Smartphones are a new generation of mobile phones, containing many communication applications (e.g., social media) constantly connected to the internet, making many aspects of people’s lives more convenient and social [1–4]. However, evidence indicates that smartphone use can be potentially addictive and/or the applications on them [5,6]. Additionally, prolonged screen time and the internet are associated with physical and psychological problems, low sleep quality, and interrupting individuals’ daily lives [7,8].

Over the past decade, the number of smartphone owners is likely to continue to rise, particularly in Asia [9]. Thailand (where the present study was carried out) is currently 16^th globally and 3^rd in Southeast Asia for time spent on smartphones with an average of 2.48 h spent a day on social media in 2021 [10]. An online Thai survey reported that there was a 3.4% increase in social media use (up to 81.2%) of the total population in 2022, from 2021 to 2022 (approximately 1.9 million more individuals) [11]. Given Thailand’s relatively high rates of smartphone and social media use [9–12], there is good reason to examine problematic (and potentially addictive) use. Currently, there is only one validated Thai instrument that assesses problematic social media use (Thai Bergen Facebook Addiction Scale [Thai-BFAS]), as well as instruments that assess social media engagement (i.e., Thai-Social Media Engagement Scale [T-SMES]), and problematic internet use more generally (i.e., Thai Internet Addiction Test [Thai-IAT]) [12–14], but none of these assess all types of problematic social media use or assess problematic smartphone use. Therefore, it is important to validate instruments assessing the risk of smartphone addiction and social media addiction (SMA) in general (rather than a specific platform). Moreover, the evidence of online-related addictions in Thailand could be extended with greater choice in available validated instruments.

The Smartphone Application-Based Addiction Scale (SABAS) and Bergen Social Media Addiction Scale (BSMAS) are commonly used instruments worldwide with established evidence of their psychometric properties [15–17]. The SABAS assesses the risk of smartphone application addiction [18]. The BSMAS assesses the risk of SMA [19,20]. The items of both the SABAS and BSMAS correspond to the six components of addiction (i.e., preoccupation, conflict, mood modification, tolerance, withdrawal) proposed by Griffiths [21,22]. Previous studies have indicated that the SABAS and BSMAS have robust reliability and validity, and that both instruments assess the risk of smartphone addiction or SMA [17–19,23,24].

Empirical evidence has shown that the SABAS and BSMAS have equivalence between genders based on measurement invariance tests in a number of languages such as Italian, Chinese, Persian, and Romanian [25–29]. The psychological literature has indicated that gender differences are an important factor in the development of internet-related addictions [30]. For example, males are more likely to have problematic online gaming while females are more likely to have problematic social media use [20]. Research carried out in the Asia-Pacific region has found that online gaming provides competitive elements that make males more vulnerable because they prefer competition over cooperation in gaming contexts [30]. In addition, males are inclined to use online gaming as a coping mechanism for emotional regulation (e.g., stress) [30]. In contrast, females are more inclined to use social media for communication and social interaction [30,31]. Therefore, confirming measurement equivalence of the SABAS and BSMAS by gender is important. It is also important to investigate the factor structures of the SABAS and BSMAS and to assess whether the factor structures are invariant across genders.

The present study is the first study in Thailand to provide preliminary evidence regarding the validation of both the Thai SABAS and BSMAS which assess the severity of smartphone addiction and SMA, respectively. Although both SABAS and BSMAS have been validated in different cultures and languages [25–29], evidence concerning the psychometric properties of Thai SABAS and BSMAS is important. Moreover, there has been an increase in smartphone and social media use in Thailand in recent years (e.g., a 3.4% increase in social media use) [11]. Therefore, Thai individuals may not view the overuse of smartphones and social media as a serious problem. In this regard, validating both the Thai SABAS and BSMAS is important because Thai researchers need to be confident that both instruments can accurately assess SMA and smartphone addiction. In addition, the validation of both scales will provide psychometrically robust instruments for researchers to assess smartphone addiction and SMA in Thailand. The present study is also the first in Thailand to investigate gender differences in smartphone addiction, SMA, and internet gaming disorder (IGD) among young adults using psychometrically validated instruments. Subsequently, the present study’s findings provide novel data and contribute to the literature regarding the understanding of gender comparisons across specific online addictions (i.e., smartphone addiction, SMA, and IGD). Moreover, gender comparisons may highlight different health issues for healthcare providers and help in developing more effective gender-based interventions in tackling smartphone addiction, SMA and IGD.

