This is the accepted version of the journal article: Panova, Tayana; Sánchez Carbonell, Xavier; Chamarro Lusar, Andrés; [et al.]. «La investigación del Internet Addiction Test desde una perspectiva intercultural : España, Estados Unidos y Colombia». Adicciones, Vol. 33 Núm. 4 (2021), p. 307-318. 27 pàg. DOI 10.20882/adicciones.1345 This version is available at https://ddd.uab.cat/record/254865 under the terms of the license 1 La investigación del Internet Addiction Test desde una perspectiva intercultural: España, Estados Unidos y Colombia Internet Addiction Test research through a cross-cultural perspective: Spain, USA and Colombia Tayana Panova*; Xavier Carbonell**; Andres Chamarro***; Diana Ximena Puerta-Cortés**** *Ramon Llull University Address: Passeig de Sant Gervasi, 47, 08022 Barcelona, Spain **Ramon Llull University Address: Passeig de Sant Gervasi, 47, 08022 Barcelona, Spain ***Autonomus University of Barcelona Address: B Building, Campus of Bellaterra, 08193 Cerdanyola del Valles, Barcelona, Spain **** University of Ibagué Carrera 22 Calle 67 B/Ambalá Ibagué, Colombia Send Correspondence to: Tayana Panova; Ginebra 25, Barcelona; tayana.panova@gmail.com “Note: This is the author’s copy of the version published in the journal Adicciones, 33(4), 307-318 (2021).” 2 Resumen español Los usuarios de Internet frecuentemente experimentan problemas relacionados con su uso de Internet y, aunque la cultura tiene una influencia importante en la forma en que las personas se comunican, en lo que valoran y, por lo tanto, en cómo utilizan Internet, las publicaciones sobre investigación intercultural del uso problemático de Internet son escasas. La herramienta más común para medir dicho uso, el Internet Addiction Test (IAT), se ha utilizado en varios países, por lo que en este estudio comparamos los resultados interculturales de las investigaciones más recientes sobre el análisis factorial del IAT. Encontramos que, en países con dos o más estudios, los resultados a menudo se replican, lo que sugiere que el contexto cultural influye en los comportamientos en Internet. Luego realizamos nuestros propios estudios de análisis factorial del IAT en tres países (España, EE. UU y Colombia) con 1,273 participantes. Al comparar nuestros resultados con los de estudios previos en esos países, encontramos que nuestros resultados también fueron similares a los de estudios anteriores. El hallazgo más notable fue que todos los análisis factoriales de IAT que realizamos y los anteriores en las mismas regiones contenían un factor relacionado con los problemas de gestión del tiempo y otro factor relacionado con problemas emocionales/psicologicos, lo que sugiere que el control de impulsos y las necesidades emocionales no satisfechas son componentes importantes en el desarrollo del uso problemático de Internet en todo el mundo. La investigación futura sobre el uso problemático de Internet debería centrarse en estos aspectos. Keywords: Adicción a Internet, Uso problemático de Internet, Intercultural, Prueba de adicción a Internet, análisis factorial English Abstract Internet users around the world often experience problems related to their Internet use, and although culture has an important influence over how people communicate, what they value, and therefore, how they use the Internet, there has been little cross-cultural research on the subject of problematic Internet use. The most common tool for measuring such use, the Internet Addiction Test (IAT), has been used in various countries, therefore in the study we compared and analyzed the cross-cultural results found in the most recent IAT factor analysis research. We found that in countries with two or more studies conducted; results are often replicated, suggesting the cultural context influences Internet behaviors. We then conducted our own IAT factor analysis studies in three countries – Spain, USA, and Colombia – with 1,273 participants. We compared our results with those from previous studies in the same countries and found that the results were similar. The most notable finding was that all the IAT factor analyses we conducted and the previous ones in the same regions contained a factor related to time management problems and another factor related to emotional/psychological problems, thereby suggesting that impulse control problems and unfulfilled emotional needs are the most important components in the development of problematic Internet use around the world. Future research on problematic Internet use should focus on these aspects. Keywords: Internet addiction, Problematic Internet Use, Cross-Cultural, Internet Addiction Test, factor analysis 3 Although Internet use is a global phenomenon, there has been little research on the topic of problematic Internet use from a cross-cultural perspective. This is a weakness in the literature that should be addressed because people from different cultural backgrounds have very different communication practices, values and motivations, and therefore have different Internet behaviors as well. Consequently, when they experience problems associated with Internet use, the nature of those problems may also differ depending on their sociocultural context. Psychometric analyses such as factorial analysis of Internet addiction questionnaires show different results in almost every country studied, and the majority of studies comment that this variation probably exists in part due to cultural differences. However, very few studies have taken a cross-cultural approach to the study of Internet addiction and little attention is given to culture in the problematic Internet use research in general. Taking a cross-cultural approach to this subject could help identify culture’s influence on problematic Internet behaviors, which would facilitate the development of customized evaluation tools and treatment practices for problematic Internet users in diverse populations. Problematic Internet Use and Internet Addiction Problematic Internet Use (PIU) is an important problem to study as half the world’s population uses the Internet regularly and the popularization of the smartphone has made Internet access even easier and more frequent (Stevens, 2018). Studies have shown that Internet Addiction (IA) is associated with disorders such as: anxiety (Ho et al., 2014; Lee & Stapinski, 2012; Younes, Halawi, Jabbour, El Osta, Karam, Hajj & Khabazz, 2016), depression (Orsala, Orsalb, Unsalc & Ozalp, 2013; Younes et al., 2016), stress (Pedrero-Pérez et al., 2018; Samaha & Hawi, 2016; Younes et al., 2016), low self-esteem (Bahrainian, Alizadeh, Raeisoon, Hashemi Gorji, & Khazaee, 2014; Bozoglan, Demirer & Sahin, 2013), loneliness (Bozoglan et al., 2013; Yao & Zhong, 2014), insomnia (Chen & Gau, 2016; Younes et al., 2016), suicidality (Lin et al., 2014), impulsivity (Lee, Choi, Shin, Lee, Jung & Kwon, 2012), substance abuse (Ho et al., 2014; Lee, Han, Kim & Renshaw, 2015) and ADHD (Ho et al., 2014; Weinstein, 4 Yaacov, Manning, Danon & Weizman, 2015), among others. That being said, it is still under discussion whether problematic Internet use can be labeled as an addiction (Kardefelt-Winther, 2014; Sánchez- Carbonell, Beranuy, Castellana & Chamarro, 2008; Starcevic, 2013; Widyanto & Griffiths, 2006). Due to issues with the concept’s theoretical development, methodology and conceptualization, and because the levels of severity of problems associated with Internet “addiction” are usually not comparable with the severity of problems caused by traditional addictions. The focus on an addiction framework may have contributed to problems with diagnosis and treatment of problematic Internet use because the leading tool for diagnosis, the Internet Addiction Test, has unstable structural validity. The IAT was developed by Young (1998), based on the DSM-IV (American Psychiatric Association, 1994) criteria for pathological gambling and has been the most widely used measure for the study of problematic Internet use around the world. It was designed to have unidimensional structure, however, it has been found to have varying numbers of factors, ranging from 1 to 6 (Laconi, Rodgers & Chabrol, 2014). Although its reliability is consistently strong (Laconi, Rodgers & Chabrol, 2014; Panayides & Walker, 2012), its factorial structure differs in almost every study, thus making it difficult to identify which components of problematic Internet use are more relevant to address in diagnosis and treatment. Many of the studies in the IAT factor analysis literature mention the potential role of culture in the psychometric differences found across studies, however IAT research with a cross- cultural perspective is scarce. Culture Although there have been few studies on the subject of Internet Addiction/Problematic Internet Use (PIU) with a cultural focus (Lopez-Fernandez, 2015), those that exist have found interesting cultural differences in Internet use. Durkee et al. (2012) studied pathological Internet use in the countries Austria, Estonia, France, Germany, Hungary, Ireland, Israel, Italy, Romania, Slovenia, Spain, and Sweden and found that the highest rate of maladaptive Internet use (18.2%) and pathological Internet use (11.8%) was found in Israel 5 and the lowest rates were found in Italy (8.8% and 1.2%). They also found that when comparing metropolitan and micropolitan areas, adolescents living in metropolitan areas showed a higher risk for PIU. They highlighted the importance of this finding and indicated that there must be a significant difference in metropolitan vs micropolitan culture, which should be further investigated. Tsitsika et al. (2014) found that prevalence rates of Internet Addictive Behavior were higher in the Southern and Eastern/Middle European countries and lower in the Northern European countries. More specifically, they found that the country with the highest rate of dysfunctional Internet behavior was Spain with a rate of 23% and the lowest was Iceland with 8%. These results contradict the results of another cross-cultural Internet study by Laconi et al. (2018) which compared problematic Internet use in Italy, Germany, France, Poland, Spain, Turkey, Hungary, England and Greece and found that the Spanish sample had one of the lowest rates of PIU. Loureiro, Pereira, Monteiro, Afonso & Esgalhado (2017) found an interesting paradox in their comparison between Portuguese and Brazilian Internet users. They expected to find more problematic use among Portuguese users, as they had more Internet users per capita and easier access to the Internet than the Brazilians. However, they found that Brazilian users had higher levels of Internet Addiction, thereby demonstrating that ease of access and usage prevalence in a country are not sufficient to predict problematic Internet use. To facilitate cross-cultural comparison of IAT factor analysis research from around the world, we organized the latest meta-analysis findings on IAT’s factor analysis (Moon, Hwang, Kim, Shin, Bae & Kim, 2018) according to geographical region and identified similarities/differences of interest (Table 1). Table 1 About here [Chart of International IAT Factor Analysis Studies]. Comparison of IAT factor analysis studies around the world Asia and Europe had the highest number of IAT factor analysis studies, so we compared their factors to identify if there were interesting similarities and/or differences. There were a couple of things to 6 note about the factor names themselves. Firstly, Asia was the only continent in which the word “withdrawal” was used in the labeling of factors and 50% of the studies contained a mention of “neglect of work/duty” whereas only one other study in all the other papers around the world mentioned such a construct (Tsimtsiou, Haidich, Kokkali, Dardavesis, Young & Arvanitidou, 2014). From a cultural perspective, these two differences may be because of the collectivist nature of most Asian communities. Being an active part of society is very important and highly valued, therefore withdrawing from the group or neglecting one’s role in the community is seen as a sign of a problem. On the other hand, in the European studies, the word “emotion/mood” was used in the factor labeling of almost 60% of the studies, whereas it was not used at all in the factor labeling of Asian studies. This could be because the personal, internal experience of the individual receives more attention in European countries, which tend to be more individualistic, than it does in Asian countries (Hofstede, 1983). There is not enough data to make conclusions in this regard, therefore more research is recommended to explore these potential cultural differences more in depth. There was too much variation in the results among studies to make reliable conclusions about broader geographical regions, and since it is unwise to rely on any single study to accurately represent a population, we instead examined countries for which two or more IAT factor