Abstract
Smartphone addiction has been widely recognized as a notable global public health issue. Researchers have already conducted numerous researches to look into various aspects of smartphone addiction such as risk/protective factors, negative consequences, prevalence, and intervention strategies. Despite the existence of these researches, they remain fragmented in the existing literature, which make it challenging for researchers to have an integrative view of smartphone addiction literature. Addressing this research gap, this study conducts bibliometric analysis on the 924 publications extracted from Scopus. More specifically, through the use of various performance indicators (e.g., total number of citations/publications and citation per publication), this study has identified the publication trend, most influential/productive journals, authors, and publications. Furthermore, through co-word analysis conducted on the title and abstract of publications, this study has also identified and briefly elaborated on the research themes in the existing literature of smartphone addiction (i.e., protective/risk factors, neurobiological mechanisms, and intervention; prevalence and negative consequences; scale development and psychometric properties). Taken together, this study provides researchers especially those new to the field of smartphone addiction a one-stop platform to gain an aggregative view of smartphone addiction literature.
Plain language summary
With the aim to provide researchers an integrative view of smartphone addiction literature, 924 publications related to smartphone addiction were extracted from Scopus and analyzed using bibliometric approach. More specifically, through examining the historical trajectory of publication quantity in the field of smartphone addiction, this study has identified the publication trend. It has found that the number of publications has been increasing in the period of 2011 to 2022 and most of the articles (approximately 67%) were published in the period between 2020 and 2022. These have reflected that smartphone addiction is a relatively new and emerging field of research that has recently started to gain popularity, which is expected to stay relevant and gain more and more attention from academic community in the coming years. Furthermore, with the use of various performance indicators (e.g., total number of citations/publications and citation per publication), this study has identified (1) most influential/productive journals, (2) most influential/productive authors, and (3) most influential publications in the existing literature of smartphone addiction, which have respectively revealed (1) the journals that can be targeted for effective search of publications related to smartphone addiction and/or submission of research output related to smartphone addiction, (2) the prominent/established authors that can be targeted for collaboration and/or trustworthy advice in smartphone addiction research, and (3) the publications that can serve as good starting points for researchers or prospective researchers to quickly pick up the knowledge needed to do smartphone addiction research. Furthermore, through co-word analysis conducted on the title and abstract of publications, this study has also identified and briefly elaborated on the research themes in the existing literature of smartphone addiction (i.e., protective/risk factors, neurobiological mechanisms, and intervention; prevalence and negative consequences; scale development and psychometric properties), which has provided an overview of what researchers have been doing in the existing literature of smartphone addiction. Taken together, this study provides researchers especially those new to the field of smartphone addiction a one-stop platform to gain an aggregative view of smartphone addiction literature.
Introduction
Smartphone is a mobile phone with advanced computing features. Since its introduction in 2007, it has penetrated into almost every aspect of people’s life. People nowadays do not only use smartphone for phone calls, text messaging, but also many other purposes such as information search, entertainment, shopping, and learning. On the other hand, smartphone is highly portable due to its tiny size. It can easily be carried along, which make it accessible for people anywhere and anytime. Due to the multifunctionality and accessibility of smartphone, people can easily spend great amount of time on using their smartphone. It may disrupt their social and/or work obligations and subsequently cause negative life consequences (Elhai et al., 2017). Although it is widely known that spending excessive time on using smartphone may disrupt social and/or work obligations and subsequently cause negative life consequences, people continue doing it (Elhai et al., 2017). This style of smartphone use is commonly termed “smartphone addiction,” which refers to the use or persistent use of smartphone despite the recognition that it can cause daily life disturbances or adverse consequences (Elhai et al., 2017). In most instances, smartphone addiction is characterized by symptoms that resemble to substance addiction (e.g., alcohol/drug), which include (1) withdrawal symptom (i.e., feeling anxiety, restless, irritability when smartphone is not available or usable), (2) disregard of adverse consequences (i.e., continue using despite the knowledge of having persistent or recurrent physical or psychological problem resulting from use or excessive use), (3) loss of control (i.e., repeated unsuccessful attempts to cut down use duration or frequency), and lastly (4) tolerance symptom (i.e., longer use duration or higher use frequency is required to achieve satisfaction; Panova & Carbonell, 2018).
Smartphone addiction has been found to be highly prevalent across countries especially among adolescents and young adults (Olson et al., 2022). Inspired by the high prevalence of smartphone addiction and the notion that smartphone addiction may cause negative life consequences, scholars have devoted great attention to the research of smartphone addiction (Khan & Khan, 2022). More specifically, scholars have attempted to (1) develop and/or validate scales for the assessment of smartphone addiction (D. Kim et al., 2014; Kwon, Kim, et al., 2013; Kwon, Lee, et al., 2013; Lin et al., 2014), (2) determine the prevalence of smartphone addiction (Cha & Seo, 2018; Lim et al., 2021; Okasha et al., 2021), (3) identify the risk and/or protective factors of smartphone addiction (Bian & Leung, 2015; Mahapatra, 2019), and lastly (4) identify the negative consequences associated with smartphone addiction (Mahapatra, 2019; Nayak, 2018; Samaha & Hawi, 2016).
The multiple research streams in the existing literature of smartphone addiction have made it necessary for researchers to have a review of what is currently known or not known about smartphone addiction (Khan & Khan, 2022). In this regard, some researchers have already conducted literature reviews on smartphone addiction (Busch & McCarthy, 2021; Harris et al., 2020; Osorio-Molina et al., 2021). For instances, Harris et al. (2020) have reviewed the measurement scales of smartphone addiction. Busch and McCarthy (2021) have provided a review of the antecedents and consequences of smartphone addiction. Osorio-Molina et al. (2021) have reviewed the risk factors and adverse effects of smartphone addiction among nursing students.