Based on the previous evidence, it was hypothesized that among Thai university students (i) both the Thai BSMAS and SABAS would have satisfactory validity and reliability (H₁), (ii) measurement invariance would be supported across gender (H₂), (iii) SMA would be significantly associated with smartphone addiction and IGD (H₃), and (iv) there would be significant differences between Thai females and males relating to smartphone addiction, SMA and IGD (H₄).

Table 1 shows the participants’ characteristics (n = 801) comprising 536 females (66.9%) and 265 males (33.1%). The participants’ mean age was 20.69 years (SD = 3.78). The mean body mass index of participants was 21.84 kg/m² (SD = 4.47). Most of the participants reported being in good health (91.5%). They spent an average of 7.02 h a day using social media and 2.55 h a day gaming. Using the cut-offs recommended by the previous studies [15,33,35], the mean scores of the SABAS, BSMAS, and IGDS9-SD were lower than the cut-off points for being at risk of addiction to these behaviors. However, 302 participants were classed as being at risk of smartphone addiction (37.7%), 170 participants were classed as being at risk of SMA (21.2%), and 17 participants were classed as being at risk of IGD (2.1%).

Table 2 shows that females had significantly higher scores than males on the SABAS (19.03 ± 5.86 vs. 17.61 ± 6.62, p = 0.002). Males had significantly higher scores than females on the IGDS9-SF (19.39 ± 7.94 vs. 14.85 ± 6.75, p < 0.001). However, there were no significant differences between males and females in BSMAS scores (15.15 ± 4.46 vs. 15.38 ± 5.13).

Table 3 shows descriptive statistics and reliability coefficients for the SABAS and BSMAS items. Acceptable psychometric properties for both SABAS and BSMAS items were found in the CFA with a confirmed unidimensional structure for both scales. In addition, the results of Table 3 indicated that all six items of the SABAS and BSMAS presented good factor loadings in the CFA (SABAS = 0.52–0.78; BSMAS = 0.56–0.76) with satisfactory item-total correlations (SABAS = 0.49–0.71; BSMAS = 0.50–0.68). More specifically, Table 4 shows good fit indices for SABAS and BSMAS in the CFA results. For the SABAS: CFI = 0.996; TLI = 0.994; RMSEA = 0.036; and SRMR = 0.036. For the BSMAS; CFI = 0.995; TLI = 0.991; RMSEA = 0.038; and SRMR = 0.038. Moreover, internal consistencies of both the SABAS and BSMAS were very good (Cronbach’s α = 0.86 and McDonald’s ω = 0.86 for SABAS; Cronbach’s α = 0.83 and McDonald’s ω = 0.83 for BSMAS).

Table 5 shows the invariance of the SABAS, and BSMAS across gender groups. The indicators of model fit presented satisfactory measurement invariance. For the SABAS, the measurement invariance was supported, with all data fit indices (∆CFI, ∆RMSEA, ∆SRMR) being acceptable between each model except for a significant χ² difference (comparing M3 and M2 models). Moreover, the BSMAS had its measurement invariance supported, with non-significant χ² difference and all data fit indices (∆CFI, ∆RMSEA, ∆SRMR) being satisfactory.

