situations that could be characterised as stressful. Around onethird of the experiences (54) were related to mild stress. Additionally, nearly a fourth of the experiences could be categorised as potentially (37) or highly (37) stressful. The remaining experiences (33) could be characterised as indicating moderate amounts of stress. Highly stressful experience encountered by a secondary school student: Maybe it’s that when I think about social media and the kinds of pressures associated with [your] looks. You also notice it in your mind or, like, in your mental health. I feel this kind of hecticness, like the presence of social media. So it can add to the anxiety related to social media, or this very stressful state of mind. And it’s precisely one of those [things] that interfere with daily life and stufflike that. That they’re all the same. Moderately stressful experience encountered by a secondary school student: Sometimes [the notifications] do disturb me. If I’m, for example, doing something, and suddenly some message arrives, I always feel like checking or viewing it. Mildly stressful experience encountered by a primary school student: It is annoying when I’m in the middle of the game and someone, for example [a friend] calls and asks if I can play with them. Potentially stressful experience encountered by a secondary school student: I should take the sounds offof my games because when I keep them on, this one time, when I did not know how to take them offof my games, my mum had said that if you don’ttakethosenotifications off,I’m going to take your phone completely away from you—oryouhavetodeleteallthegamesbecauseshelost her nerve, because of the constant sounds. Based on their predominant qualities, the stressful experiences encountered by the participants can be 6S. MEHTÄLÄ ET AL. further divided into four main categories: 1) online social (media) behaviour, 2) notification-driven behaviour, 3) social distancing and online education behaviour, and 4) gameplay and other aspects of IT use behaviour. Table 2 presents these categories together with the degree of stress identified among the mentions within each category. In our data, the stressful experiences encountered by the participants most often related to social activities and social media or notifications. Similarly, highly stressful experiences were quite evenly distributed among online social activities and social media, notifications, and social distancing and online education. Moderately and mildly stressful experiences, in turn, were found to be most prevalent in the first two categories. This provides a connection to the technostress literature because prior research has identified stressors related to social networking services (SNSs) use (see, e.g. Maier et al. 2015b; Salo, Pirkkalainen, and Koskelainen 2019). Because notifications have a strong link with IT design itself, the stressful experiences related to these could be contrasted with the technology environmental conditions. These include the IT characteristics that can create a demand for the user (e.g. presenteeism and mobility, see, e.g. Tarafdar, Cooper, and Stich 2019). A more profound division of the categories is presented in Table 3. It is visible that the stressful experiences encountered by adolescents relate to different dimensions of IT use. In relation to social distancing and online education, the negative experiences were most often associated with the feeling of using too much IT. Regarding the technostress literature, this can be connected to information/IT overload (see, e.g. Tarafdar et al. 2007; Fischer and Riedl 2017). The experiences regarding gameplay and other aspects of IT use had to do with negative feelings associated with gameplay, number of advertisements or device use in general. Much like with the notifications, many of these aspects could be connected to the IT characteristics dimension. However, a more careful analysis of the experiences is still needed to evaluate their nature as a whole, as described in the following sections. 4.1. Online social (media) behavior The participants’stressful experiences with online social activities and social media were most often associated with reaction time to messages. Many of the participants felt that there was some kind of pressure associated with how fast they reacted to the messages sent to them. Additionally, some of the participants associated different negative feelings, such as frustration, with a situation where a friend did not answer their message. Maybe in the case of some people. Not with my closest friends, but with more distant friends because I seriously don’t have the energy to open everything that I receive. Social media has a kind of this culture that if you don’t react right away, people will immediately be like, ‘Hey, why aren’tyouanswering?’You sometimes feel like you have to ‘announce that you’re alive.’Because the lives of everyone, or many people, revolve around social media, so maybe in that sense. (Secondary school student) Table 2. Stressful experiences with IT categorised by the perceived degree of stress. Online social (media) behaviour 64 Highly stressful 13 Moderately stressful 10 Mildly stressful 17 Potentially stressful 24 Notification-driven behaviour 56 Highly stressful 12 Moderately stressful 19 Mildly stressful 18 Potentially stressful 7 Social distancing and online education behaviour 28 Highly stressful 9 Moderately stressful 3 Mildly stressful 10 Potentially stressful 6 Gameplay and other aspects of IT use behaviour 13 Highly stressful 3 Moderately