The Psychology of Smartphone: The Development of the Smartphone Impact Scale (SIS)

Smartphone Overuse and Smartphone Addiction

Smartphone overuse, also referred to as problematic smartphone use (PSU), is defined as spending too much time using a smartphone, to the extent that it negatively affects different aspects of people’s lives (Ding & Li, 2017). Accordingly, a number of negative outcomes have been associated with smartphone overuse. Smartphones are a primary driving distraction, and the most recent estimates indicate that these devices cause at least one of every four accidents on U.S. streets (National Safety Council, 2015). Prolonged smartphone sessions can provoke musculoskeletal damage, including back and neck pain (H.-J. Kim & Kim, 2015; Y.-G. Kim, Kang, Kim, Jang, & Oh, 2013; Park et al., 2015). Smartphone addiction was found to negatively affect academic performance (Hawi & Samaha, 2016; Samaha & Hawi, 2016). The more the time spent on smartphones, the lower the grade point average of college students, even after controlling for key demographic and psychological variables (Lepp, Barkley, & Karpinski, 2015). Moreover, Rozgonjuk, Saal, and Täht (2018) found that, in contrast with a deep approach oriented to understanding, a surface approach to learning was associated to PSU and frequency of social media use during university lectures.

The likelihood of smartphone overuse and the adverse consequences that result from it have led scholars to focus on the phenomenon of PSU, generally labeling it as a behavioral (i.e., nonchemical) addiction. According to Marlatt, Baer, Donovan, and Kivlahan (1988), behavioral addictions are repetitive habit patterns associated with personal and social problems that might be subjectively experienced as out of one’s control. Behavioral addictions that concern human–machine interaction are labeled technological addiction (Griffiths, 1995). Smartphone addiction falls under this wide range of phenomena, and it can be defined as excessive smartphone use that can interfere with users’ daily lives (King & Dong, 2017; Lee, Chang, Lin, & Cheng, 2014). In this sense, Lin et al. (2016) have recently proposed a set of diagnostic criteria of smartphone addiction that include the following: (a) six symptom criteria (e.g., manifestation of withdrawal, presence of tolerance); (b) four functional impairment criteria (e.g., use of smartphone in hazardous situations, impairment of social relationships due to smartphone use); and (c) an exclusion criteria (i.e., smartphone addiction should not be accounted for by obsessive–compulsive disorder or by bipolar I disorder). However, there is an ongoing debate about whether technological devices can be considered addictive, and issues concerning the definition, etiology, course, diagnosis, and treatment of the supposed technological addictions are far from being addressed (for an overview, see Gentile, Coyne, & Bricolo, 2013). Moreover, research on this topic should rely on the operational definition of behavioral addiction (and its related exclusion criteria) given by Kardefelt-Winther et al. (2017) to avoid overpathologizing common behaviors (such as smartphone overuse) as behavioral addictions. Research on smartphone use is just at the beginning stage, and several processes should be investigated before considering it a possible behavioral addiction, including the identification of a neurophysiological basis for overuse.

Thus, we noted that the current self-report scales on the impact of smartphones in everyday life were primarily developed to measure the possible dimensions of smartphone addiction, with the aim of providing a practical tool to detect individuals at risk of smartphone overuse. Indeed, the Smartphone Addiction Scale (SAS; Kwon, Lee, et al., 2013), the Smartphone Addiction Inventory (SPAI; Lin et al., 2014), and the Smartphone Addiction Proneness Scale (SAPS; D. Kim, Lee, Lee, Nam, & Chung, 2014) focus on the measurement of PSU. Among others, these scales measure the two pharmacological criteria typical of addiction, as defined by the Diagnostic and Statistical Manual of Mental Disorder–Fifth edition (American Psychiatric Association, 2013), namely, tolerance and withdrawal. Although developed in Asian countries, the SAS and the SPAI have been adopted worldwide to measure smartphone addiction, demonstrating good psychometric properties across different cultures (Demirci, Orhan, Demirdas, Akpinar, & Sert, 2014; Lopez-Fernandez, 2017; Pavia, Cavani, Di Blasi, & Giordano, 2016). In addition to the SAS, SPAI, and SAPS, two other scales have recently been developed to assess specific aspects of human–smartphone interaction: the “Young Adult Attachment to Phone Scale” (YAPS; Trub & Barbot, 2016) and the “Nomophobia Questionnaire” (NMP-Q; Yildirim & Correia, 2015). These scales both measure smartphone attachment from different perspectives, again, using an addiction-oriented approach. Although these early scales have paved the way to measuring the psychological impact of smartphones, their framework is deeply rooted in the psychiatric tradition of substance use and dependence.