While these reviews have enriched the extant body of knowledge, they are fragmented and narrowly focused. More specifically, most of the existing literature reviews are systematic literature review (SLR; refer to Table 1 for more details). To identify the research themes in a given field, this type of literature review typically requires researchers to manually examine the full content of publications and thus tend to be less comprehensive in coverage (e.g., tens to low hundreds of articles). In such case, systematic literature review (SLR) has been widely argued by scholars to better suit for confined research area or narrow scope of review (Donthu et al., 2021). These have explained why most of the existing literature reviews have contents pertaining to only one or two aspects of smartphone addiction (e.g., risk and protective factors/prevalence/measurement scales/negative consequences; refer to Table 1 for more details), which make it difficult for researchers to have a complete overview of the literature of smartphone addiction.
Previous Literature Review on Smartphone Addiction.
There are also reviews of smartphone addiction literature conducted with meta-analysis (refer to Table 1 for more details). Scholars have repeatedly highlighted that meta-analysis is an approach entirely developed to synthesize multiple existing statistical evidences and draw an overall conclusion on a single aspect of certain phenomenon (e.g., prevalence of smartphone addiction) or relationship between two variables (e.g., relationship between personality traits and smartphone addiction). Thus, it is essentially unable to provide a thorough retrospective or assessment of smartphone addiction literature (Donthu et al., 2021). These circumstances have highlighted the need for a comprehensive review of the existing smartphone addiction literature, which can bring together the fragmented knowledge and serves as a one-stop platform that allows researchers to gain a complete overview of smartphone addiction literature (Khan & Khan, 2022). In this regard, bibliometric analysis seems to be the solution. Bibliometric analysis, benefits from the technology-enabled automated analysis of quantitative measures (e.g., citations and co-occurrences), tend to be more extensive in coverage (e.g., high hundreds to thousands of articles) than other types of reviews (e.g., systematic literature review, meta-analysis). It has also been widely argued and proven to be more effective in presenting a full sketch of a specific scientific field (Donthu et al., 2021; Gao et al., 2020; Goyal & Kumar, 2021).
Inspired by these, Khan and Khan (2022) have conducted bibliometric analysis with the aim to identify the publication trends, most productive/influential authors, publications, and journals as well as research themes in the literature of smartphone addiction. However, the analysis conducted by Khan and Khan (2022) suffers from two main limitations. Firstly, they have only used “smartphone addiction” and “problematic smartphone use” as their keywords for literature search. However, in fact, researchers in the field of smartphone addiction have also used other terms to represent the same construct including “compulsive smartphone use,”“excessive smartphone use,”“smartphone overuse,” and “smartphone dependence.” In such circumstances, it is reasonable to expect that they may miss out on some valuable research papers related to smartphone addiction. Secondly, they have only covered studies up to May 2021. However, since then, the literature of smartphone addiction has grown rapidly. As a result, many newer studies on smartphone addiction are not covered in their review.
Addressing the research gaps discussed above, the bibliometric analysis conducted in this study aims to provide researchers a one-stop platform to gain good understanding of the literature of smartphone addiction and address the following research questions (RQs):
RQ1: What are the publication trends in the literature of smartphone addiction?
RQ2: Which are the most productive or influential journals in the literature of smartphone addiction?
RQ3: Who are the most productive or influential authors in the literature of smartphone addiction?
RQ4: Which are the most influential publications in the literature of smartphone addiction?
RQ5: What are the major research themes in the literature of smartphone addiction?
Methodology
Data Collection and Cleaning
To conduct bibliometric analysis on the literature of smartphone addiction, bibliographic data first needs to be retrieved from bibliographic databases. Scholars have pointed out that “the number of bibliographic databases is high (e.g., PubMed, EMbase, SpringerLink, etc.), but not all of them provide information that allows easily performing bibliometric analyses” (Moral-Muñoz et al., 2020, p. 3) and “existing software can only process documents from some databases (Gan et al., 2022, p. 2). For examples, Vosviewer can only analyze bibliography data downloaded from Web of Science (WoS), Scopus, and PubMed while SciMat can only analyze bibliographic data downloaded from Web of Science (WoS) and Scopus (Gan et al., 2022).
Among various bibliographic databases, due to the fact that “most existing software has been developed for WoS and Scopus databases” (Gan et al., p. 2) and “almost all software tools and libraries can import data downloaded from WoS and Scopus” (Moral-Muñoz et al., 2020, p. 8), Web of Science (WoS) and Scopus have been the two most employed databases for bibliometric analysis (Caputo & Kargina, 2022; Singh et al., 2021). Although both databases provide comprehensive coverage of academic journals and allows efficient retrieval of complete and accurate bibliographic information needed for bibliometric analysis (e.g., citation information, abstract, keywords, author details, references of each article), it should be noted that Scopus has been found to have a wider coverage of academic journal than Web of Science (WoS; Falagas et al., 2008; Singh et al., 2021). For examples, Falagas et al. (2008) have found that Scopus has approximately 20% wider coverage of academic journals than Web of Science (WoS) database.
Apart from the two databases noted above, Google Scholar, Dimensions, and PubMed have also been employed by previous bibliometric reviews (AlRyalat et al., 2019; Falagas et al., 2008; Singh et al., 2021). However, it should be noted that (1) the bibliographic data provided by Google Scholar is often inaccurate or incomplete (Falagas et al., 2008), (2) Dimensions include preprints or full draft research papers that are shared publicly before it has been peer reviewed (Singh et al., 2021), and (3) PubMed has a narrow coverage of academic journal because it focuses only on life sciences and biomedical disciplines (AlRyalat et al., 2019). Thus, these databases (i.e., Google Scholar, Dimensions, PubMed) may not be credible or reliable data sources for bibliometric analysis. Building on the discussions above, Scopus has been widely recommended by numerous researchers for the extraction of data needed for bibliometric analysis (Kumar et al., 2022; Nagariya et al., 2020).