stressful 1 Mildly stressful 9 Potentially stressful 0 Total references 161 Table 3. Factors affecting stressful IT use experiences. Online social (media) behaviour 64 The pressure associated with reacting to messages and being present 27 Own reaction time and presence 22 Friends’reaction time 5 A negative disposition toward specific content and communications 20 Received by others 10 Received by oneself 10 The pressure associated with content published by oneself 14 Doubts about information privacy 3 Notification-driven behaviour 56 Disturbance to other activities 34 The feeling of receiving too many notifications 19 Disturbance to family members 3 Social distancing and online education behaviour 28 The feeling of using too much IT 19 Difficulties with participating in online education 8 Disturbance to family members 1 Gameplay and other aspects of IT use behaviour 13 A mismatch between own skills and level required by the game 6 Missing a deleted game 2 Game feels annoying or addictive 1 Too many advertisements 2 Negative feelings associated with device use in general 2 Total references 161 BEHAVIOUR & INFORMATION TECHNOLOGY 7 This one time my friend had shared this funny, umm, video to a group, and then I felt like I commented it too late—that is when I felt that, am I too slow with these things or what—it’s nothing more than that, but, I’m usually the person who puts in the last message. (Primary school student) Well then I kind of, maybe wonder a bit about that. When you’ve been [sending instant messages] a little more actively, and suddenly there is a longer pause, it kind of makes you wonder. (Secondary school student) Prior research has noted that social pressure can be present in instant messaging situations, fostered by IT characteristics such as indicators of availability (Church and De Oliveira 2013; Pielot et al. 2014). The participants also referred to their disposition toward messages and comments that they encounter online. For example, the interviewees expressed sympathy toward the victims of bullying and users who received other unpleasant comments. Yeah, you see that [bullying] every single day. Umm, I have never been bullied, and I have [not] bullied anyone online because I see no sense in that. But you see it quite often. And people make all these completely stupid [online] questionnaires where people can anonymously say anything. It also counts as online bullying if someone calls you names anonymously. (Secondary school student) —For example, there can be pretty mean comments underneath a picture in [social networking service] or something. —In both services [social networking service and messaging service], there have been cases where an embarrassing or otherwise unfavorable picture could have been sent or published. So I think that it would at least be pretty unpleasant if that was done to me. (Secondary school student) I mean, there is no point to go and put comments like that to someone else. —It’s pretty awful for the person who is getting the comments. (Primary school student) Well, I think it feels pretty bad when some people are arguing and the others have to just watch when they send these nasty messages to one another. (Primary school student) Some of the participants also chose to share their own experiences, including disputes with other users. These often occurred with people that they knew in real life as well but in situations related to online game play or social media use. Well, maybe in a game, someone has a bit called you names or something –It’s like, the players have become better and better [in playing the game], and they keep calling the worse players names and that is why I stopped [playing] it. It’s like this, unnecessary, umm, criticizing of others. (Primary school student) And then, umm, sometimes I argue with [a friend]. (Primary school student) When I started this one user [in a social networking service] that I still use, in one video, I received like [tens of] hateful comments—I don’t care about them —I might think about them for a moment, but it’s like, they don’taffect the things I do in any way. (Secondary school student) While adolescents can experience different types of bullying online (see, e.g. Slonje, Smith, and Frisén 2013), the actual extent of this phenomenon as contrasted with traditional modes of bullying has been questioned (Olweus 2012). However, IT environments can work as platforms for bullying, due to different characteristics such as anonymity and not being bound to certain time and place (Slonje, Smith, and Frisén 2013), making cyberbullying a potential threat to adolescents’online social activity (Pujazon-Zazik and Park 2010) as well. The participants were also asked whether they felt pressure in relation to the content that they published. They mentioned spending time thinking about what they published, comparing themselves to others, and feeling pressure in relation to their own appearance. The remaining few mentions were related to the information security of the services. One of the participants, for example, discussed a data breach in a widely used social networking service. —I think about it very carefully, like is it a picture that I want, for example, my grandma to see. If the picture is not something that I would voluntarily show to all of my relatives for example, then I wouldn’t publish anything like that. I want to publish pictures that I don’t have to regret afterwards. (Secondary school student) Well yeah—like, I wouldn’t publish a picture, where you can see that I have just woken up and