The adverse consequences of smartphone overuse are not negligible; indeed, a recent systematic review (Elhai, Dvorak, Levine, & Hall, 2017) has shown that PSU is related to depression, anxiety, stress, and low self-esteem, ranging from small to medium effects. Moreover, a recent study showed that PSU is associated to posttraumatic stress disorder symptoms, in particular with two symptom clusters, namely negative alterations in cognitions and mood and physiological arousal symptoms (Contractor, Frankfurt, Weiss, & Elhai, 2017). Still from an addiction perspective, scholars have shown links between PSU and the personality traits typically associated with substance use, such as impulsivity (Roberts, Pullig, & Manolis, 2015; Ryu & Cho, 2015) and specific Big Five traits. In particular, Lane and Manner (2011) showed that more extraverted individuals were more likely to own a smartphone and to use it for texting, whereas more agreeable ones preferred calling than texting. In a study on WhatsApp use, Montag et al. (2015) found that daily use of this instant messaging app was positively associated with extraversion and negatively with conscientiousness. Finally, Roberts et al. (2015) found that smartphone addiction was positively predicted by emotional instability, whereas conscientiousness seemed to negatively predict smartphone addiction via impulsivity.

However, limiting the perspective to addiction neglects other potentially significant psychological dimensions. Indeed, the available current research suggests a variety of psychological and social effects from smartphones that extend beyond the adverse consequences that are strictly related to the construct of addiction.

Moving Beyond the Addiction-Oriented Approach

Given the pervasiveness and versatility of smartphones, dividing the world into addicted versus nonaddicted users seems simplistic and inadequate to describe the different nuances in which smartphones have an impact on people’s everyday life. Smartphone use can be seen through the lenses of a recent conceptualization originated from the literature on Internet addiction, namely the Compensatory Internet Use Theory (CIUT; Kardefelt-Winther, 2014). The CIUT combines the research traditions on Internet addiction and motivations for Internet use, showing an interesting perspective that can account for the relationship between offline and online life. Specifically, Kardefelt-Winther (2014) claims that motivation is a key factor that mediates the link between psychosocial well-being and Internet use, so that the latter is seen as a coping strategy by which people try to compensate their psychosocial problems or unmet real-life needs. The actual advantage of the CIUT is that it can account for both positive and negative outcomes deriving from Internet use. Indeed, excessive Internet use is not seen as a compulsive, pathological phenomenon, but as a compensatory strategy that allows to cope with negative life events and that can also increase well-being.

Existing scales on smartphone addiction described in the previous paragraph are focused on detecting only dimensions that are clearly and uniquely related to an addiction-oriented approach to PSU, but there are a number of other smartphone impact dimensions that can be associated with both positive and negative outcomes of smartphone use. Given that smartphones made the Internet ubiquitous and that most of the activities on these devices need Internet connection, smartphone use can be interpreted through the CIUT. Indeed, the CIUT can offer guidance to explore psychosocial impacts of smartphones in everyday life that remain poorly investigated and need proper measurement. Scholars have just begun to look at a series of cognitive (i.e., smartphone anthropomorphization and awareness about adverse consequences related to overuse), affective (i.e., regulation of inner states), social (i.e., preference for online vs. offline communications and enhancement of romantic relationships), and behavioral (i.e., usefulness in daily activities) impacts of smartphones.

