1 Introduction

Research conducted to date on problematic Internet use (sometimes referred to as Internet addiction) shows that this phenomenon is changing dynamically with the development of the information society and Information and Communications Technology (ICT) (Tomczyk et al., 2020). Problematic Internet use, i.e. the situation in which there is a loss of control over the way ICTs are used, with a concomitant negative impact of this phenomenon on family and social life, is now an area intensively studied by social scientists (Bányai et al., 2017; Fineberg et al., 2018). Over the years, there has been an increase in research on the mechanisms that determine various forms of problematic Internet use, namely addiction to social networks, dating sites, online shopping, gaming and gambling addiction, FOMO, Nomophobia, and Phubbing (Fineberg et al., 2022). However, despite the exponential increase in reports on various forms of problematic Internet use, there is still a lack of consensus on the theoretical frameworks and diagnostic thresholds that show a clear boundary between destructive and normal ICT use. Such a state of affairs also applies to the problematic use of smart phones (PSUs) (Montag et al., 2021).

PSU has, in recent years, become the focus of discussion among experts on the prevention of risk behaviours mediated by digital media (Gentina & Rowe, 2020; Lopez-Fernandez et al., 2018). Research on problematic smartphone use in vulnerable groups has become essential (Busch & McCarthy, 2021; Pereira et al., 2020) due to several important societal processes that can be observed in recent times. The increasing level of computerisation of various spaces of social and professional life is involuntarily forcing an increase in screen time associated with the use of e-services and cyberspace resources (Abhari & Vaghefi, 2022). Increased ICT use (including smartphones in particular) has been particularly observed among adolescents (Ranjit et al., 2022). The recent increase in screen time is mainly linked to autonomous processes, such as the COVID-19 pandemic, which shifted many activities from the offline world to the online space (Király et al., 2020; Limone & Toto, 2021; Toto & Limone, 2021). The increasing intensity of new media use is also linked to the natural features of the development of the information society (Ziemba, 2019).

The increase in the intensity of smartphone use is of interest and concern to experts involved in the diagnosis and prevention of online risk behaviour. Young people are at particular risk (Smahel et al., 2020). This text seeks to measure the scale of PSU among young people in Bosnia and Herzegovina. The text fills a gap related to the knowledge of the scale of the phenomenon in the indicated country due to the lack of research in this area. The article also considers one particular form of problematic Internet use in relation to the increasing use of smartphones for young people’s communication, entertainment, and educational needs. The study allows for a critical-constructive analysis of the relationship between the increasing use of smartphones in young people’s daily lives and the increasing symptoms of problematic use due to the type of software that is installed on the device.

2 Theoretical Framework and Overview of Research Results

PSU is a type of problematic internet use, i.e. a situation in which there is a loss of control over how the internet and devices that enable access to cyberspace are used. This phenomenon is very often described as risky behaviour mediated by new media (Fischer-Grote et al., 2019). The riskiness of this phenomenon manifests itself in the occurrence of negative phenomena in both individual and social dimensions (Mac Cárthaigh et al., 2020). The question of the scale of the PSU phenomenon and how to measure it is still open (Roig-Vila et al., 2020). Analysing the literature, it can be seen that there are many methodological approaches and sets of factors that are likely causes and consequences of PSU (Davey & Davey, 2014; Elhai et al., 2017; Hale & Guan, 2015).