stufflike that. (Primary school student) Yeah, as I mentioned, I don’t have [an instant messaging service], but I had it for like six months. The same goes with [a social networking service]. But then I deleted them because I want to be unreachable and also because it’sso…How do you put it …Very superficial, like for example. you show in [the social networking service] the best aspects of your life and like that. And so, sometimes that has given me this [feeling] that geez, the life of that person is like a bed of roses, but on the other hand, that is not the reality. (Secondary school student) Well, it’s like when you think about the fact that all the information is taken from a hundred thousand people, both visible and not visible. They see all your passwords, too, email passwords as well. That would surely be pretty unpleasant. (Secondary school student) 8S. MEHTÄLÄ ET AL. O’Keeffe, G. S., K. Clarke-Pearson, & Council on Communications and Media. 2011.“The Impact of Social Media on Children, Adolescents, and Families.”Pediatrics 127 (4): 800–804. Myers, M. D., and M. Newman. 2007.“The Qualitative Interview in IS Research: Examining the Craft.” Information and Organization 17 (1): 2–26. Olweus, D. 2012.“Cyberbullying: An Overrated Phenomenon?”European Journal of Developmental Psychology 9 (5): 520–538. Orben, A., and A. K. Przybylski. 2019.“The Association Between Adolescent Well-Being and Digital Technology Use.”Nature Human Behaviour 3 (2): 173–182. Panda, A., and N. K. Jain. 2018.“Compulsive Smartphone Usage and Users’Ill-Being among Young Indians: Does Personality Matter?”Telematics and Informatics 35: 1355– 1372. Pawlowski, S. D., E. A. Kaganer, and J. J. Cater. 2007. “Focusing the Research Agenda on Burnout in IT: Social Representations of Burnout in the Profession.”European Journal of Information Systems 16 (5): 612–627. Pielot, M., R. De Oliveira, H. Kwak, and N. Oliver. 2014, April. “Didn’t you see my Message? Predicting Attentiveness to Mobile Instant Messages,”In Proceedings of the SIGCHI conference on human factors in computing systems, Toronto, CA, 26 April-1 May, pp. 3319–3328. Pirkkalainen, H., M. Salo, M. Tarafdar, and M. Makkonen. 2019.“Deliberate or Instinctive? Proactive and Reactive Coping for Technostress.”Journal of Management Information Systems 36 (4): 1179–1212. Pujazon-Zazik, M., and M. J. Park. 2010.“To Tweet, or not to Tweet: Gender Differences and Potential Positive and Negative Health Outcomes of Adolescents’Social Internet Use.”American Journal of Men’s Health 4(1): 77–85. Rabionett, S. E. 2011.“How I Learned to Design and Conduct Semi-Structured Interviews: An Ongoing and Continuous Journey.”Qualitative Report 16 (2): 563–566. Ragu-Nathan, T. S., M. Tarafdar, B. S. Ragu-Nathan, and Q. Tu. 2008.“The Consequences of Technostress for End Users in Organizations: Conceptual Development and Validation.”Information Systems Research 19 (4): 417–433. Roberts, D. F., L. Henriksen, and U. G. Foehr. 2009. “Adolescence, Adolescents, and Media.”In Handbook of Adolescent Psychology: Contextual Influences on Adolescent Development, edited by R. M. Lerner, and L. Steinberg, 314–344. New York: John Wiley & Sons, Inc. Salo, M., H. Pirkkalainen, C. E. H. Chua, and T. Koskelainen. 2022.“Formation and Mitigation of Technostress in the Personal Use of IT.”MIS Quarterly, Available in Early View 46 (2): 1073–1107. Salo, M., H. Pirkkalainen, and T. Koskelainen. 2019. “Technostress and Social Networking Services: Explaining Users’Concentration, Sleep, Identity, and Social Relation Problems.”Information Systems Journal 29 (2): 408–435. Salo, M., H. Pirkkalainen, M. Makkonen, and R. Hekkala. 2018.“Distress, Eustress, or No Stress?: Explaining Smartphone Users™Different Technostress Responses,” in International Conference on Information Systems, San Francisco, CA, 13–16 December. Sawyer, S. M., P. S. Azzopardi, D. Wickremarathne, and G. C. Patton. 2018.“The age of Adolescence.”The Lancet Child & Adolescent Health 2 (3): 223–228. Silic, M., and P. B. Lowry. 2020.“Using Design-Science Based Gamification to Improve Organizational Security Training and Compliance.”Journal of Management Information Systems 37 (1): 129–161. Slonje, R., P. K. Smith, and A. Frisén. 2013.“The Nature of Cyberbullying, and Strategies for Prevention.”Computers in Human Behavior 29 (1): 26–32. Srivastava, S. C., S. Chandra, and A. Shirish. 2015. “Technostress Creators and Job Outcomes: Theorizing the Moderating Influence of Personality Traits.” Information Systems Journal 25 (4): 355–401. Tams, S., J. B. Thatcher, and V. Grover. 2018.“Concentration, Competence, Confidence, and Capture: An Experimental Study of Age, Interruption-Based Technostress, and Task Performance.”Journal of the Association for Information Systems 19 (9): 857–908. Tarafdar, M., C. L. Cooper, and J. F. Stich. 2019.“The Technostress Trifecta-Techno Eustress, Techno Distress and Design: Theoretical Directions and an Agenda for Research.”Information Systems Journal 29 (1): 6–42. Tarafdar, M., Q. Tu, B. S. Ragu-Nathan, and T. S. RaguNathan. 2007.“The Impact of Technostress on Role Stress and Productivity.”Journal of Management Information Systems 24 (1): 301328. Williams, M., and T. Moser. 2019.“The art of Coding and Thematic Exploration in Qualitative Research.” International Management Review 15 (1): 45–55. Yang, C. C. 2016.“Instagram use, Loneliness, and Social Comparison Orientation: Interact and Browse on Social Media, but Don’t Compare.”Cyberpsychology, Behavior, and Social Networking 19 (12): 703–708. BEHAVIOUR & INFORMATION TECHNOLOGY 15