analysis studies had been conducted in order to see if results were replicated, which would add support to the theory that cultural context influences Internet behaviors. We identified the four countries from the meta-analysis in which two or more studies were conducted – Korea, Portugal, Italy, and Turkey – and we compared their findings to see how closely the results were replicated. In both of the Portuguese studies there was missing data regarding the factor analysis, therefore we dropped Portugal from the comparison. In the Korean studies the findings were similar: both studies had a primary factor centered around time management, with almost all the items from Sung, Shin and Cho’s (2014) Factor 1 included in Lee et al.’s (2013) Factor 1. Both studies also had a Withdrawal factor, with all of the items in Lee et al.’s Withdrawal factor included in Sung et al.’s Withdrawal Factor. The studies had different samples, with Sung et al. using teenage participants aged 13-15 and Lee et al. using university students. This age 7 difference may be an important contributor to the findings that they did not share such as that Sung et al. found four factors and Lee et al. found three. The three studies from Italy (Faraci et al., 2013; Fioravanti & Casale, 2015; Servidio, 2017) all showed nearly the same findings. Each study used university students in the sample (Fiorvanti & Casale used university students as well as high school students). Each had a two factor structure with the primary and secondary factors containing almost the same items across all studies; one factor was related to emotional/psychological problems and the other factor was related to loss of control of time and interference with daily life. The results from Turkey (Boysan, Kuss, Barut, Ayköse, Güleçe & Özdemir, 2017; Kaya, Delen & Young, 2016) were different: although both studies explained nearly the same percent variance (46% in Kaya et al. and 45% in Boysan et al.), Boysan et al. found a unidimensional structure whereas Kaya et al. identified four separate factors. As the samples were very similar culturally and demographically, differences may be attributed to the fact that different statistical analyses were used. Present study For the present study, we conducted factor analyses on the IAT in three regions with distinct cultural profiles – the United States (Mid-West region), Colombia (Ibagué), and Spain (Barcelona). We then compared the findings of our own factor analyses to previous IAT factor analysis studies conducted in those countries to see whether findings were consistent within country as they were in Korea and Italy. We conclude by offering suggestions about future research on IA and PIU. Methods Participants The participants were 1,516 university students from 3 universities, one in each country, who filled out an online questionnaire. After dropping the incomplete responses, 451 were left from the USA, 8 467 from Spain and 355 from Colombia. The American participants were 293 females (64.9%), the Spanish participants were 240 females (79.2%) and the Colombian students were 202 females (64.2%). The mean age of American students was 19.59 (SD = 1.43; range 18-30), the mean age of Spanish students was 21.45 (SD = 2.41; range 18-30), and the mean age of Colombian student was 19.95 (SD = 2.00; range 18-30). The Spanish participants were older than the other two groups (F=111.05; p<.001). Measures Internet Addiction Test (Young, 1998): a 20-item self-report questionnaire based on the DSM- IV criteria for pathological gambling. Respondents are asked to rate items on a 5-point Likert scale covering the degree to which their Internet use affects their daily routine, social life, productivity, sleeping pattern, and feelings. The minimum score is 20 and the maximum is 100. The higher the score, the greater the problems that the Internet causes. Young suggested that a score ranging from 20 to 39 is a typical online user who has no problems with their Internet usage. A score ranging from 40 to 69 signifies frequent problems due to Internet usage. Finally, a score ranging from 70 to 100 signifies that the Internet is causing significant problems for the user. The IAT was designed as a unidimensional instrument, however, subsequent studies have found between one and six factors (see Moon et al. 2018). In online applications, reliability varies between 0.83 and 0.91(Korkeila, Karlaas, Jääskeläinen, Vahlberg & Taiminen, 2010; Barke, Nyenhuis & Kröner-Herwig; Jelenchik, Becker & Moreno, 2012). The reliability (Cronbach’s Alpha) for the present study was .91. Procedure Participants were recruited via email in Spain and Colombia, and via the University of Illinois Subject Pool website in the United States (used by students to find and participate in research projects). Participants who chose to complete the study clicked the link provided to them from either the email or the Subject Pool website (USA) and were redirected to the questionnaires on the web host Qualtrics. When a participant accessed the questionnaire, they were presented with a document explaining the study 9 and were asked to provide their informed consent in order to continue. No identifying information was collected from the participants and their responses were encoded as a set of random numbers and letters. IP numbers were not tracked. Some data collected were not related to the Internet focus of the current study and will therefore be presented elsewhere. Data Analysis Principal components analysis with Varimax rotation was used for factor extraction. Prior to exploratory factor analysis, data were inspected to ensure items were significantly correlated, using Bartlett’s Test of Sphericity. In addition, in order to evaluate whether items shared sufficient variance to justify factor extraction, KMO’s Test of Sampling Adequacy was used. Factor loadings resulting from the Varimax rotation were evaluated using the threshold of 0.40. If an item loads on more than one factor, then it is bonded with the factor with the highest loading unless there is a compelling reason to attach it to another factor in order to improve factor interpretability. The IAT factor structure that emerged from exploratory factor analysis was verified using confirmatory factorial analysis (CFA Least Square, which is applicable when data do not meet the assumption of multivariate normality, was selected as the procedure for estimation). Model fit was evaluated based on the comparative fit index (CFI), Tucker- Lewis index (TLI), root-mean-square error of approximation (RMSEA), and standardized root mean square residual (SRMR). CFI and TLI > .90, RMSEA < .08 and SRMR < .1 typically reflect acceptable fit, and CFI and TLI > .95, RMSEA < .06 and SRMR < .08 indicate excellent fit (Brown, 2006). In addition, descriptive and correlational analyses were performed. To test country and sex differences in the study, a bifactorial (sex by country) analysis of variance (General Linear Model procedure) was performed. When main effects were significant, post-hoc comparisons (with Bonferroni adjustment for multiple comparisons) were computed. SPSS 19.0 was used for descriptive statistics, General Linear Model and exploratory factor analysis. EQS 6.1 (Bentler, 2006) was used for CFA. 10 Results Descriptive analysis In the total sample, 72% of respondents showed scores ranging from 20 to 39, meaning no problems with their Internet usage. 