From a cognitive perspective, the literature has well documented how people are prone to anthropomorphize (i.e., ascribe humanlike characteristics) nonhuman agents (Epley, Waytz, Akalis, & Cacioppo, 2008), highlighting the positive and negative consequences of parasocial relationships that might arise from such a phenomenon. Observing this phenomenon from the CIUT perspective, the lack of chances to interact with others might motivate people to compensate their need for social connection by using their smartphone as a social surrogate, ascribing human features to it. Although it has been demonstrated that there is a tendency to anthropomorphize technological devices (Luczak, Roetting, & Schmidt, 2003), research on this topic is limited to the study of W. Wang (2017), who found that chronic loneliness was weakly and positively associated with a proneness to anthropomorphize smartphones. Finally, despite the focus on negative aspects, existing scales did not account for individual awareness about the adverse consequences of smartphone pervasiveness. This dimension might be particularly important when PSU is detected because it might allow researchers to distinguish those individuals who recognize it as an actual problem (and might look for help) from those who did not realize it at all and must first become aware of it. Indeed, being aware of smartphone negative impact might constitute a CIUT-like motivation to reduce smartphone use or to keep it under control.

Smartphones and their functionalities might be used to regulate internal states and, specifically, to cope with negative emotions, escape from boring situations, or avoid unpleasant activities. Within the CIUT framework, this means that people might find a way to compensate their negative inner states by using the smartphone (e.g., playing games or surfing the Internet to distract themselves from offline problems). Moreover, the use of smartphone to regulate emotions was theorized by Billieux (2012), who described four pathways that can originate problematic mobile phone use, one of which (i.e., the impulsive pathway) is mainly driven by maladaptive emotion regulation. In line with this interpretation, Grellhesl and Punyanunt-Carter (2012) have found that the second most reported reason for texting was “relaxation and escape,” whereas “socialization and affection,” supposedly the main functionality of texting, was rated as the sixth reason. Consistently, Panova and Lleras (2016) have found that smartphone use after an experimentally induced anxiety situation initially reduced the perceived stress. Moreover, Hoffner and Lee (2015) asked their participants to imagine that they lost their smartphones; the authors reported that the extent to which the participants missed common smartphone functionalities (i.e., interpersonal contacts, social support, entertainment, and information) was positively associated with using the device as a method of coping with negative emotions.

From a social perspective, recent studies have shown that loneliness and social anxiety are positively related to smartphone addiction (Bian & Leung, 2015; Enez Darcin et al., 2016). This phenomenon aligns with the study by J.-H. Kim (2017), who found that lonely individuals tend to rely more on smartphone-mediated communication than face-to-face interactions to alleviate their loneliness, both increasing the likelihood to develop PSU and decreasing their perceived social support. Thus, from a CIUT perspective, loneliness can motivate people to find a way to connect with other people through their smartphone. Moreover, individuals who experience high anxiety in real-life interactions, might prefer virtual and more “safe” ways to connect with others, which are made easily available by smartphones (i.e., instant messaging or social networking apps). Although the SAS (Kwon, Lee, et al., 2013) includes a dimension that addresses the importance of smartphone-mediated relationships (i.e., Cyberspace-oriented relationships), no measures that directly compare online and offline communication with others have been developed yet. Currently, the literature has particularly focused on phubbing, which is snubbing face-to-face interactions in favor of those mediated by a smartphone, showing that “phubbed individuals experience a sense of social exclusion” (David & Roberts, 2017, p. 155) that has harmful consequences on romantic relationships (Roberts & David, 2016; X. Wang, Cie, Wang, Wang, & Lei, 2017). Nevertheless, new technologies can also improve sociability when they are used to support deep offline relationships (Waytz & Gray, 2018), such as intimate ones. According to Billieux (2012), maintenance of relationships with partner, family, and friends represents another path that can lead to problematic mobile phone use, typical of individuals with low self-esteem and insecure attachment, who need to be constantly connected with significant others to obtain reassurance. However, the positive effects of a smartphone in developing and maintaining a relationship with a partner are often neglected and lack proper measurement, even though phone calls and texting (within the context of a couple) were found to be positively related to the quality of a relationship (Morey, Gentzler, Creasy, Oberhauser, & Westerman, 2013).