Due to the variety of diagnostic tools that measure PSU, there is currently a lack of consensus on any single, universal set of indicators and thresholds for the phenomenon under analysis. The abundance of ways to measure PSU allows, on the one hand, easy access to research tools while, on the other hand, it introduces a lack of continuity of measurement. This means that that richness translates into a lack of longitudinal research. This is particularly important in the context of the development of the information society, in which the style of use of new media and e-services is changing, with this also having a bearing on the scale of PSU. Therefore, given the aforementioned conditions, it was decided to use the PSU scale developed by Kwon et al. (2013) in this article. It is a tool that has been used across many different countries and age groups (with a main orientation towards children and adolescents) (Carbonell et al., 2018; Haug et al., 2015). The theoretical framework developed by Kwon et al. (2013) assumes that excessive smartphone use is associated with the following problems: cancellation of meetings in favour of smartphone use, concentration problems due to intensive use of the device, occurrence of physical ailments (pain in the eyes, hands, or neck), negative attitude towards daily functioning without the device, the occurrence of nervousness when unable to use the phone, persistently thinking about the smartphone when unable to use the smartphone, constantly checking information on websites and applications without any real need, loss of control over the period of use, and receiving signals from the immediate environment about the negative aspects associated with smartphone use. The ten PSU indicators listed are presented in Figure number 1. The smartphone indicators listed represent a set of diagnostic criteria that also appear in general definitions of problematic Internet use (Tomczyk et al., 2020), or relate to related behaviours such as Phubbing, Nomophobia (Tomczyk & Lizde, 2022), and FOMO (Tomczyk and Selmanagic-Lizde, 2018). Therefore, when analysing the theoretical and diagnostic framework used, it is important to be aware that PSU encapsulates components of different types of Problematic Internet Use.

Fig. 1
figure 1

Problematic use of smart phones (PSU)—diagnostic indicators

Full size image

In any analysis of the scale of PSU it is important to recall previous research findings in this area, as they provide a benchmark that not only allows comparison of the scale of the phenomenon, but also provides an opportunity to capture trends in the development of undesirable behaviour in groups requiring educational support. A 2020 study of Chinese young people (N = 2090, 12–19 years) found that 16.4 per cent of respondents could be considered to have significant levels of PSU. The study was conducted using the Mobile Phone Addiction Index (MPAI) tool (Hu et al., 2021). Another study conducted in 2017–2018 involving Hungarian young people and those in early adulthood (N = 249) showed that 15.7 per cent of the respondents are categorised as having a very high level of PSU (Kiss et al., 2020). In contrast, diagnoses conducted in Bangladesh (N = 601) in the second half of 2020 using the Smartphone Application-Based Addiction Scale showed that for 86.9 per cent of the respondents 21 or more of the 36 indicators in the scale measuring PSU were elevated (Hosen et al., 2021). On the other hand, a study conducted a few years ago in South Korea using the Korean Smartphone Addiction Proneness Scale (2017, N = 370 high school students) found that 13.5% had high levels of smartphone addiction (Lee & Lee, 2017). Interesting results are brought by data collected by researchers in Lebanon who, using the MPPUS-10, obtained responses from 207 respondents between the ages of 18 and 65. The results explicitly show that younger individuals have much higher PSU rates than older groups (Nahas et al., 2018). These studies offer several important findings. Firstly, younger users are at much higher risk of PSU than older groups due to their style of new media use. Secondly, there are many ways to measure and analyse PSU (with differing thresholds for considering whether the phenomenon is realistically problematic and therefore harmful), which is why there are so many methodological rifts and disagreements about the true extent of the negative impact of smartphones on young people.

When analysing research findings, it should also be emphasised that very often PSU themes are not shown directly through descriptive statistics that present the scale, but with an emphasis on co-occurring factors (sociodemographic, social determinants, biopsychic characteristics, new media use style) (Lo Coco et al., 2020; Yang et al., 2020; Romero-Rodríguez et al., 2020; Eichenberg et al., 2021). Clearly, research that triangulates multiple tests that measure factors that are determinant variables of PSU is extremely valuable; however, the absence of data on the magnitude of PSU (particularly with representative samples) causes issues, particularly for those involved in the diagnosis and prevention of cyberhazards.