27% scored from 40 to 69, meaning frequent problems due to Internet usage, and 1% scored from 70 to 100, for whom the Internet may be a significant problem. Regarding country differences, Spanish participants showed lower scores on the IAT (Mean = 33.50; DT = 9.44) than USA participants (Mean = 36.82; DT = 10.82) and Colombian participants (Mean = 36.70; DT = 11.05). Differences were statistically significant (F = 12.55; p = .000). Factor Analysis The KMO’s Test of Sampling Adequacy was .94 and Bartlett’s Test of Sphericity (χ =9490.9) was significant (P=.000), indicating that the IAT items were appropriate for a factor analysis. For both USA and Spain, this criterion resulted in a three-factor solution whereas in the case of Colombia there were two underlying factors. Table 1 shows the factor loadings of the items for the USA, Spain, and Colombia respectively. For the USA, the three factors explained 51.91% of the variance (see Table 1). Factor 1 (twelve items) accounted for 25.65 % of the variance and appeared to measure psychological conflict. Factor 2 (five items) accounted for 16.67% of the variance and appeared to measure inability to control use. Factor 3 (three items) accounted for 9.66% of the variance and appeared to measure social and work dysfunctions. The fit of this model was excellent (CFI= .985; TLI= .983, RMSEA= .027; SRMR= .042). For Spain, the three factors explained 46.68% of the variance. Factor one (seven items) accounted for 18.16% of the variance and measured social/work dysfunctions and difficulties with time management. Factor two (eight items) accounted for 15.55 % of variance and measured psychological conflicts related to Internet use. Factor three (four items) accounted for the 13.14% of variance and measured affective reaction. Item 14 did not charge at any factor. The fit of this model was excellent (CFI= .989; TLI= .987, 11 RMSEA= .023; SRMR= .040). For Colombia, the two factors explained 54.7% of the variance. Factor one (eleven items) accounted for 30.72% of variance and measured psychological conflicts. Factor two (eight items) accounted for 23.97% of variance and measured inability to control Internet use. Item seven did not charge at any factor. The fit of this model was acceptable (CFI= .978; TLI= .975, RMSEA= .033; SRMR= .049). In Summary: USA Factor 1: Satisfaction of Emotional Needs and Dependence: 3, 4, 5, 9, 10, 11, 12, 13, 15, 18, 19, and 20 Factor 2: Inability to Control use and Neglect of Important Activities: 1, 2, 14, 16, and 17 Factor 3: Neglect of Duties in Favour of the Internet: 6, 7, 8 Spain Factor 1: Inability to Control Use and Neglect of Duties: 1, 2, 6, 7, 8, 16, and 17 Factor 2: Satisfaction of Emotional Needs: 3, 4, 9, 10, 13, 18, 19, and 20 Factor 3: Dependence: 5, 11, 12, and 15 Colombia Factor 1: Satisfaction of Emotional Needs and Dependence: 3, 4, 9, 10, 11, 12, 13, 15, 18, 19, and 20 Factor 2: Inability to Control use and Neglect of Duties: 1, 2, 5, 6, 8, 14, 16, and 17 [Table 2 about Here] Discussion Comparing the IAT factor analyses 12 When we compare our findings to the previous IAT factor analyses in the same countries, we see similarities. We found three factors in the US sample: 1. Satisfaction of Emotional Needs and Dependence, 2. Inability to Control Use and Neglect of Important Activities and 3. Neglect of Duties in Favour of the Internet. When we compare these findings to the previous IAT factor analysis in the US (Jelenchik et al., 2012), we find that both studies have an identical Factor 1 and very similar Factor 2 (all the items from our study’s Factor 2 were included in Jelenchick’s Factor 2). The main difference was that the three additional items in Jelenchick’s Factor 2 were a separate Factor in our study – those related to neglect of work or studies in favour of the Internet. The shared items in the psychological/emotional problems factor relate to dependence on the Internet for positive affect and preference for the Internet over reality. The shared items in the time management problems factor relate to inability to control time online and prioritization of Internet time over other tasks. We found three factors in our Spanish sample as well: 1. Inability to Control Use and Neglect of Duties, 2. Satisfaction of Emotional Needs, and 3. Dependence. When we compare our results to the previous IAT factor analysis in Spain (Fernández-Villa et al., 2015), we again see similarities. Nearly all of the items in our Factor 2 were included in Fernández-Villa et al.’s Factor 1, although their Factor 1 had an additional 4 items, 3 of which composed our Factor 3. Additionally, our Factor 1 was almost identical to Fernández-Villa et al.’s Factor 2. Although the two factors are in switched positions for the two studies, their similarities are important to note; as in the US sample, one was centered on psychological/emotional problems and the other around time management problems. The shared items in the psychological/emotional problems factor primarily related to dependence on the Internet for positive affect. The shared items in the time management problems factor relate to inability to control time online and neglect of important duties in favor of the Internet. We found two factors in our Colombian sample: 1. Satisfaction of Emotional Needs and Dependence, and 2. Inability to Control use and Neglect of Duties. Once again, when compare them to the previous IAT factor analysis study, conducted in Colombia (Puerta-Cortes et al., 2012), we see many similarities. Puerta-Cortes’ Factor 2 and our Factor 1 are nearly identical, with only one item difference. 