As for romantic relationships, smartphones can support daily activities (e.g., increasing efficiency in studying/working, helping to recall information) and users might be strongly motivated to use their device to facilitate daily tasks; however, the focus on PSU has led to neglecting the positive effects and improvements that this device has brought into everyday lives. These behavioral aspects deserve to be investigated and measured because they might play a fundamental role in smartphone usage.

The Present Research

The novelty of the smartphone phenomenon, the predominance of an addiction-oriented theoretical approach, and the resultant focus on the adverse consequences of PSU have driven research to a partial view of the psychosocial impact of smartphones. Although a growing number of studies are expanding the boundaries of the fragmented literature currently available, there are a lack of instruments able to properly address and measure different smartphone impacts. Hence, this obstructs the possibility of assembling an overview of the smartphone phenomenon and systematically studying it. To broaden the perspective of existing measures from the theme of dependence to a wider range of impacts, the present research (in line with the CIUT framework; Kardefelt-Winther, 2014) developed a brief, comprehensive, psychometrically valid measure that can account for the cognitive, affective, social, and behavioral impacts of smartphones in everyday lives. More specifically, the Smartphone Impact Scale (SIS) aimed at (a) improving the existing measurement of PSU and (b) introducing dimensions of impact that move beyond the addiction perspective. Such an instrument could provide researchers with a comprehensive tool to investigate human–smartphone interactions.

Study 1

Study 1 was designed to develop a preliminary version of the Smartphone Impact Scale (SIS-PV). Specifically, an item pool was generated to cover different aspects of smartphone impact, and it was administered to a representative sample of smartphone users. The specific aims of Study 1 were to (a) identify the latent dimensions of the impact underlying the item pool, (b) evaluate the dimensions’ reliability, and (c) investigate the convergent validity with an existing measure of smartphone addiction, namely the SAS (Kwon, Lee, et al., 2013).

Results

Dimensionality of the SIS-PV

All analyses were conducted using SPSS software, version 24.0 (IBM Corp., 2016). No missing data were detected on the SIS-PV items. A parallel analysis (Horn, 1965) was initially used to identify the maximum number of latent factors that could be extracted from the item pool. Specifically, the eigenvalues extracted from 100 simulations were compared with those obtained empirically, retaining only the factors associated with an eigenvalue higher than the 95th percentile of the simulated ones. The results, reported in Table 1, indicated retaining eight factors. Then, an eight-factor exploratory factor analysis (EFA) was conducted using the principal axis factoring method with Promax oblique rotation. The solution explained 35.30% of the total variance. Factor correlations ranged from |.02| to |.60|, revealing high associations among some factors and confirming the adequacy of the oblique rotation.

OPEN IN VIEWER

Table 1.

The First 10 factors Extracted by the EFA.

	Empirical eigenvalue	Percentage of explained variance	Percentage of cumulative variance	95th Percentile of the simulated eigenvalues
Factor 1	32.37	18.71	18.71	2.75
Factor 2	7.19	4.16	22.86	2.64
Factor 3	5.53	3.20	26.06	2.58
Factor 4	4.37	2.53	28.59	2.51
Factor 5	3.37	1.95	30.53	2.46
Factor 6	2.83	1.64	32.17	2.43
Factor 7	2.76	1.59	33.76	2.38
Factor 8	2.66	1.54	35.30	2.33
Factor 9	2.22	1.28	36.59	2.30
Factor 10	2.01	1.16	37.75	2.26

Note. EFA = exploratory factor analysis. Empirical eigenvalues and percentages of explained variance are reported along with the 95th percentile of the simulated eigenvalues.