Although Scopus has been usually considered as one of the largest abstract and citation databases of peer-reviewed literature, scholars have pointed out that its journal coverage differs substantially from the other databases (e.g., Web of Science, Dimensions; Caputo & Kargina, 2022). For examples, Singh et al. (2021) have found that (1) only 33.93% of journals indexed in Scopus (N = 13,489) overlap with those indexed in Web of Science (WoS) and (2) only 17.78% of journals indexed in Dimensions (N = 13,149) overlap with those indexed in Web of Science (WoS). Accordingly, scholars have recommended merging the bibliographic data downloaded from several databases because it is likely to result in more comprehensive bibliometric analysis (Caputo & Kargina, 2022).
However, due to the fact that different databases have different format of bibliographic data, some scholars have explicitly pointed out that merging bibliographic data downloaded from different databases is a difficult and error-prone task. It requires researchers to have coding skills or knowledges that are beyond the skill sets of many researchers especially those in the field of social sciences (Caputo & Kargina, 2022). Also, the merge of bibliographic data downloaded from different databases requires access to several software packages such as Endnote, R studio, Word and Excel VBA macros, which may not be widely and freely accessible. As a result, the majority of scholars tend to use only one database to extract bibliographic data for their bibliometric analysis (Caputo & Kargina, 2022). From the same vein, the recommendation of Donthu et al. (2021, p. 293) is “to settle on one appropriate database to mitigate the need for that consolidation, as minimizing unnecessary action items can help to mitigate potential human errors.” Inspired by the notions highlighted above, this study has opted Scopus as the only database to extract the bibliographic data needed for bibliometric analysis in this study (e.g., citation information, abstract, keywords, author details).
With regard to the identification of literature search terms, Donthu et al. (2021, p. 293) have highlighted that researchers should “consult the literature to identify a relevant combination of search terms.” Accordingly, this study has consulted the literature of smartphone addiction and found that there were other terms used by researchers to represent the same construct such as “problematic smartphone use” (Elhai et al., 2017), “compulsive smartphone use” (Panda & Jain, 2018), “excessive smartphone use” (Shen & Wang, 2019), smartphone overuse (Kim et al., 2019), “smartphone dependence” (J. L. Wang et al., 2020), “problematic mobile phone use” (Li et al., 2021), and “mobile phone addiction” (Gao et al., 2018).
However, scholars have pointed out that “problematic mobile phone use” and “mobile phone addiction” are different from “smartphone addiction.” They represent a broader construct than smartphone addiction (Duke & Montag, 2017; Montag et al., 2015). More specifically, the term “mobile phone” covers both older cellular phone (those without operating system and access to internet and many other multimedia contents such as online games, YouTube, social media) and smartphone. Accordingly, when comes to the study of smartphone addiction, scholars have argued that it would be good for researchers to use smartphone-specific terms (e.g., smartphone addiction, problematic smartphone use) and employ measurement scale specifically tailored for smartphone addiction (Bian & Leung, 2014; D. Kim et al., 2014; Rozgonjuk et al., 2016). These have explained why some researchers have insisted the use of term “smartphone” rather than “mobile phone” in their study of smartphone addiction (Bian & Leung, 2014). For example, Bian and Leung (2014, p. 9) have modified the item “I have attempted to spend less time on my mobile phone but am unable to” to be “You have attempted to spend less time on your smartphone but are unable to.” Building on the notions highlighted above, given the fact that the focus of this bibliometric review was on smartphone addiction, the keywords (“problematic smartphone use” OR “smartphone addiction” OR “compulsive smartphone use” OR “excessive smartphone use” OR “smartphone overuse” OR “smartphone dependence”) were used in combination with Boolean operators (e.g., OR) to search in the title, abstract, and keywords of articles.
As the first ever smartphone was introduced in 2007, this study has only included studies published after 2007. More specifically, this study has limited the time span of its literature search to 2008 to 2022 and thus studies published before 2008 were excluded. Such approach was expected to greatly exclude studies related to the use of older cellular phone (those without operating system and access to internet and many other multimedia contents such as online games, YouTube, social media; Elhai et al., 2017; Marino et al., 2021). The search was further refined to include only articles written in English. Articles written in other languages (e.g., Spanish, German, Portuguese, Italian, Turkish, Chinese, Korean, French, Hungarian, Bosnian, Czech, Japanese, Persian, Polish) were excluded, which has been a common practice in bibliometric studies (Nagariya et al., 2020). Furthermore, in order to ensure the quality of dataset, this study has included only articles published in peer-reviewed journal as they have already been evaluated by peers to be of a suitable quality for publications in academic journals and thus more likely to provide trustworthy research findings (Stehmann, 2020). Accordingly, publications of other types such as conference paper, review, book chapter, conference review, erratum, editorial, letter, note, data paper were excluded.
A literature search was conducted with the aforementioned criteria in Scopus database on 15 August 2022, which has resulted in 1,088 published articles. However, the data retrieved should not be directly analyzed because it is likely to contain studies that fall beyond the scope of smartphone addiction literature. Accordingly, in order to ensure the representativeness of sample data, this study has only included the studies with primary focus on smartphone addiction and excluded those studies that have only mentioned the term “smartphone addiction,”“problematic smartphone use,”“compulsive smartphone use,”“excessive smartphone use,”“smartphone overuse” or “smartphone dependence” in their main texts but primarily focused on other topics (e.g., internet addiction; social media addiction; problematic use of short-form video applications; phubbing behavior; fear of missing out; metacognitions about smartphone use; flow experience). For example, Huang et al. (2022) have repeatedly mentioned the term “smartphone addiction” and “problematic smartphone use” in their study but their core focus was on problematic use of short-form video applications. It was outside the scope of smartphone addiction literature and thus excluded from the dataset of this study.
To do these, the title, abstract, and keywords of each article were screened. Full text of articles was also accessed if there was still doubt on the relevance of studies after screening the title, abstract, and keywords. Building on the processes noted above, 164 articles were identified to fall beyond the scope of smartphone addiction literature and thus excluded from the dataset of this study. Accordingly, only 924 articles were taken as the samples for bibliometric analysis in this study.