13 Puerta-Cortes’ Factor 1 also shares a majority of its items with our Factor 2. The shared items in the psychological/emotional problems factor were nearly the same as those in the US sample, related to dependence on the Internet for positive affect and preference for the Internet over reality. The shared items in the time management problems factor related to excessive time spent online and neglect of important duties in favour of the Internet. The Shared Factors and Implications for IA Research All of our samples and the studies we compared them to contained one of their top two factors related to time management problems and the other of the top two factors related to emotional/psychological problems, although these factors manifested somewhat differently across the samples. This finding confirms the finding from Moon et al.’s meta-analysis (2018) which determined that when considering only the studies that strictly follow the factor analysis guidelines; the IAT most likely has one or two real factors. Two items in the time management factor were shared among all six studies: questions 1 and 2. Six items in the emotional/psychological factor were shared among all six studies: questions 3, 9, 10, 13, 19 and 20. The Italian studies all showed the same two factors and the Korean studies also included a time management factor and a factor related to emotional problems/dependence despite the fact that Lee et al. (2012) found four factors and Sung et al. found three (2014). Therefore, we can conclude that although there are differences in the IAT factor analysis findings around the world with factors ranging from 1-5 in the most recent meta-analysis (Moon et al., 2018), there are usually two primary factors that emerge: one related to emotional conflict/dependence and another related to time management problems regarding the Internet. Future diagnosis and treatment efforts should focus on these two factors if more detailed information is not available about the specific population being studied. On the topic of the disputed existence of Internet Addiction as a disorder, judging from the two most common factors identified in the IAT, it appears that there are two primary underlying components of problematic Internet use – impulse control problems and the presence of unsatisfied emotional needs 14 that the Internet is employed to satisfy. It would appear that the comorbidity of these two components manifests as problematic Internet use in the modern technological age. With this in mind, it may be more beneficial from a diagnostic and treatment perspective to focus on these two cognitive-emotional components rather than on “Internet Addiction” as a single construct, which continues to be a somewhat vague concept with various interpretations. As Starcevic (2012, p.93) argues, “‘New’ disorders or patterns of behaviour may be no more than alternative expression of various psychopathological entities.” As seen from the factor analyses conducted around the world, although the IAT may have been designed as a unidimensional measure, this unidimensionality has not been proven. Therefore, perhaps Internet Addiction should also not be considered as a single construct, but rather problematic Internet behaviours should be viewed as the modern-day manifestation of a combination of cognitive-emotional disorders (Starcevic, 2010) that manifested in different ways before the existence of the Internet, but maintain the same basic constructs independently of it. As “Internet Addiction” has been difficult to describe psychometrically due to the instability in diagnostic measures, since problematic Internet use presents differentlty in people around the world, and since clinical cases of Internet addicts are scarce, perhaps there is insufficient support for the diagnosis of “addiction” at this time. It may benefit the diagnosis and treatment of those suffering from problematic Internet use more if research explores what cognitive-emotional profiles are susceptible to Internet use problems and why, and what benefits people with those profiles receive from their Internet use which reinforces it so much that it leads to its prioritization over other aspects of daily life (Kardefelt-Winther, 2014). Conclusion The findings of our three factor analysis studies, in the USA, Spain and Colombia, showed similar results as the previous findings in those same regions. In the USA, we found three factors: 1. Satisfaction of Emotional Needs and Dependence, 2. Inability to Control use and Neglect of Important Activities and 3. Neglect of Duties in Favour of the Internet. In Spain, we also found three factors: 1. Inability to Control Use and Neglect of Duties, 2. Satisfaction of Emotional Needs, and 3. Dependence. 15 And in Colombia we found 2 factors: 1. Satisfaction of Emotional Needs and Dependence, and 2. Inability to Control Use and Neglect of Duties. Spanish participants were found to have the lowest IAT scores among the three countries studied, consistent with a previous study by Laconi et al. (2018). All factor analyses contained a factor related to emotional/psychological problems and another factor related to time management problems, thereby suggesting that impulse control problems and unfulfilled emotional needs are the most important components in the development of problematic Internet use around the world. We, therefore suggest a move away from the addiction framework in problematic Internet use research, which puts the focus on the Internet as a kind of addiction-causing entity like a drug, and instead shift the focus onto the motivations and gratifications of Internet users when engaging with the Internet, and re-conceptualizing PIU as the technological age’s manifestation of an interaction between impulse-control problems and unfulfilled emotional needs. Limitations This study is not without limitations. Firstly, the IAT is a self-report measure, which means results may not be fully accurate, as respondents often have a mistaken perception of their own Internet behaviours. Secondly, in all samples the participants were university students, which means caution should be taken when generalizing results to a more diverse population. Regarding the comparative analysis of Moon et al.’s IAT meta- analysis (2018), this paper did not aim to analyze the statistical procedures associated with each study