Item loadings were evaluated by computing the marker index (Gallucci & Perugini, 2007), an item selection method that compares primary and secondary loadings to avoid cross-loadings and reach the clearest solution. A total of 67 items exceeded the recommended marker index cutoff of .40 (Gallucci & Perugini, 2007); hence, they were considered to be good indicators of their respective factors. The results, which are reported in Table 2, confirmed that each item had a clear primary loading on one factor and secondary loadings close to zero on the other factors. A further eight-factor EFA conducted on the 67 selected items explained 45.95% of variance, with factor correlations that ranged from |.05| to |.52|. The identified factors are described below, in order of the explained variance. Factor 1 was loaded by 16 items that refer to the use of smartphone as a method of coping with negative internal states, thus, it was called “Emotion regulation through smartphone usage.” Factor 2 was loaded by 10 items related to either the inability to be without smartphone or the negative consequences deriving from problems with smartphone’s functionalities; thus, it was called “Nomophobia.” Factor 3 included 14 items that refer to a great amount of time spent on smartphone, its interference with daily life and social relationships, and failures in controlling smartphone usage; thus, it was called “Loss of control of smartphone use.” The six items that loaded on Factor 4 recall the awareness of the negative influences of smartphones in everyday life and the positive consequences that are derived from smartphone nonuse; thus, it was called “Awareness of smartphone negative impact.” The five items included in Factor 5 refer to the ascription of human-like characteristics to smartphones; thus, it was called “Smartphone anthropomorphization.” Factor 6 gathered three items related to smartphone usefulness in creating and maintaining sentimental relationships; thus, it was called “Smartphone support to romantic relationships.” Factor 7 was loaded by six items that refer to the preference for communications through smartphones rather than face-to-face interaction; thus, it was called “Smartphone-mediated communication.” Finally, the seven items from Factor 8 focused on the usefulness of different smartphone functionalities in everyday life; thus, it was called “Smartphone tasks support.”

OPEN IN VIEWER

Table 2.

Eight-Factor Solution of the EFA in Order of Explained Variance.