Another reason for PSU being characterised by variable diagnostic criteria is due to the intense development of the information society. This phenomenon can be linked to the overarching concept of technological determinism (Dafoe, 2015; Hauer, 2017; Wielgosz, 2017). When analysing the mechanisms determining PSU, it is worth noting the determinants of the development of the information society, linked to the increase in the efficiency of digital devices, the exponential growth of e-services and digital content, the transfer of some services and activities from the offline world to the online space, the increasing speed of Internet connection, the possibility of free and low-cost Internet access using WiFi networks and data packages offered by ISPs, and the development of software aimed at creating spaces for intensive interaction by users of selected e-services, e.g. social networks. This last phenomenon is of particular interest in the context of PSUs. Technological determinism in this case manifests itself in an increase in the period for which software installed on smartphones is used, the purpose of which applications is not only to facilitate access to selected e-services, but also to increase the time in which the user views the application on their device (Ashford et al., 2020). In this case, users receive many of the gratifications associated with accessing SNSs (including likes, interactions, and ready access to information), which enables the rapid satisfaction of needs of various orders (Korhan & Ersoy, 2016). This type of software is therefore one of the factors that can lead to increased screen time and, in some cases, loss of control over smartphone use, which fits into the theoretical model of the concept of problematic internet use, or PSU (Anderson et al., 2017; Gioia et al., 2021; Kuss & Lopez-Fernandez, 2016). The conditions outlined above regarding PSU and software for accessing e-services (including SNSs, which are very popular among young people) suggest a relationship between smartphone use style and possible negative consequences resulting from the impact of digital devices on young people's behaviour.

This text is an exemplification of research attributed to the risk paradigm of media pedagogy (Tomczyk, 2021). The research is part of an attempt to show the scale of PSU while referring to the phenomenon of technological determinism, exemplified by behavioural change resulting from the additional software installed on smartphones. The inclusion of the risk paradigm in the theoretical framework is linked to the emphasis on the importance of diagnostic research targeting new media mediated behaviour, which should be used to design preventive measures and critically-constructively interpret the scale of the negative impact of new media on the quality of human life.

3 Research Methodology

3.1 Aim and Subject of the Study

The aim of the research was to reveal the level of problematic smartphone use among young people in Bosnia and Herzegovina. An intermediate objective was to contrast this phenomenon with the frequency of use of social networking applications. The research is part of the discussion not only on the scale of the phenomenon of problematic smartphone use, but also reveals a gap in the diagnosis of selected forms of problematic Internet use (Fineberg et al., 2022). The objectives of the research yield the following research problems:

  • RQ1: What is the level of problematic smartphone use among young people in Bosnia and Herzegovina?

  • RQ2: What is the scale of the use of social networking applications on smartphones?

  • RQ3: To what extent is the co-occurrence of problematic smartphone use linked to the use of social networking applications installed on smartphones?

The subject of the study was the answers provided by young people. For research problems 1 and 2, calculations showing gender differences were also carried out.

3.2 Research Tool

The main variable, namely the level of PSU, was measured using a popular diagnostic questionnaire developed by Kwon et al. (2013). The tool consisted of 10 questions with the option to answer on a scale from: 1—I strongly disagree to 6—I definitely agree. The internal psychometric properties were checked using EFA—Exploratory Factor Analysis—which can be found in appendix number 1. The survey tool was also checked for internal consistency, with McDonald's ω = 0.838 and Cronbach's α = 0.836 both suggesting a high level of consistency.

In addition, an in-house (proprietary) survey tool was added to the diagnostic battery to measure the intensity of the use of social networking applications on the smartphone. The scale consisted of 6 items measuring the frequency of use of Facebook, Instagram, Snapchat, TikTok, Twitter, and Pinterest. Respondents could provide answers ranging from 1—Never—I don't have the app installed to 5—Very often—practically all the time. The internal consistency of this tool was McDonald's ω = 0.562 and Cronbach's α = 0.525. The internal consistency showed strong heterogeneity in the frequency of use of the different types of application, which will have implications for how the results are analysed and interpreted.

The relationship between the indicators from the tool is included in Appendix 2.