A flow diagram of data collection was presented in Figure 1. After identifying the samples for bibliometric analysis of this study, bibliographic data corresponding to the samples (i.e., title of articles, names/affiliation of authors, abstract, keywords, references) were exported from Scopus database and saved as Comma-Separated Values (.csv) and Research Information System (.ris) file. It is important to note that the bibliographic data exported from Scopus database may contain erroneous entries. Specifically, the name of same authors or title of same journals may be inconsistent in the dataset. In this regard, scholars have highlighted that the inconsistencies of authors name and journal titles should be corrected in order to avoid distorting the results (Donthu et al., 2021; Zupic & Čater, 2015). Inspired by these, this study has ensured the name of same authors or title of same journals to be consistent and represented author or journal data under one specific form of representation if there was any inconsistency to assure accurate analytical findings.
Flow diagram of the data collection
Data Analysis
There are two major techniques for bibliometric analysis, namely (1) performance analysis and (2) science mapping analysis. Performance analysis is a descriptive technique that focuses on examining the performance of different research constituents (authors, journals, articles) in a given field. Two most prominent measures in performance analysis are the quantity of publications and quantity of citations, which respectively reflect the productivity and impact/influence of research constituents (authors, journals, articles, countries). There are also performance indicators that combine the quantity of publications and quantity of citations such as number of cited publications, and citation per publication (Donthu et al., 2021). The aforementioned performance indicators were applied in this study to provide insights on (1) the current publication trends (RQ1), (2) the most productive and influential journals (RQ2), (3) the most productive and influential authors (RQ3), (4) the most influential publications (RQ4). Performance indicators used to address each of the research questions (RQs) were highlighted in detailed in Figure 2. With regard to the software applied in the performance analysis of this study, Harzing’s Publish and Perish was deployed to compute various citation metrics (i.e., citation per publication, number of cited publications), whereas Microsoft Excel was used to conduct frequency analysis on the quantity of publications/citations (i.e., total number of publications/citations by year) and generate corresponding graphical representations.
Data analysis methods
On the other hand, science mapping analysis focuses on identifying the relationship between research constituents (e.g., keywords; Donthu et al., 2021). Among various science mapping techniques, co-word analysis was conducted in this study on the title and abstract of publications in smartphone addiction literature. This technique assumes that subject terms that frequently appear together are likely to be talking about same topic or have a thematic relationship with each other. Accordingly, by identifying and clustering the subject terms that often appear together, researchers would be able to identify research themes in a specific scientific field (Donthu et al., 2021). Co-word analysis has been widely used and proven to be effective in uncovering the research themes in the field of medical tourism (De la Hoz-Correa et al., 2018), environmental crisis management (Dai et al., 2020), child–computer interaction (Giannakos et al., 2020), and knowledge-hiding (Bernatović et al., 2022). Inspired by these, co-word analysis was applied in this study to uncover the research themes in the existing literature of smartphone addiction (RQ5).
Co-word analysis in this study was carried out with Vosviewer, a software designed to visualize the connections among subject terms. In the network visualization of Vosviewer (refer to Figure 4), all items are circled and represented by their respective labels. The size of the circles and labels (font size) is determined by the weight of items, that is, the number of occurrences of subject terms in the title and/or abstract of publications. The higher the occurrences of subject terms in the title and/or abstract of publications, the larger the circles and labels of the items. It also shows the strength of relation between subject terms. Specifically, the higher the number of co-occurrence of two subject terms in the title and/or abstract of publications, the stronger the link strength and the thicker the line connecting the two subject terms. More importantly, Vosviewer uses different colors to represent different research cluster. Accordingly, the more colors, the more research clusters are identified (Van Eck & Waltman, 2011). These are particularly relevant for the objective of identifying the research themes in the existing smartphone addiction literature.
Findings and Discussions
Publication Trends
In bibliometric analysis, the developmental trend of a scientific field is often represented by the quantity of publications (Goyal & Kumar, 2021). In this regard, Figure 3 shows the historical trajectory of publication quantity in the research field of smartphone addiction. The number of publications in the field of smartphone addiction has been increasing over the years. From just 2 articles published in 2011 to 268 articles published in 2021. With regard to the growth trajectory of publication quantity, this study has observed that the growth of publication quantity in the period between 2011 and 2017 was slow and accelerated after 2017. Furthermore, it was seen to have sharp surge starting from 2020 and recorded highest number of publications in 2021 with 286 publications.
Number of publications
Although Figure 3 has shown that there was a drop of publication quantity in 2022, it should not be considered as an actual research output reduction in the literature of smartphone addiction because the literature search for this study was conducted on August 15, 2022 and thus the publication quantity in Figure 3 could not represent the actual overall publication quantity in 2022 in smartphone addiction literature. However, building on the overall upward growth trend observed for the publication quantity in the literature of smartphone addiction, this study is optimistic that the publication quantity in 2022 is likely to surpass the publication quantity in 2021 and continue growing in the coming years. Furthermore, it is also worth to point out that the majority of the articles were published in the period between 2020 to 2022 August, which accounts for approximately 67% of total publications in smartphone addiction literature. These have reflected that smartphone addiction is a relatively new and emerging field of research that has recently started to gain popularity and expected to gain more and more attention from academic community in the coming years.
Most Productive and Influential Journals
Building on the claim that “different journals are most influential in different subareas” (Baumgartner & Pieters, 2003, p. 123), researchers have argued that it is essential to identify the publication outlets or journals that are productive and influential in a specific scientific field because it provides a guide for researchers especially those new to that scientific field about which journals to target for effective search of related publications and the potential good destinations for the submission of research paper in that scientific field. The 924 articles considered in this study were published in 361 journals. Table 2 shows the top 15 journals with highest number of total publications (TP), along with their total citations (TC) accumulated, average citation received per publication (TC/TP), and Scimago Journal Ranking (SJR; Quartile). These journals have been considered as the hotspot of publications related to smartphone addiction. Specifically, 394 articles were published by the top 15 journals, which had accounted approximately 43% of the total outputs in the literature of smartphone addiction.
Top 15 Productive and Influential Journals.