included in that paper, therefore some of the differences found between or within countries could be attributed to different statistical procedures used, not culture. One must also take into account how the passage of time may have affected the effectiveness of the IAT to measure problems with Internet use. The IAT was developed in 1998 before the extensive use of the Internet around the world. These past 22 years of development have probably affected the relevance of the questionnaire and the importance of certain items. Therefore, it would benefit the 16 literature on this subject if the IAT, being the most popular questionnaire in the field, were updated to better reflect Internet behaviors that are popular and problematic today. Conflicts of Interest The authors declare no conflict of interest 17 References Ahmad, M., Alzayyat, A. & Al-Gamal, E. (2015). The factor structure of the Internet Addiction Tool with university students in Jordan. Issues in Mental Health Nursing, 36(9), 725– 731. doi: 10.3109/01612840.2015.1033041. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Bell, C. C. Bahrainian, S. A., Alizadeh, K. H., Raeisoon, M., Hashemi Gorji, O. & Khazaee, A. (2014). Relationship of Internet addiction with self-esteem and depression in university students. Journal of Preventive Medicine and Hygiene, 55(3), 86-89. Barke, A., Nyenhuis, N. & Kroner-Herwig, B. (2012). The German version of the Internet Addiction Test: A validation study. Cyberpsychology, Behavior and Social Networking, 15(10), 534–542. doi: 10.1089/cyber.2011.0616 Bentler, P. M. (2006). EQS 6 structural equations program manual. Encino: Multivariate Software Boysan, M., Kuss, D. J., Barut, Y., Ayköse, N., Güleçe, M. & Özdemir, O. (2017). Psychometric properties of the Turkish version of the Internet Addiction Test (IAT). Addictive Behaviors, 64, 247–252. doi: 10.1016/j.addbeh.2015.09.002 Bozoglan, B., Demirer, V. & Sahin, I. (2013). Loneliness, self-esteem, and life satisfaction as predictors of Internet addiction: A cross-sectional study among Turkish university students. Scandinavian Journal of Psychology, 54, 313-319. doi: 10.1111/sjop.12049 Brown, T.A. (2006). Confirmatory factor analysis for applied research. New York: Guilford Press. Chen, Y-L. & Gau, S. S-F. (2016). Sleep problems and Internet addiction among children and adolescents: A longitudinal study. Journal of Sleep Research, 25(4), 458-465. doi: 10.1111/jsr.12388. 18 Conti, M. A., Jardim, A.P., Hearst, N., Táki, A. C., Tavares, H. & Nabuco de Abreu, C. (2012). Evaluation of semantic equivalence and internal consistency of a Portuguese version of the Internet Addiction Test (IAT). Revista de Psiquiatria Clinica, 39(3), 106–110. doi: 10.1590/S0101- 60832012000300007 Dhir, A., Che, S. & Nieminen, M. (2015). Psychometric validation of Internet Addiction Test with Indian adolescents. Journal of Educational Computing Research, 53(2), 15–31. doi: 10.1177/0735633115597491 Durkee, T., Kaess, M., Carli, V., Parzer, P., Wasserman, C., Floderus, B., Apter, A., ... Wasserman, D. (2012). Prevalence of pathological Internet use among adolescents in Europe: Demographic and social factors. Addiction, 107(12), 2210-2222. doi:10.1111/j.1360-0443.2012.03946.x. Faraci, P., Craparo, G., Messina, R. & Severino, S. (2013). Internet Addiction Test (IAT): which is the best factorial solution? Journal of Medical Internet Research, 15(10), 220–230. doi: 10.2196/jmir.2935. Fernandez-Villa, T., Molina, A. J., García-Martín M, Llorca, J., Delgado-Rodríguez, M. & Martín, V. (2015). Validation and psychometric analysis of the Internet Addiction Test in Spanish among college students. BMC Public Health, 15(1), 1-9. doi: 10.1186/s12889-015-2281-5 Fioravanti, G. & Casale S. (2015). Evaluation of the psychometric properties of the Italian Internet Addiction Test. Cyberpsychology, Behavior, and Social Networking, 18(2), 120–128. doi: 10.1089/cyber.2014.0493. Guan, N. C., Isa S. M., Hashim, A. H., Pillai, S. K. & Harbajan Singh, M. K. (2015). Validity of the Malay version of the Internet Addiction Test: A study on a group of medical students in Malaysia. Asia-Pacific Journal of Public Health, 27(2). NP2210-NP2219. doi: 10.1177/1010539512447808. 19 Hawi, N. S. (2013). Arabic validation of the Internet Addiction Test. Cyberpsychology, Behavior, and Social Networking, 16, 200–204. doi: 10.1089/cyber.2012.0426 Hawi, N. S., Blachnio, A. & Przepiorka, A. (2015). Polish validation of the Internet Addiction Test. Computers in Human Behavior, 48, 548–553. doi: 10.1016/j.chb.2015.01.058 Ho, R. C., Zhang, M. W. B., Tsang, T.Y., Toh, A. H., Pan, F., Lu, Y., Cheng, C., ... Mak, K-K. (2014). The association between internet addiction and psychiatric co-morbidity: a meta-analysis. BMC Psychiatry, 14(1), 183. doi: 10.1186/1471-244X-14-183 Hofstede, G. (1983). National Cultures in Four Dimensions: A Research Based Theory of Cultural Differences Among Nations. International Studies of Management & Organization, 13(1-2), 46- 74. doi: 10.1080/00208825.1983.11656358 Jelenchick, L. A., Becker, T. & Moreno, M. A. (2012). Assessing the psychometric properties of the Internet Addiction Test (IAT) in US college students. Psychiatry Research, 196(2-3), 296–301. doi: 10.1016/j.psychres.2011.09.007. Kardefelt-Winther, D. (2014). A conceptual and methodological critique of internet addiction research: Towards a model of compensatory internet use. Computers in Human Behavior, 31, 351–354. doi:10.1016/j.chb.2013.10.059 Kaya, F., Delen, E. & Young, K. S. (2016). Psychometric properties of the Internet Addiction Test in Turkish. Journal of Behavioral Addictions, 5(1),130–134. doi: 10.1556/2006.4.2015.042. Korkeila, J., Kaarlas, S., Jääskeläinen, M., Vahlberg, T., & Taiminen, T. (2010). Attached to the web – harmful use of the Internet and its correlates. European Psychiatry, 25(4), 236–241. https://doi.org/10.1016/j.eurpsy.2009.02.008 20 Laconi, S., Kaliszewska-Czeremska, K.. Gnisci, A., Sergi, A., Barke, A., Jeromin, F., Groth, J., ...Kuss, D. J. (2018). Cross-cultural study of Problematic Internet Use in nine European countries, Computers in Human Behavior, 84, 430-440. doi.org/10.1016/j.chb.2018.03.020. Laconi, S., Rodgers, R. F., Chabrol, H. (2014). The measurement of Internet addiction: A critical review of existing scales and their psychometric properties. Computers in Human Behavior, 21, 190-202. doi: 10.1016/j.chb.2014.09.026 Lai, C. M., Mak, K. K., Watanabe, H., Ang, R. P., Pang, J. S. & Ho, R. C. (2013). Psychometric