3.3 Research Procedure and Characteristics of the Research Sample

The study included 1,024 young people aged 14–19 from the Federation of Bosnia and Herzegovina, and was implemented online The distribution of the questionnaire was carried out in a non-random manner using the personal contacts of the interviewers at the University of Sarajevo, and letters sent directly to school managements requesting participation in the survey. Data collection took place in the first half of 2022, so the results presented refer to the post-pandemic period. The research was conducted using the LimeSurvey web-based system.

A total of 1,024 questionnaires were obtained that met the criterion of being fully and correctly completed. The sample included 627 girls (61.23%) and 397 boys (38.77%). The mean age of the respondents was 15.854 with a standard deviation of 1.687. The young people participating in the study resided in the following areas: City (N = 673, 65.723%), Countryside (N = 149, 14.551%), and Suburban area (N = 202, 19.727%).

3.4 Research Ethics

The research was conducted in compliance with the General Data Protection Regulation (GDPR) procedure and is in accordance with the Helsinki Declaration. Each participant was informed of the purpose of the survey in a cover letter. The online survey was constructed in such a way that ensured anonymity. The design of the survey tool and the data processing procedure did not allow for the identification of respondents. During the course of the study, each respondent was able to opt out of the research procedure at any time. The research was approved by the Ethical Committee of the Ministry for Education of Bosnia and Herzegovina (approval number: 11–04-34–120491/22).

3.5 Research Results

3.5.1 RQ1: Level of Problematic Smartphone Use among Young People in Bosnia and Herzegovina

The PSU scale among young people is the main variable in this article. PSU is a complex phenomenon that does not possess indicators of equal saturation. Analysing the data in Table 1, it can be noted that the most common symptom of this phenomenon is a loss of control over the time spent using the device. One in four young people is constantly checking for new information on the SNS on their smartphone (40% of respondents confirm that this is the symptom that affects them most often). In contrast, one in five respondents note that they are currently unable to stop using their smartphone because the device has a major impact on their lives. The least frequently confirmed symptoms of PSU are the cancellation of meetings due to the need to use the device, pain due to overuse of the phone, and feeling anxious when unavailable online. A detailed description of the PSU scale is presented in Table 1.

Table 1 Scale of problematic use of smart phones among young people—descriptive statistics
Full size table

Using the k-means cluster analysis technique, four groups of users can be distinguished, each having different intensities of the individual factors depicting PSU levels. Cluster 1 is represented by young people who have a high intensity of 7 of the 10 factors covering problematic Internet use; this cluster represents 17.19% of the study sample. Cluster 2 and Cluster 3 together represent 51.95% of the study sample, and respondents belonging to these communities show only selected PSU indicators at medium and high levels. This is a collective that mainly has a problem with controlling how long smart phones are used for. The last Cluster, number 4 containing 30.86%, are young people who use smartphones in a non-problematic manner. A graphical depiction of the different communities is presented in Fig. 2

Fig. 2
figure 2

Scale of problematic smartphone use among young people—results of cluster analysis

Full size image

Detailed data related to descriptive statistics for the four groups are presented in Table 2. Each group is characterised by a different saturation of individual indicators of the main variable. The data presented in Table 2 clearly suggest that the population of young people is heterogeneous in terms of how they use the smartphone. Particular attention, in the context of preventive measures, should be given to those classified in Cluster 1. Elsewhere, more than half of the respondents (Clusters 2 and 3) require reinforcement of self-control skills related to the time spent using the smartphone.

Table 2 Descriptive statistics for the four clusters—k-means cluster analysis
Full size table

When analysing the variation in individual indicators related to PSU, it was noted that half of the ten indicators were more common among girls than boys. On the basis of the data obtained, it was noted that the greatest gender differences related to loss of control over how long was spent on the smartphone, the occurrence of nervousness and impatience when the smartphone was not in hand, and the occurrence of symptoms of physical pain. Analyses were performed using the non-parametric Mann–Whitney test due to the distribution of indicators. Detailed differences resulting from the PSU scale and gender are presented in Table 3 and in Fig. 2.