	F1, ER	F2, LC	F3, NP	F4, SC	F5, AN	F6, TS	F7, SR	F8, SA	MI
Percentage of explained variance	22.9	5.4	4.7	3.2	2.7	2.5	2.4	2.1
When I’m very upset, the smartphone helps me get better	.88	−.09	.02	−.07	.03	.01	.01	.01	.83
When I’m stressed, the smartphone helps me feel better	.88	.07	−.13	.03	−.03	.00	−.03	−.05	.81
When I’m anxious, using my smartphone helps me feel better	.87	.04	−.02	.00	−.01	.03	−.04	.02	.82
If I’m sad, using the smartphone makes me feel better	.82	−.04	−.09	−.04	−.03	−.07	.06	.04	.73
If I’m very tense, using my smartphone helps me feel better	.82	.00	.06	.04	−.01	.03	−.07	−.04	.79
When I’m angry, using smartphone makes me feel better	.81	.03	−.08	.07	−.01	−.03	−.01	−.02	.78
If I’m aloof for something, using the smartphone makes me feel better	.81	.01	.02	−.03	.01	−.08	.01	.06	.76
When I feel depressed, using my smartphone helps me feel better	.80	−.11	.05	−.02	.01	.05	−.05	.01	.77
When I feel pressured, using the smartphone makes me feel better	.79	−.02	.08	−.06	−.03	−.01	.01	−.02	.77
If something worries me too much, I use my smartphone to make myself feel better	.78	.07	−.05	.11	.03	−.03	−.05	.03	.72
When I’m nervous, using my smartphone makes me feel better	.77	.01	.00	−.06	.02	−.02	.06	.05	.69
When I’m feeling down, using the smartphone helps me feel better	.77	.06	−.08	.00	.05	.09	−.03	.06	.74
If I had a hard day, using the smartphone helps me relax	.75	.13	.03	−.08	−.02	−.02	−.07	.02	.70
In highly ansiogenic situations, using smartphones helps me feel better	.67	.00	.11	.04	−.05	.04	−.08	−.07	.64
When I say or do something wrong with other people, using the smartphone helps me get on	.65	−.08	.03	.10	.03	−.02	.13	.04	.57
When I get bored, the smartphone keeps me entertained	.47	−.04	.11	.06	−.10	.06	.13	−.05	.44
Others tell me I spend too much time on the smartphone	−.04	.84	−.14	.02	−.11	.12	−.09	.09	.64
Sometimes I have discussions with those around me about my excessive use of the smartphone	−.14	.74	−.09	.02	−.07	.08	−.09	.16	.52
People around me often find my use of the smartphone excessive	−.04	.71	−.12	−.05	−.08	.10	−.03	.13	.50
Even when I think I should stop, I still use my smartphone too much	.15	.59	.02	.06	.08	−.01	−.01	−.12	.54
I do not realize how much time I spend on my smartphone	.17	.53	−.02	−.08	.10	.06	.05	−.06	.50
I have to check my smartphone often, even when I have conversations	.00	.53	.13	−.02	−.02	−.08	.04	.08	.54
I have tried several times to reduce the time spent on the smartphone but I did not succeed	.07	.52	.04	.02	.15	.01	−.02	−.01	.56
I spend a lot of time using my smartphone	.11	.46	.10	−.03	.22	−.03	.10	−.12	.47
I can keep track of the amount of time I use the smartphone	−.01	−.46	−.03	.00	−.01	.12	−.05	.03	.54
I can spend as much time as I want without checking my smartphone	.02	−.45	−.29	.02	.03	.10	.01	.04	.44
If I find out my smartphone is affecting my life, I would not have any problems reducing the time using it	.06	−.45	−.07	−.09	.08	.09	−.01	.17	.50
When I’m with my friends, I often feel the need to look at the smartphone	.06	.43	.14	.00	−.08	−.07	.06	−.05	.40
I cannot stop using the smartphone even when there are more important things to do	.06	.38	.10	.01	.11	−.04	.08	−.01	.49
When I walk down the street I do not pay attention because of the smartphone	.04	.35	.05	.01	.12	.06	.06	.05	.42
I would panic if I realized I had forgotten the smartphone at home after going out to go to school/university/work	.05	−.07	.79	−.05	.00	.03	−.01	−.07	.80
If the smartphone turns off, I feel lost	.01	−.03	.72	.00	.04	.06	−.06	.14	.66