In which, International Journal of Environmental Research and Public Health has published 55 articles in total. It was followed by Computers in Human Behavior and Journal of Behavioral Addictions, which have respectively published 54 and 40 articles. These have led them to be identified as the three most productive journal in the literature of smartphone addiction. Remarkably, although International Journal of Environmental Research and Public Health was the journal with highest number of publications (TP), its total citations (TC = 383) and average citation per publication (TC/TP = 6.96) were much lower than those of Computers in Human Behavior (TC = 4,120; TC/TP = 76.30) and Journal of Behavioral Addictions (TC = 2,664; TC/TP = 66.60). These have reflected the fact that the articles published by International Journal of Environmental Research and Public Health have received significant lesser attention from researchers than those published by Computers in Human Behavior and Journal of Behavioral Addictions.
Another journal worth mentioning was PLoS One. Despite its lower total publications (TP) than the journals highlighted above, it has accumulated considerable number of total citations (TC = 2,398). Also, it was surprisingly found to have the highest average citation per publication (TC/TP = 88.81) among the top 15 journals publishing smartphone addiction research, which make it one of the most notable journals in the literature of smartphone addiction. Furthermore, apart from the four journals particularly highlighted above, the other journals listed in Table 2 were all worth noting. Specifically, despite their relatively low total publications (TP), total citations (TC), and average citation per publication (TC/TP), they were still ranked at Quartile 1 or 2 in Scimago Journal Ranking (SJR). Accordingly, they have been widely deemed as reliable and established journals that publish smartphone addiction research. Building on the discussions above, this study argues that the top 15 journals listed in Table 2 could all be deemed as good platforms for the search of publications related to smartphone addiction and good destinations for the submission of research output related to smartphone addiction.
Most Productive and Influential Authors
Researchers have claimed that it is valuable to identify the most productive and influential authors in a scientific field because it helps “equipping prospective scholars with valuable information to reach out and collaborate with established and trending scholars in the research field” (Donthu et al., 2021, p. 290). The 924 articles considered in this study were authored by 3,927 researchers. Table 3 lists the top 15 authors with highest number of total publications (TP), along with their accumulated total citations (TC), and average citation received per publication (TC/TP). These authors have been considered as the top dogs in the literature of smartphone addiction.
Top 15 Productive or Influential Authors.
Among them, Elhai, J.D. has published a total of 36 articles (TP). It was followed by Griffiths, M.D. and Kim, D.J., which have respectively published 33 and 27 articles (TP). These have made them to be identified as the most productive authors in the literature of smartphone addiction. Furthermore, the articles written by these authors were also widely cited in the existing literature of smartphone addiction. Specifically, Elhai, J.D. has accumulated a total of 2,040 total citations (TC). It was followed by Kim, D.J. and Griffiths, M.D., which have respectively accumulated 2,023 and 943 total citations (TC). These have led them to be identified as the most influential authors in smartphone addiction literature. Remarkably, Griffiths, M.D. has obtained 28.58 average citation per publication (TC/TP). It was much lower than those of Elhai, J.D. (TC/TP = 56.67) and Kim, D.J., (TC/TP = 74.93). It has reflected the fact that the articles written by Griffiths, M.D. have received significantly lesser attention than those written by Elhai, J.D. and Kim, D.J. However, this study argues that these may not be sufficient to cause any doubt to the prominent status of Griffiths, M.D. in the literature of smartphone addiction.
Another two authors worth highlighting were Hall, B.J. and Cho, H. Despite their lower number of publications (TP), they have respectively accumulated 979 and 749 total citations (TC). Also, their average citation per publication (TC/TP) were surprisingly found to be the largest (TC/TP = 97.90) and second largest (TC/TP = 83.22) among the top 15 authors that have published in smartphone addiction literature. These have reflected that the articles written by them have received great attention in the literature of smartphone addiction, laying their prominent status in the literature of smartphone addiction.
Apart from the five authors particularly highlighted above, the other authors listed in Table 3 were also worth noting. Although their overall contribution may not be comparable to the five authors previously highlighted, these authors have published considerable quantity of articles (TP) and also obtained sizeable number of total citations (TC) as well as average citation per publication (TC/TP). Accordingly, they have also been widely deemed as established and trending authors in the existing literature of smartphone addiction. Building on the discussions above, this study argues that the top 15 authors in the existing literature of smartphone addiction (as listed in Table 3) could all be considered as the top choices to reach out for collaborations and/or seek advice regarding smartphone addiction research.
Most Influential Publications
Researchers have repeatedly pointed out that the most cited publications are usually the seminal publications or publications that can provide rich and/or trustworthy information needed to do research in a particular scientific field (Gao et al., 2020). Accordingly, it has been argued that it is valuable for researchers especially those new to a particular scientific field to identify the most cited publications in that scientific field because these publications may be good starting points for them to quickly build up the knowledge needed to do research in that scientific field (Kumar et al., 2022). Table 4 lists the top 15 publications with highest number of total citations (TC) among the 924 publications considered in this study. Among these publications, one publication particularly worth highlighting was “Is smartphone addiction really an addiction?” (rank 9 in Table 4). It has extensively discussed the concept of smartphone addiction and questioned whether the behavior “persistent smartphone use despite the recognition that it can cause negative life consequences” should be conceptualized as an addiction or just a problematic way of smartphone use. This publication has been considered to be highly relevant for researchers especially those new to the field of smartphone addiction because it helps in understanding how smartphone addiction has been conceptualized in the existing literature (Harris et al., 2020).
Top 15 Top Cited Publications.