properties of the Internet addiction test in Chinese adolescents. Journal of Pediatric Psychology, 38(7), 794–807. doi: 10.1093/jpepsy/jst022. Lee, B. W. & Stapinski, S. L. (2012). Seeking safety on the internet: Relationship between social anxiety and problematic internet use. Journal of Anxiety Disorders, 26(1), 197-205. doi: 10.1016/j.janxdis.2011.11.001. Lee, H. W., Choi, J-S., Shin, Y-C., Lee, J-Y., Jung, H. Y. & Kwon, J. S. (2012). Impulsivity in Internet Addiction: A Comparison with Pathological Gambling. Psychology, Behavior, and Social Networking, 15(7), 373 – 377. doi: 10.1089/cyber.2012.0063 Lee, K., Lee H-K., Gyeong H., Yu, B. & Kim, D. (2013). Reliability and validity of the Korean version of the Internet Addiction Test among college students. Journal of Korean Medical Science, 28(5), 763–768. doi: 10.3346/jkms.2013.28.5.763. Lee, Y. S., Han, D. H., Kim, S. M. & Renshaw, P. F. (2015). Substance abuse precedes Internet addiction. Addictive Behaviors, 38(4), 2022-2025. doi: 10.1016/j.addbeh.2012.12.024. Lin, I-H., Ko, C-H., Chang, Y-P., Liu, T-L., Wang, P-W, Lin, H-C., Huang, M-F., ... Yen, C-F. (2014). The association between suicidality and Internet addiction and activities in Taiwanese adolescents. Comprehensive Psychiatry, 55(3), 504-510. doi: 10.1016/j.comppsych.2013.11.012. 21 Lopez-Fernandez, O. (2015). Cross-cultural research on Internet addiction: A systematic review. International Archives of Addiction Research and Medicine, 1(11). Mohammadsalehi N, Mohammadbeigi, A., Jadidi, R., Anbari, Z., Ghaderi, E. & Akbari, M. (2015). Psychometric properties of the Persian language version of Yang Internet Addiction Questionnaire: an explanatory factor analysis. International Journal of High Risk Behaviors and Addiction, 4(3), e21560. doi: 10.5812/ijhrba.21560. Moon, M. S., Hwang, J. S., Kim, J. Y., Shin, A. L., Bae, S. M., & Kim, J. W. (2018). Psychometric properties of the Internet Addiction Test: a systematic review and meta-analysis. Cyberpsychology, Behavior, and Social Networking, 21(8). doi: 10.1089/cyber.2018.0154. Orsala, O., Orsalb, O., Unsalc, A., Ozalp, S. S. (2013). Evaluation of Internet addiction and depression among university students. Procedia - Social and Behavioral Sciences, 82, 445-454. doi: 10.1016/j.sbspro.2013.06.291 Panayides, P. & Walker, M. J. (2012). Evaluation of the psychometric properties of the Internet Addiction Test (IAT) in a sample of cypriot high school students: The Rasch Measurement Perspective. Europe’s Journal of Psychology, 8(3), 327-351. doi: 10.5964/ejop.v8i3.474 Pedrero-Pérez, E. J., Ruiz-Sánchez de León, J. M., Rojo-Mota, G, Llanero-Luque, M., Pedrero-Aguilar, J. Morales-Alonso, S., Puerta-García, C. (2018), Tecnologías de la Información y la Comunicación (TIC): uso problemático de Internet, videojuegos, teléfonos móviles, mensajería instantánea y redes sociales mediante el MULTICAGE-TIC. Adicciones, 30(1), 19-32. doi: 10.20882/adicciones.806 Pontes, H. M., Patrão, I, M. & Griffiths, M. D. (2014). Portuguese validation of the Internet Addiction Test: an empirical study. Journal of Behavioral Addictions, 3(2), 107–114. doi: 10.1556/JBA.3.2014.2.4. 22 Puerta-Cortés, D. X., Carbonell, X. & Chamarro, A. (2012). Analysis of the psychometric properties of the Spanish version of Internet Addiction Test. Trastornos Adictivos, 14(4), 99– 104. doi: 10.1016/S1575-0973(12)70052-1 Rezaul Karim, A. K. M. & Nigar, N. (2014). The Internet Addiction Test: Assessing its psychometric properties in Bangladeshi culture. Asian Journal of Psychiatry, 10, 75–83. doi: 10.1016/j.ajp.2013.10.011. Sánchez-Carbonell, X., Beranuy, M., Castellana, M., Chamarro, A. (2008). La adicción a Internet y al móvil. ¿Moda o trastorno? Adicciones, 20(2), 149-160. doi: 10.20882/adicciones.279 Samaha, M. & Hawi, N. S. (2016). Relationships among smartphone addiction, stress, academic performance, and satisfaction with life. Computers in Human Behavior, 57, 321-325. doi: 10.1016/j.chb.2015.12.045 Seabra, L., Loureiro, M., Pereira, H., Monteiro, S., Afonso, R. M. & Esgalhado, G. (2017). Relationship between Internet addiction and self-esteem: Cross-cultural study in Portugal and Brazil. Interacting with Computers, 29(5), 767–778. doi:10.1093/iwc/iwx011. Servidio, R. (2017). Assessing the psychometric properties of the Internet Addiction Test: a study on a sample of Italian university students. Computers in Human Behavior, 68, 17–29. 10.1016/j.chb.2016.11.019 Starcevic, V. (2010). Problematic Internet use: a distinct disorder, a manifestation of an underlying psychopathology, or a troublesome behaviour? World Psychiatry, 9(2): 92–93. Starcevic, V. (2013). Is Internet addiction a useful concept? Australian & New Zealand Journal of Psychiatry, 47(1), 16–19. doi:10.1177/0004867412461693 Stevens, J. (2018, December 17). Internet Stats & Facts for 2019. Hosting Facts. Retrieved from https://hostingfacts.com/internet-facts-stats/ 23 Sung, M., Shin, Y. M. & Cho, S. M. (2014). Factor structure of the Internet Addiction Scale and its associations with psychiatric symptoms for Korean adolescents. Community Mental Health Journal, 50, 612–618. doi: 10.1007/s10597-013-9689-0. Tsimtsiou, Z., Haidich, A. B., Kokkali, S., Dardavesis, T., Young, K. S. & Arvanitidou, M. (2014). Greek version of the Internet Addiction Test: a validation study. Psychiatric Quarterly, 85(2), 187–195. doi: 10.1007/s11126-013-9282-2. Tsitsika, A., Janikian, M., Schoenmakers, T. M., Tzavela, E. C., Ólafsson, K., Wójcik, S., Macarie, G. F., ... Richardson, C. (2014). Internet addictive behavior in adolescence: a cross- sectional study in seven European countries, Cyberpsychology, Behavior, And Social Networking, 17(8), 528-535. doi: 10.1089/cyber.2013.0382 Waqas, A., Farooq, F., Raza, M., Javed, S. T., Khan, S., Ghumman, M. E., Naveed, S. & Haddad, M. (2017). Validation of the Internet Addiction Test in students at a Pakistani medical and dental school. Psychiatric Quarterly, 89(1), 235–247. doi: 10.1007/s11126-017-9528-5. Watters, C. A., Keefer, K. V., Kloosterman, P. H., Summerfeldt, L. J. & Parker, J. D. A. (2013). Examining the structure of the Internet Addiction Test in adolescents: a bifactor approach. Computers in Human Behavior, 29(6), 2294–2302. doi: 10.1016/j.chb.2013.05.020 Weinstein, A., Yaacov, Y., Manning, M., Danon, P. & Weizman, A. (2015). Internet Addiction and Attention Deficit Hyperactivity Disorder among schoolchildren. The Israel Medical Association Journal, 