I’m terrified at the idea of losing my smartphone	−.02	−.04	.67	.07	.05	.09	.02	−.01	.65
I never leave home without the smartphone, not even for short periods	−.05	.03	.64	−.04	.00	−.01	−.09	.03	.65
If my smartphone has a problem (e.g., it does not turn on, it breaks) it is the only thing I can think about	.01	.07	.60	.04	.00	.01	−.03	.08	.61
I’m scared to find myself alone in an open natural space (e.g., in the mountains, in a woods, open country) with an uncharged smartphone	.06	−.08	.59	−.04	−.03	−.11	.02	−.04	.62
I would be able to spend a whole day without smartphone	.13	−.14	−.55	.01	.10	.16	.04	−.13	.43
Losing the smartphone would be worse than losing the wallet	.04	.04	.49	.09	−.04	.06	−.03	−.07	.50
I’m very annoyed when there is no signal or I have no connection	.11	.07	.48	.03	−.04	.10	.01	−.02	.46
I always carry my smartphone with me	−.08	.13	.44	−.05	−.02	.02	.08	−.02	.50
I prefer smartphone communications because I can interrupt it at any time, unlike those face-to-face	−.05	−.03	.04	.77	−.07	.07	−.03	.07	.61
I prefer smartphone communications because you can decide if and when to intervene, unlike those face-to-face	.00	.03	−.01	.76	−.04	.04	−.06	−.02	.63
I prefer to talk about my feelings via smartphone than face-to-face	.05	−.02	−.03	.73	.04	−.05	.07	−.01	.62
I prefer to talk about my problems via smartphone than face-to-face	.04	.07	−.07	.70	.06	−.06	−.01	−.03	.58
I can be more sincere with someone using the smartphone than I can face-to-face	.08	−.03	.07	.65	−.03	.01	−.01	−.08	.53
I find it easier to keep virtual relationships than face-to-face relationships	−.05	.02	.00	.60	.01	−.02	.08	.13	.50
If I do not use my smartphone, I feel more relaxed	−.07	−.05	−.02	.01	.76	.07	−.01	−.01	.60
When I do not use the smartphone, I feel more serene	.11	−.08	−.06	−.02	.67	.04	.04	.08	.54
When I do not use the smartphone, I feel better	.00	−.08	−.08	.06	.64	−.01	−.08	.01	.52
The smartphone is an overwhelming device	−.16	.14	−.03	.01	.57	.01	.07	−.05	.49
I felt better when I had a normal mobile phone	−.03	.15	−.01	.00	.51	−.19	−.09	.01	.45
When I use the smartphone too much, I feel dizzy	.05	.02	.09	−.09	.46	.06	−.01	.00	.51
The smartphone helps me remember what I have to do	−.04	−.11	.05	.04	.15	.64	.01	−.03	.50
Without my smartphone I would not be able to remember my appointments	−.10	−.03	.08	.06	.19	.61	−.03	−.01	.51
The smartphone makes me faster and more efficient in work and/or study	.04	−.04	−.14	.01	−.07	.60	−.06	.05	.51
My performance in studying/working has improved since I used the smartphone	.08	−.01	.02	−.02	−.07	.55	−.03	.07	.45
My smartphone helps me perform tasks faster	.04	.07	.01	−.13	−.05	.52	.01	−.09	.44
The smartphone helps me in the day-to-day activities	−.06	.15	.06	.00	−.13	.44	.10	−.08	.42
The smartphone helps me carry out tasks that I would otherwise not know how to complete	.05	.13	−.08	.02	−.09	.44	.12	−.06	.46
The relationship with my partner would be affected by the absence of the smartphone	−.14	.05	−.07	.02	−.01	.02	.81	.07	.54
The smartphone helped me (or helps me) keep my relationship alive	.07	−.05	−.06	.01	−.07	−.03	.78	.03	.65
An important part of my relationship with my partner comes from smartphone communication	.03	.00	.04	−.01	.05	.03	.77	−.01	.63
The smartphone is conscious	−.02	.02	.10	−.04	.09	.03	.02	.67	.45
The smartphone perceives the stimuli that I send	.16	−.08	.10	−.01	.06	.05	.03	.59	.41
The smartphone feels emotions	−.04	.04	−.02	.11	−.03	−.05	−.03	.54	.44
The smartphone comes with free will	.04	.13	−.08	−.02	−.03	−.08	.02	.47	.43
The smartphone understands me	.09	−.02	.03	.02	−.07	−.05	.12	.46	.43