Another four of them (rank 1, 3, 10, and 15 in Table 4) have focused on developing and validating the measurement scales for smartphone addiction. The scales developed and validated in these publications were ““Smartphone Addiction Scale Short Version (SAS-SV), “Smartphone Addiction Scale” (SAS), Smartphone Addiction Inventory (SPAI), and Smartphone Addiction Proneness Scale (SAPS) respectively. These scales have been widely used to measure smartphone addiction in the existing literature (Harris et al., 2020; Yu & Sussman, 2020), which could be partly reflected by the sizeable number of citations (TC) they have accumulated (refer to Table 4 for the exact number of total citations). Among the four scales previously highlighted, the total citations (TC) of “Smartphone Addiction Scale Short Version (SAS-SV) and “Smartphone Addiction Scale” (SAS) were the highest and second highest respectively. Furthermore, they have also been found to be the two most frequently used scales to measure smartphone addiction in the existing literature (Yu & Sussman, 2020). Building on these, this study argues that the four scale previously discussed especially “Smartphone Addiction Scale Short Version (SAS-SV) and “Smartphone Addiction Scale” (SAS) are particularly worth noting for researchers when comes to the measurement of smartphone addiction.
The others have focused on the identification of the risk/protective factors of smartphone addiction (rank 4, 6, 7, 11, 12, 13, and 14 in Table 4) and negative consequences associated with smartphone addiction (rank 2, 5, and 8 in Table 4). These publications have accumulated considerable number of citations (TC) and been widely considered as highly influential publications in their respective field of investigations (e.g., risk/protective factors; negative life consequences; scale development/validation; conceptualization) in the existing literature of smartphone addiction. Building on the discussions above, this study argues that these publications could all be considered as good starting points for researchers especially those new to the field of smartphone addiction to quickly pick up the necessary knowledge for smartphone addiction research.
Major Research Themes
Co-word analysis was carried out through Vosviewer on the title and abstract of publications to uncover the major research themes in the existing literature of smartphone addiction (RQ5). With full counting method, it has initially identified a total of 13,837 terms from the title and abstract of the 924 publications considered in this study. Although it is ideal to provide as much information on research themes as possible, it is not always practically viable or manageable to visualize a very large number of terms and their relationship (Zupic & Čater, 2015). Accordingly, this study has only considered terms with a minimum five occurrences in the title and abstract of 924 publications considered in this study and revealed 1,572 terms meeting this threshold.
It should be noted that terms with same meaning may have different forms of representation. Accordingly, as advised by Zupic and Čater (2015) and Donthu et al. (2021), this study has represented all terms with same meaning under one label in order to avoid inaccurate analytical findings. For examples, the terms “problematic smartphone use,”“compulsive smartphone use,”“smartphone overuse,”“addictive smartphone use,” and “excessive smartphone use” were all represented by “smartphone addiction.” The terms “depression severity,”“depression symptom,”“depression symptom severity” and “depressive disorder” and “depressive symptom” were all represented by “depression.” Furthermore, this study has also aggregated the terms usually perceived to be under same umbrella. For examples, the terms “life stress,”“academic stress,”“interpersonal relationship stress,”“job stress,”“social stress,”“perceived stress” were all aggregated as “stress.” The terms “sleep disturbance,”“sleeping disorder,”“sleep disorder,”“sleep interference,”“insomnia disorder,”“insomnia symptom,”“insomnia,”“poor sleep quality,”“poorer sleep quality,” and “poor sleep” were all aggregated as “sleep problem.” In addition, this study has also excluded numerous terms that are too general and cannot reflect research themes in the literature of smartphone addiction such as “university,”“potential,”“personal,”“number,” and “missing.”
After going through the aforementioned processes, only 48 terms were finally retained for science mapping analysis. Three research clusters were identified among these terms that include (1) protective/risk factors, neurobiological mechanisms, and intervention, (2) prevalence and negative consequences, and (3) scale development/psychometric properties test, which were respectively represented by red, green, and blue color in the network visualization of this study (refer to Figure 4). Other key takeaways from the network visualization of this study were that, in the title and abstract of publications, terms related to protective/risk factors (e.g., depression, anxiety, stress), negative consequences (e.g., sleep problem, academic performance, physical activity), and prevalence appear much more often and co-occur much more often with the term “smartphone addiction (thus higher link strength in Vosviewer) than those related to intervention and neurobiological mechanism (as illustrated in Table 5).
Term co-occurrences network visualization.
Research Clusters and Their Corresponding Terms.
These could also be reflected by the fact that, in the network visualization of this study (refer to Figure 4), (1) the size of the circles and labels (font size) of terms related to protective/risk factors, negative consequences, and prevalence was much bigger than the terms related to intervention and neurobiological mechanism (occurrences) and (2) the lines connecting terms related to protective/risk factors, negative consequences, and prevalence to the term “smartphone addiction” were much thicker than those of intervention and neurobiological mechanism (co-occurrences and/or link strength). These have all reflected that (1) studies identifying protective/risk factors, negative consequences, and prevalence have thus far dominated the existing literature of smartphone addiction and (2) the investigation of neurobiological mechanisms underlying smartphone addiction and formulation/test of intervention programs have received relatively scant attention from researchers. In order to provide more insights into the research themes of the existing literature of smartphone addiction, each of the previously identified research clusters was further elaborated in the following subsections.
Protective/Risk Factors, Neurobiological Mechanisms, and Intervention
The first focal point of this research cluster was to identify protective/risk factors of smartphone addiction. In this regard, in the existing literature, various individual/contextual factors were examined as the protective/risk factors of smartphone addiction such as psychological characteristics (e.g., stress, anxiety, depression, self-esteem; Elhai et al., 2017; J. L. Wang et al., 2020), individual traits (e.g., personality trait, impulsivity, self-control; Contractor et al., 2017; Horwood & Anglim, 2018; Servidio, 2021), and interpersonal relationship (e.g., family relationship, peer relationship; Ouyang et al., 2020; P. Wang et al., 2017). For examples, Peng et al. (2022) has found that perceived stress is positively related to smartphone addiction (risk factor) and self-control negative moderate the relationship (protective factor).