17(12), 731-734. Widyanto, L., & Griffiths, M. (2006). “Internet Addiction”: A critical review. International Journal of Mental Health and Addiction, 4(1), 31–51. doi:10.1007/s11469-006-9009-9 24 Widyanto, L., Griffiths, M. D. & Brunsden, V. A. (2011). Psychometric comparison of the Internet Addiction Test, the Internet-related problem scale, and self-diagnosis. Cyberpsychology, Behavior, and Social Networking, 14(3), 141–149. doi: 10.1089/cyber.2010.0151 Yao, M. Z. & Zhong, Z. (2014). Loneliness, social contacts and Internet addiction: A cross-lagged panel study. Computers in Human Behavior, 30, 164-170. doi: 10.1016/j.chb.2013.08.007 Younes,F., Halawi , G., Jabbour, H., El Osta, N., Karam, L., Hajj, A., Khabbaz, L. R. (2016). Internet addiction and relationships with insomnia, anxiety, depression, stress and self-esteem in university students: A Cross-Sectional Designed Study. PLoS ONE, 11(9), e0161126. doi:10.1371/journal.pone.0161126 Young, K. (1998). Caught in the net. New York: John Wiley 25 Tables and Figures Table 1 Total Population # of # of Variance Name of Factors (and number of items) Studied Participants Factors (%) East Asia 1. Middle-school 1. 1,722 1. 3 1. 50.0 1. Time management (6), withdrawal behaviors (8), 1. Korea: Sung, M., Shin, Y-M., students 2. 162 2. 5 2. 64.0 neglect work (4) Cho, S-M. (2014). 2. College 3. 1,914 3. 1 3. 41.4 2. Lack of control (8), neglect of duty (7), 2. Malaysia: Guan NC, Isa SM, Medical 4. 172 4. 4 4. 55.7 problematic use (2), social relationship disruption (2), Hashim AH, et al. (2015). Students 5. 279 5. 4 5. 58.9 e-mail primacy (1) 3. India: Dhir A, Chen S, Nieminen 3. High school 6. 844 6. 3 6. - 3. – M. (2015). students 4. Neglect of duty (6), online dependence (4), virtual 4. Bangladesh: Karim AKMR, Nigar 4. University fantasies (5), privacy and self-defense (3) N. (2014). students 5. excessive Internet use (9), dependence (5), 5. Korea: Lee K, Lee H-K, Gyeong 5. University withdrawal (3), avoidance of reality (3) H, et al. (2013). students 6. withdrawal and social problems (9), time 6. China: Lai C-M, Mak K-K, 6. Middle and management and performance (6), reality substitute Watanabe H, et al. (2013). high-school (3) students Europe 1. University 1. 659 1. 2 1. 41.0 1. Interpersonal, emotional, and obsessive conflict as 1. Italy: Servidio R. (2017). students 2. 453 2. 1 2. 44.9 a result of Internet use (11), online time management 2. Turkey: Boysan M, Kuss DJ, 2. College 3. 407 3. 4 3. 46.0 and compromised personal well-being (7) Barut Y, et al. (2016) students 4. 1,245 4. 2 4.44.6 2. – 3. Turkey: Kaya F, Delen E, Young 3. University 5. 840 5. 2 5. 45.6 3. Mood (8), relationship (4), responsibilities (5), KS. (2016). students 6. 851 6. 2 6. 55.0 duration (3) 4. Polish: Hawi NS, Blachnio A, 4. College 7.151 7. 3 7. 55.3 4. Mental disorder (11), time management disorder Przepiorka A. (2015) students 8. 593 8. – 8. – (9) 5. Italy: Fioravanti G, Casale S. 5. Students 9. 485 9. 2 9. 42.2 5. Emotional and cognitive preoccupations with the (2015) between 14 and 10. 77 10. – 10. – Internet and social consequences (11), loss of control 6. Spain: Fernandez-Villa T, Molina 26 11. 1,041 11. 2 11. online: and interference with daily duties (9) AJ, Garcıa-Martın M, et al. (2015) 6. College online & 12. 3 46.7, 6.Emotional investment (11), time management and 7. Greek: Tsimtsiou Z, Haidich A-B, students 841 offline offline: performance (8) Kokkali S, et al. (2014). 7. College 12. 225 42.0 7. Psychological/emotional conflict (10), time 8. Portugal: Pontes HM, Patrao IM, students: 12. 56.3 management (6), neglect work (4) Griffiths MD. (2014). medical 8. – 9. Italy: Faraci P, Craparo G, 8. High school 9. Emotional and cognitive preoccupation with the Messina R, et al. (2013). and university Internet (11), loss of control and interference with 10. Portugal: Conti MA, Jardim AP, students daily life (7) Hearst N, et al. 9. College 10. – 11. Germany: Barke A, Nyenhuis N, students 11. Preoccupation (12), loss of control (8) Kroner-Herwig B. (2012) 10. University 12. Emotional/psychological conflict (9), time 12. UK? Widyanto L, Griffiths MD, students management issues (5), mood modification (6) Brunsden V. (2011) 11. College students: psychology 12. General population (Internet users) North America 1. High-school 1. 1,948 1. – 1. – 1. – 1. Canada: Watters CA, Keefer KV, students 2. 215 2. 2 2. 91.0 2. Dependent use (12), excessive use (8) Kloosterman PH, et al. (2013). 2. 2. USA: Jelenchick LA, Becker T, undergraduate Moreno MA. students aged 18–20 26 South America 1. Colombia: Puerta-Cortes DX, 1. General 1. 1,117 1. 3 1. 47.8 1. Consequences for the use of the Internet (7), Carbonell X, Chamarro A. (2012). population cognitive and emotional dimension (10), control of (Internet users) time (3) Africa - - - - - Middle East Asia 1. Pakistan: Waqas, A., Faroog, F., 1. College 1. 522 1. 1 1. 34.1 1. – Javed, ST., Khan, S., Ghumman, students 2. 254 2. 3 2. 55.8 2. Personal activities disorder (11), emotional and ME., Naveed, S., Haddad, M. (2018). (medical and 3. 587 3. 4 3. 52.7 mood disorder (6), social activities disorder (3) 2. Persia: Mohammadsalehi N, dental) 4. 817 4. 1 4. 40.6 3. Excessive use (6), loss/suffer (6), attached to (4), Mohammadbeigi A, Jadidi R, et al. 2. College impaired social relations (4) (2015). students 4. – 3. Jordan: Ahmad M, Alzayyat A, (medical Al-Gamal E. (2015). sciences) 4. Lebanon: Hawi NS. (2013). 3. University students 4. Middle and High-school students 27 Table 2 Factor analysis of the IAT in the three countries USA Spain Colombia 1 2 3 1 2 3 1 2 Q1 -.013 .709 .342 .584 -.050 .398 .039 .767 Q2 .182 .569 .380 .725 .170 .260 .299 .731 Q3 .561 -.068 .355 .022 .435 -.001 .569 .131 Q4 .522 .090 .080 .193 .474 .120 .629 .303 Q5 .626 .146 .118 .250 .264 .464 .277 .673 Q6 .301 .336 .600 .707 .278 .081 .540 .555 Q7 -.087 .167 .661 .466 -.035 .260 .225 .253 Q8 .347 .232 .657 .725 .247 .006 .482 .612 Q9 .595 .139 .204 .207 .615 .177 .651 .349 Q10 .407 .349 .346 .165 .502 .286 .679 .356 Q11 .588 .403 .029 .245 .190 .649 .667 .368 Q12 .566 .425 -.050 .095 .083 .701 .626 .436 Q13 .697 .084 .016 .173 .538 .408 .713 .223 Q14 .246 .647 .104 .357 .353 .305 .355 .587 Q15 .732 .389 -.024 .215 .249 .715 .700 .377 Q16 .154 .806 .102 .680 .210 .251 .283 .707 Q17 .254 .688 .203 .655 .330 .112 .374 .651 Q18 .709 .194 .152 .407 .514 .123 .571 .488 Q19 .669 .225 .192 .100 .765 .099 .796 .211 Q20 .772 .093 .084 .114 .554 .450 .812 .222