Table 3 Gender differences in problematic smartphone use (descriptive statistics)
Full size table

3.5.2 RQ2: Scale of Intensity of Use of Social Networking Sites through Additional Software

Not all young people use social networking applications on their smartphones, with some instead accessing such sites via their phone’s browser instead; or they simply do not use that social networking service. Considering the features of the internal consistency scale of the intensity of use of social networking applications on the smartphone, the fact that not all types of software enjoy the same popularity (and therefore intensity of use) among young people also becomes apparent. From the responses collected, it is noticeable that just over half of the respondents use Instagram practically all the time. Almost one in four checks for new information on TikTok all the time. The least popular applications are Twitter, Pinterest, and Facebook. Detailed statistics on the frequency of SNS use via applications on the smartphone are presented in Table 4.

Table 4 Scale of intensity of SNS use via applications on the smartphone—descriptive statistics
Full size table

Considering the results of the Mann–Whitney test analysis, it appears that girls are slightly more likely to use Instagram, Snapchat, and Pinterest. For other types of software, the differences by gender are not statistically significant. For Pinterest software, gender differences are strongly noticeable as evidenced by the data presented in Table 5 and Figs. 3 and 4.

Table 5 Scale of intensity of use of social networking sites via applications on the smartphone vs. gender
Full size table
Fig. 3
figure 3

Problematic use of smartphones vs gender

Full size image
Fig. 4
figure 4

Scale of intensity of use of social networking sites via applications on the smartphone vs. gender

Full size image

3.5.3 RQ3: PSU Co-Occurrence and Use of Social Networking Applications

The PSU concept is a process made up of multiple indicators of a phenomenon’s occurrence. Therefore a multivariate regression analysis was carried out to compare each factor with the intensity of use of the selected application. Based on the detailed analysis, it was noted that the increase in most PSU indicators is mostly associated with two types of software, namely Instagram and TikTok. However, the β coefficient clearly suggests that this relationship is characterised by a low strength of dependence and a low measure of model fit (R2). This is a very important indication that we cannot directly link the level of PSU to the intensity of use of smartphone applications. The details of the multivariate regression analysis are presented in Tables 6 and 7.

Table 6 Multivariate regression analysis (Part I)
Full size table
Table 7 Multivariate regression analysis (Part II)
Full size table

4 Discussion

PSU is an issue of increasing visibility in analyses of the literature attributed to the pedagogy, psychology, and sociology of new media. Regardless of the differences due to the diagnostic tools used and the thresholds at which diagnostic results are classified as threatening, PSU is becoming an increasingly discussed e-risk (Harris et al., 2020; Ryding & Kuss, 2020). The results obtained within Bosnia and Herzegovina do not significantly deviate from similar studies conducted with other survey tools in other countries (Hu et al., 2021; Kiss et al., 2020; Lee & Lee, 2017). This means that several percent of young people (in the case of Bosnian and Herzegovinian young people, 17.19%) have an intensity of PSU symptoms to classify this group as requiring intensive preventive actions (Gacka, 2019). These activities should primarily be aimed at strengthening self-control, as well as the ability to segregate relevant from unimportant information. This indication follows on directly from the data obtained, as loss of control of the time period of smartphone use and FOMO are the most highly saturated diagnostic indicators of PSU. At the same time, it should be emphasised that the tool used did not confirm that all PSU indicators occur at the highest intensity in the most at-risk group. Therefore, when analysing the PSU phenomenon, it is necessary to analyse very precisely the individual areas (indicators) in which the young person is characterised by aberrant behaviour. Referring to RQ1, a further important conclusion also emerges that should be clearly articulated—PSU does not affect all or most young people, but only a select group. The research on PSU in terms of measuring the magnitude of the phenomenon is consistent with previous findings on problematic internet use (Pyżalski et al., 2019, 2022). The data presented therefore confirm the further need to address stereotypes related to the style of use of new media by young people. The collected data also confirm one regularity that has been noted in other studies. The sociodemographic variable gender differentiates selected PSU indicators. Confirmation that girls have higher levels of PSU than boys can also be found in other studies of this type that have been conducted in other geographical spaces (Fischer-Grote et al., 2019; Pereira et al., 2020).