On the other hand, researchers have also devoted their attention to the neurobiological mechanisms underlying smartphone addiction (Horvath et al., 2020; Seo et al., 2020; West et al., 2021). In this regard, researchers have argued that addictive behavior is mainly characterized by an imbalance between increasing urges/desires to engage in certain behavior and decreasing inhibitory control over these urges/desires (Brand et al., 2019; Dong & Potenza, 2014). Accordingly, some studies have applied neuro imaging techniques (e.g., electro-encephalography or structural/functional magnetic resonance imaging) to record the structure or activity of brain regions associated with inhibitory control and reward/loss sensitivity (e.g., anterior cingulate cortex and orbitofrontal cortex) in the settings of experimental task (e.g., go/no-go task/lowa gambling task; Horvath et al., 2020; West et al., 2021). For examples, in the setting of gain and loss guessing tasks, Horvath et al. (2020) has found that smartphone addicts demonstrate stronger reward sensitivity (increased activation in orbitofrontal cortex) and weaker loss sensitivity (reduced activation in anterior cingulate cortex). These may explain why smartphone addicts often (1) show more preference to immediate gratifying experiences (e.g., using smartphone for gaming) than long term gains (e.g., using the time spent gaming to perform tasks associated with long term career success) and (2) unable to inhibit their impulse to use or continue using smartphone for immediate gratification, even they have already recognized that it can cause adverse consequences.
These studies have been repeatedly highlighted to be highly valuable because they provide insights into how to reduce the likelihood/incidence of smartphone addiction (Busch & McCarthy, 2021; Horvath et al., 2020; Osorio-Molina et al., 2021), which can serve as inputs for the formulation and development of intervention/corrective strategies of smartphone addiction. Apart from these, building on the risk/protective factors and neurobiological mechanisms identified by previous studies, a stream of studies has focused on identifying/developing intervention programs for smartphone addiction such as cognitive behavioral therapy (Lan et al., 2018; Seo et al., 2020), exercise intervention programs (e.g., taichi, basketball, badminton, dance, run, and bicycle; Choi et al., 2020; Xiao et al., 2021), and digital intervention (e.g., smartphone addiction management system; Lee et al., 2022) as well as testing their effectiveness on treating/correcting smartphone addiction. For examples, Lan et al. (2018) has highlighted that cognitive behavioral therapy can help treating smartphone addiction because it improves the ability to inhibit the impulse to continue using smartphone when users recognize its associated adverse consequences and found that cognitive behavioral therapy has successfully reduced the score of Mobile Phone Internet Addiction Scale (MPIAS) in their intervention group.
Taken together, with the aim to provide information about how to reduce the likelihood/incidence of smartphone addiction, this research cluster focuses on (1) identifying the risk/protective factors, (2) neurobiological mechanism underlying smartphone addiction, and (3) intervention/corrective measures of smartphone addiction.
Prevalence and Negative Consequences
The first focal point of this research cluster was on identifying the prevalence of smartphone addiction. Through the use of smartphone addiction measurement scales such as smartphone addiction scale-short version (SAS-SV; Kwon, Lee, et al., 2013), malay version of smartphone addiction scale (SAS-M; Ching et al., 2015), or smartphone addiction scale (SPAS; Bian & Leung, 2014), numerous studies have examined the prevalence of smartphone addiction in order to identify whether smartphone addiction is an issue that really exists among the community (Bian & Leung, 2014; Lim et al., 2021; Okasha et al., 2021).
Although these studies have usually found that smartphone addiction is prevalent across countries especially among emerging adults and adolescents, it should be noted that the source of their data was unitary, or more specifically, their data was collected only from a specific group of objects. For examples, Cha and Seo (2018), Lim et al. (2021), and Okasha et al. (2021) have respectively collected data only from middle school students in South Korea, patients with depression in Malaysia, and university students in Egypt. Accordingly, the findings generated by these studies may not be sufficient to support the idea that smartphone addiction is a global public health concern. These have partially explained why there has been continuous effort among researchers in the field of smartphone addiction to identify the prevalence of smartphone addiction with survey data collected from samples with different characteristic around the world (Olson et al., 2022).
Another focus of this research cluster was on the identification of adverse consequences associated with smartphone addiction. In this regard, studies have examined the relationship of smartphone addiction with a broad range of adverse consequences such as poor academic/job performance (Hessari & Nategh, 2022; Nayak, 2018), poor sleep quality (Chen et al., 2017), reduced physical activity (S. E. Kim et al., 2015), fatigue (Sert et al., 2019), and musculoskeletal problems (Soliman Elserty et al., 2020). The main motivation of these investigations was to identify whether smartphone addiction is threatening or disturbing to human daily life (Busch & McCarthy, 2021).
Taken together, with the aim to identify (1) whether smartphone addiction is an issue that really exists among the community and (2) whether smartphone addiction is threatening or disturbing to human daily life, this research cluster focuses on examining the prevalence of smartphone addiction and its associated adverse consequences. These have been particularly highlighted to be valuable because they provide evidences that may support the idea that smartphone addiction is already a public health issue that deserves more attention and reflect the need to develop intervention measures for smartphone addiction (Olson et al., 2022).
Scale Development and Psychometric Properties
This research cluster mainly focuses on developing scales to measure smartphone addiction and examining their psychometric properties. In this regard, since smartphone addiction has been generally conceptualized as a behavioral addiction, studies developing scale to measure smartphone addiction have often transposed the criteria established to diagnose other addictive behavior (e.g., gambling, substance use) to measure smartphone addiction (e.g., tolerance, withdrawal, relapse, preoccupation, difficulty in control; D. Kim et al., 2014; Kwon, Lee, et al., 2013; Lin et al., 2014). After establishing the criteria to measure smartphone addiction, these studies have conducted factor analysis (exploratory and/or confirmatory factor analysis) to examine the psychometric properties of the scales developed by them (content validity, convergent validity, discriminant validity, internal consistency reliability, test-retest reliability) in order to find out how well these scales measure or reflect smartphone addiction (D. Kim et al., 2014; Kwon, Lee, et al., 2013; Lin et al., 2014).