Table 6 shows the correlation matrix regarding scores on the SABAS, BSMAS, and IGDS9-SF, as well as social media use frequency and gaming frequency. SABAS scores were significantly correlated with BSMAS scores (r = 0.58; p < 0.001), IGDS9-SF scores (r = 0.31; p < 0.001) and social media frequency (r = 0.21; p < 0.001). BSMAS scores were significantly correlated with IGDS9-SF scores (r = 0.48; p < 0.001), and social media use frequency (r = 0.10; p = 0.005). IGDS9-SF scores were significantly correlated with gaming frequency (r = 0.22; p < 0.001). Moreover, social media use frequency was significantly correlated with gaming frequency (r = 0.12; p < 0.001).

The results of the present study are in line with the three hypotheses. First, both Thai versions of SABAS and BSMAS presented satisfactory validity and reliability (H₁). Second, their measurement invariance was supported across gender (H₂). Third, SMA was significantly associated with smartphone addiction and IGD (H₃). Finally, significant differences were found between females and males in regard to smartphone addiction and IGD (H₄) whereas there were no significant gender differences in SMA. Therefore, the findings indicated that the Thai version of SABAS and BSMAS appear to be valid, reliable, and gender-invariant scales to examine the risk of developing smartphone addiction and SMA among Thai young adults. Additionally, the present study had similar findings to those of other studies confirming that the SABAS and BSMAS had a one-factor structure together with significant correlations with the IGDS9-SF [15,29]. The results suggest that individuals in different countries (including those in Thailand) appear to have equivalent interpretations of SABAS and BSMAS items. Therefore, it is concluded that SABAS and BSMAS are psychometrically robust instruments across different populations. Moreover, Thai males and females tended to have different levels of smartphone addiction and IGD but similar levels of SMA.

Consistent with the findings from previous studies [15,18,27,29], both the SABAS and BSMAS had very good internal consistencies and one-factor structures. More specifically, the present findings demonstrated that both the SABAS and BSMAS had a unidimensional structure and showed very good psychometric properties which were comparable with the original version (α = 0.86 for SABAS; α = 0.88 for BSMAS). In addition, the internal consistency of the Thai BSMAS was found to be comparable with the previous versions of the BSMAS, BFAS and the Thai BFAS (Cronbach’s α = 0.91) [13,18–20]. Given the good psychometric properties found in the present study, the Thai versions of the SABAS and BSMAS could be additional options for healthcare providers assessing SMA and smartphone addiction in addition to the currently existing Thai-BFAS, T-SMES, and Thai-TAI [12–14].

Additionally, the present study is one of the few that has compared smartphone addiction, SMA, and IGD simultaneously among Thai young adults. Consistent with the previous findings outside of the Thai population, results indicated that females had higher levels of smartphone addiction than males [46–48]. A previous review indicated that the purposes underlying the time spent using smartphones were different between females and males [48]. More specifically, it has been reported that females are more likely to use smartphones as tools to maintain intense social relationships (i.e., phone calls, texting, and social networking) and spend greater time on each of these smartphone-related applications whereas males use smartphones less for maintaining relationships and more for such activities as accessing news feeds [48,49]. Because spending time socially interacting is usually more time-intensive than activities such as reading social media news feeds, females are thus more likely than males to be addicted to smartphone use. Furthermore, the results indicated that males had higher levels of IGD than females which concurs with prior studies, almost all have shown that gaming disorder is more highly prevalent among males [46–48]. The literature has indicated that online gaming can be a coping mechanism to escape from unpleasant situations and helps improve individuals’ self-esteem, decreases stress, alleviates poor mood, and enhances social contact [47,50]. In addition, the family has the most important role in Asian culture from childhood to adulthood (which is especially important given the study was carried out among Thai students) [51].