Turning to RQ2, an interesting trend regarding the style of smartphone use of young people becomes apparent. Firstly, not all young people use applications that offer quick access to SNS content. Many young people prefer traditional access via a web browser. This is a solution that may minimise FOMO to some extent, as well as screen time length, as applications are designed specifically to encourage users to increase their SNS activity, such as via push notifications that appear on the smartphone screen (Rozgonjuk et al., 2020, 2021; Yang et al., 2019). Nevertheless, based on the data collected, it was noted that more than half of young people are logged in almost all the time to Instagram, and a quarter also use the TikTok application all the time with little interruption. These two pieces of software have significantly overtaken Facebook. The results from Bosnia and Herzegovina are in line with findings from other countries, where variation in the frequency of use of individual SNSs by metric age is apparent (Coman et al., 2021; Gentina et al., 2021; Li, 2022). Variation in the popularity of individual SNSs is already apparent at a very early stage in the use of cyberspace resources (Iwanicka, 2021) and changes with circumstances (Pyżalski & Walter, 2021). The data from Bosnia and Herzegovina also provided one very important piece of information. As in the case of PSU syndromes, which are more common among girls, the use of the applications mentioned in this study is also more common to the group. Above all, girls are more likely to use Instagram and Snapchat, which is explained by experts as satisfying the need for social belonging, or building one's own identity and self-worth (Anderson & Jiang, 2018; Wilksch et al., 2020).

The increase in selected PSU indicators is most often associated with the use of TikTok and Instagram. This software is at the same time the most frequently used on smartphones. However, it should be made clear that the strength of this correlation is at a low level. This means that one has to be very careful in linking the use of particular software with an increase in undesirable behaviour. The reasons for the high level of PSU should therefore be sought in other bio-psycho-social factors than the style of use of applications, one of the aims of which is to increase the intensity of SNS use and thus increase screen time. The data collected thus provide a voice in the discussion on the need to discover the determinants of high PSU levels hidden outside the smartphone (Kim, 2017).

4.1 Limitations and New Research Directions

The present research fills a gap in the diagnosis of PSU among adolescents within Bosnia and Herzegovina. This research is based on data related to self-declarations among young people using an online questionnaire. Therefore, a general weakness of this type of research is the lack of any real confirmation of the style of smartphone use, which does not include device data or targeted observation. Quantitative studies using standardised diagnostic questionnaires may therefore be subject to the error of subjectivity.

Furthermore, the present research is narrowed down to show the magnitude of the phenomenon of two processes related to the use of smartphones, i.e. the negative consequences that are determinants of PSU, as well as the frequency of use of social networking applications. Given the bio-psycho-social complexity of PSU determinants, there is a need to extend the research model by triangulating data relating to broader contexts, where PSU can be both an independent variable and an indicator of other processes, not necessarily related to the technological dimension of device use (Busch & McCarthy, 2021; Sohn, et al., 2019).

5 Conclusions

The research findings presented here are linked to the intensely developing information society (Ziemba, 2021). The increasingly widespread digitisation of everyday life, manifested, among other things, in increased screen time, raises questions about the extent of PSU. Young people seem to be a particularly vulnerable group (Lee et al., 2018; Meng et al., 2020), which, as it were, forces an increase in specific diagnostic and preventive interest in that group’s style of new media use. The present study, however, does not confirm that all or most young people have a high saturation of all PSU indicators. Furthermore, it was not observed that the use of social networking applications was in a strong relationship with PSU indicators. The collected data therefore provide a voice in the discussion on the psychosocial functioning of global adolescents in the information society, thus bridging the stereotypes that tend to dominate the wider discussion. The data collected also show the complexity of diagnosing PSU, which is still characterised by an open theoretical framework that changes with the evolution of the understanding and interpretation of problematic Internet use, or the debatable concept of Internet addiction.