Although most of the existing measurement scales have demonstrated adequate reliability and validity in the original studies that have developed them, the source of data involved in their development was unitary (Kwon, Lee, et al., 2013; Lin et al., 2014). For examples, the survey data involved in the development of smartphone addiction scale-short version (SAS-SV) and smartphone addiction inventory (SPAI) was respectively collected only from high school students in South Korea (Kwon, Lee, et al., 2013) and university students in Taiwan (Lin et al., 2014). In this regard, researchers have repeatedly highlighted that samples with different characteristics (e.g., country, culture, age, education level, occupation) may interpret and/or respond differently to same set of questions and thus a particular measurement scale may demonstrate different psychometric properties in different contexts (Harris et al., 2020).
These have partly explained why numerous studies have validated the existing measurement scales of smartphone addiction with survey data collected from samples with different characteristics (e.g., country, occupation, developmental stage) and found inconsistent factor structure (Sfendla et al., 2018; Vintilă et al., 2021). For examples, smartphone addiction scale (SAS) was developed based on the survey data collected from Korean participants and found to have six-factor structure (Kwon, Lee, et al., 2013). However, it was found to have seven-factor structure in Turkish context (Demirci et al., 2014), five-factor structure in Romanian context (Vintilă et al., 2021), and five-factor structure in Brazilian context (Andrade et al., 2021).
To conclude, this clusters mainly focuses on developing scales to measure smartphone addiction and identify whether these scales are reliable and valid forms of smartphone addiction measurement.
Conclusion
This study is among the first to provide holistic review of smartphone addiction literature through bibliometric analysis. Firstly, this study has examined the publication trend of smartphone addiction literature through tracking the historical trajectory of its publication quantity. It has revealed that the number of publications has been increasing in the period of 2011 to 2022. However, the growth was slow in the period of 2011 to 2017 and only accelerated after 2017, which was subsequently pushed to its peak in 2021 (286 publications). This study has also highlighted that most of the articles (approximately 67%) were published in the period between 2020 and 2022. These have together reflected that smartphone addiction is a relatively new and emerging field of research that has recently started to gain popularity, which is expected to stay relevant and gain more and more attention from academic community in the coming years.
Through the use of various performance indicators (e.g., total number of citation/publication, citation per publication), this study has also identified the (1) most productive/influential journals, (2) most productive/influential authors, and (3) the most influential publications in the literature of smartphone addiction. These are respectively expected to reveal (1) the journals that can be targeted for effective search of publications related to smartphone addiction and/or submission of research output related to smartphone addiction, (2) the prominent/established authors that can be targeted for collaboration and/or trustworthy advice in smartphone addiction research, and (3) the publications that can serve as good starting points for researchers or prospective researchers to quickly pick up the knowledge needed to do smartphone addiction research. Apart from these, this study has also identified and elaborated on the research themes that exist in the current literature of smartphone addiction. It is expected to provide researchers a one-stop overview of what research has already been done in the existing smartphone addiction literature (e.g., protective/risk factors, neurobiological mechanisms, and intervention; prevalence and negative consequences; scale development/psychometric properties test) and enable researchers to position the contribution of their own research against the established research themes. Furthermore, the co-word analysis conducted in our review has also revealed that (1) studies identifying protective/risk factors, negative consequences, and prevalence have thus far dominated the existing literature of smartphone addiction and (2) the investigation of neurobiological mechanisms underlying smartphone addiction and formulation/test of intervention programs have received relatively scant attention from researchers. These have enabled scholars to gain a rough insight of the knowledge gaps in the current literature of smartphone addiction and identify a general direction for future research. Taken together, this study provides a one-stop platform for researchers (especially those new to the field) to quickly and largely grasp the information needed to conduct smartphone addiction research (i.e., publication trends, top journals that publish smartphone addiction research, prominent authors, seminal/highly cited publications, existing research themes).
Despite the contribution to the literature of smartphone addiction, this study is not without limitations. Firstly, the dataset for bibliometric analysis in this study is retrieved only from Scopus. Although it is among the largest scientific databases, it may not contain all publications related to smartphone addiction. Secondly, this study has only considered journal article published in English. Accordingly, this study is likely to miss out some valuable publications such as papers listed only in other databases (e.g., web of science), non-English publications, conference proceedings, and/or books chapters.
Lastly, co-word analysis identifies research themes only through clustering “terms” that co-occur in the title and/or abstracts of publications (e.g., smartphone addiction and risk factor, smartphone addiction and protective factor, smartphone addiction and psychometric property) and does not provide in-depth information regarding the full content of publications (Nagariya et al., 2020). It has implied that the information (e.g., topical focus, research design, sample characteristic, sampling method, data collection/analysis process, findings) needed to identify the specific limitations of existing studies and specific areas that require further scholarship in the literature of smartphone addiction was not extracted. In such case, this study may not be able to provide clear or specific indication of potential avenue for future research of smartphone addiction.
Accordingly, future bibliometric review aiming to identify the directions for future research of smartphone addiction could complement their findings (e.g., identification of research themes) with other literature review techniques that enable them to examine the full content of smartphone addiction publications (e.g., systematic review). For examples, future bibliometric review could first identify research themes through clustering “terms” that co-occur in the title and/or abstracts of publications (e.g., smartphone addiction and risk factor, smartphone addiction and protective factor, smartphone addiction and psychometric property; i.e., co-word analysis). Subsequently, systematic review (SLR) could be employed to identify and locate studies related to each of the identified research themes and further examine their full content. Such complementary approach first enables researchers to efficiently identify the major research themes from large volume of unstructured smartphone addiction literature with technology-enabled automated analysis of quantitative measures (e.g., occurrences/co-occurrences of terms) and subsequently allow them to manually extract the information needed (e.g., topical focus, research design, sample characteristic, sampling method, data collection/analysis process) to identify the specific limitations of existing smartphone addiction studies and specific areas that require further scholarship in the literature of smartphone addiction.
Footnotes
Acknowledgements
We are grateful to the anonymous reviewers for their insightful suggestions and constructive comments. We deeply thank the editors for their patient work on our manuscript.
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) received no financial support for the research, authorship, and/or publication of this article.
Data Availability Statement
Data availability is not applicable to this article as no new data were created or analysed in this study.