A previous Taiwanese study indicated that Asian parents protect and constrain their daughters’ leisure activities more than their sons (e.g., daughters should not stay outside overnight). Therefore, Asian females are likely to be constrained from spending time on gaming compared to Asian males [47]. Similarly, Thai females are likely to experience greater day-to-day supervision from their parents which results in lower levels of IGD compared to males. However, further study is needed to address the causal relationship between IGD and gender differences among Thai individuals. The present study also found no differences between males and females in terms of SMA, which does not concur with previous findings [52]. A meta-analysis reported that females were more inclined to spend greater amounts of time on social media than males because females enjoy social interactions more than males [52]. However, another review reported that the relationship between gender and SMA is complex because of moderating effects from confounding variables (e.g., psychological distress and cultural variables) [53]. Therefore, it is possible that Thai males and females have similar levels of SMA. However, future studies are needed for corroboration.

In line with previous studies [24,26,27,34], the present study’s findings indicated the factorial structures of the SABAS and BSMAS were similar across genders. This means that when using the SABAS and BSMAS to assess addictions to smartphones and social media, gender will not be a factor influencing the respondents’ interpretations of individual items. However, as previously suggested, further studies should confirm the present findings in other age-related and cross-cultural contexts to verify the validity and reliability of the SABAS and BSMAS [24,34].

The results here provide further support for the notion made in previous studies that scores on the SABAS and BSMAS have strong and positive associations with IGDS9-SF scores [15,16,24,34]. A recent study indicated that SMA and IGD were both associated with psychological distress (i.e., depression, anxiety, and stress) [26]. Nevertheless, some studies have argued that types of specific internet addiction (i.e., SMA, smartphone addiction, IGD) are different from each other [15,54]. In other words, there are fundamental differences between types of internet addiction, and the SABAS, BSMAS, and IGDS9-SF assess different types of internet addictions [15,55]. Moreover, the literature has suggested that further studies should examine overlapping and important components of specific internet addiction which could prevent the different potential risks of internet-related addictions [6].

Several limitations of the present study should be noted. First, the participants consisted of Thai university students, therefore the findings are difficult to generalize to more diverse cultural demographics, age groups (e.g., children, adults), and populations other than university students. Second, the present study was a cross-sectional design and had weak evidence concerning any causal relationships among the studied variables. Third, the participants were recruited using a self-selected convenience sampling method, and the results may be biased due to memory recall and/or social desirability. Moreover, the risk of SMA, smartphone addiction, and IGD were identified through self-report instruments. Therefore, single-rater bias may exist, and future studies are encouraged (where possible) to use clinical diagnosis to replace self-reported psychological distress to avoid this type of bias. Fourth, the present study had more female participants (66.9%) than male participants (33.1%). Therefore, the imbalanced gender distribution may limit the generalizability of the findings. Fifth, the present study used the addiction risk cut-off points of SABAS, BSMAS and IGDS9-SF recommended in previous studies not on Thai populations [15,33,35]. However, these cut-offs were developed in other populations and may not directly equate to Thai university students. Therefore, the severity estimations of the SMA, smartphone addiction, and IGD for the present sample might not be accurate. Future studies are needed to determine an accurate cut-off of these instruments in screening for the risk of smartphone addiction, SMA and IGD for the Thai population. Finally, the present study collected data using online surveys. Such data may not be of the highest quality due to the aforementioned limitations. However, the data quality appears to be acceptable given the results of psychometric properties of SABAS and BSMAS concur with the findings of previous studies [15,18,27,29]. Nevertheless, future studies need to consider using control check questions to further ensure data quality.

In the present study, the Thai SABAS and BSMAS had very good psychometric properties and can be used to assess internet-related addictions (i.e., smartphone addiction and SMA) among Thai university students. Moreover, the measurement invariance of the SABAS and BSMAS across gender was supported. Therefore, the SABAS and BSMAS can be additional tools for Thai healthcare providers and researchers to use to assess smartphone addiction and SMA. However, the present findings were obtained using a university student sample. Therefore, it is recommended that the psychometric properties of the Thai SABAS and BSMAS be evaluated using more diverse population groups.