Note. EFA = exploratory factor analysis; ER = emotion regulation through smartphone usage; LC = loss of control of smartphone use; NP = nomophobia; SC = smartphone-mediated communication; AN = awareness of smartphone negative impact; TS = smartphone tasks support; SR = smartphone support to romantic relationships; SA = smartphone anthropomorphization; MI = marker index. Loadings of the final solution on 67 items and MI of the initial solution on 173 items are reported.

Primary loadings are highlighted in bold, whereas percentage of explained variance and marker index are italicized.

Cronbach’s alpha was computed for each latent factor to assess internal consistency. The highest Cronbach’s alpha was observed for emotion regulation through smartphone usage (α = .96), followed by loss of control of smartphone use (α = .87), nomophobia (α = .86), smartphone-mediated communication (α = .86), smartphone support to romantic relationships (α = .82), awareness of smartphone negative impact (α = .77), smartphone tasks support (α = .75), and smartphone anthropomorphization (α = .68).

Convergent Validity

Scores of each SIS-PV subscale were computed, averaging the responses of the items that loaded onto their respective dimensions. Table 3 displays the means and standard deviations of the SIS-PV subscales and their correlations with the SAS dimensions. Mean scores ranged from 1.73 to 3.13, except for Smartphone anthropomorphization, which showed a very low value indicating a strong floor effect. Given the multiple correlations estimated, the Bonferroni correction was applied to the p value, dividing the ordinary cutoff of .05 for the number of coefficients estimated (i.e., 48), thus lowering the level of significance at p < .00104. List wise deletion was used to handle missing data. Loss of control of smartphone use, nomophobia, emotion regulation through smartphone usage, and smartphone-mediated communication showed the highest correlation coefficients, with all the dimensions of the SAS, suggesting that these factors were the most related to PSU. In particular, loss of control of smartphone use showed the high associations with withdrawal, tolerance, and overuse, which are typical dimensions of addiction. In contrast, awareness of smartphone negative impact showed low to zero correlations with the majority of the SAS dimensions, indicating that being aware or not aware of the smartphone’s negative influence on everyday life was relatively independent from the problematic use of the device. The other SIS-PV dimensions showed low to moderate correlation coefficients with those of the SAS.

OPEN IN VIEWER

Table 3.

Descriptive Statistics and Pearson Correlations Between the Dimensions of the SIS-PV (In Rows), in Order of Explained Variance, and the SAS (In Columns).

	M (SD)	DD	PA	WD	CR	OU	TO
Emotion regulation through smartphone usage	2.23 (0.96)	.40*	.71*	.54*	.50*	.50*	.45*
Loss of control of smartphone use	2.19 (0.74)	.55*	.44*	.67*	.55*	.68*	.78*
Nomophobia	2.90 (0.90)	.30*	.47*	.68*	.47*	.56*	.38*
Smartphone-mediated communication	1.73 (0.84)	.34*	.40*	.42*	.47*	.38*	.29*
Awareness of smartphone negative impact	3.13 (0.80)	.36*	−.13	.00	−.06	.12	.27*
Smartphone tasks support	3.02 (0.77)	.16	.40*	.27*	.23*	.33*	.18*
Smartphone support to romantic relationships	2.40 (1.18)	.31*	.34*	.33*	.31*	.33*	.22*
Smartphone anthropomorphization	1.25 (0.46)	.17	.30*	.27*	.32*	.08	.18*

Note. SIS-PV = preliminary version of the Smartphone Impact Scale; SAS = Smartphone Addiction Scale; DD = daily-life disturbance; PA = positive anticipation; WD = withdrawal; CR = cyberspace-oriented relationship; OU = overuse; TO = tolerance.

*

p < .00104.

Study 2

The present study was devised to complete the development process of the SIS. The main aim was to confirm the factor structure found in Study 1 and, based on the results and on theoretical considerations, shorten the scale to increase its usability. Once the final version of the SIS was developed, its dimensionality and psychometric properties were tested. Finally, we investigated the association between the SIS dimensions and psychosocial constructs that might be related to PSU, human–smartphone interaction, and smartphone usage regarding daily time spent on the device as well as its specific functionalities/apps. Based on the results on convergent validity reported in Study 1, we hypothesized that the dimensions more correlated to PSU (i.e., loss of control of smartphone use, nomophobia, smartphone-mediated communication, and emotion regulation through smartphone usage) would show the highest associations with personality traits commonly related to addictions (i.e., conscientiousness, neuroticism, and impulsivity), psychosocial variables negatively related to well-being (i.e., depression, anxiety, stress, loneliness, perceived social support, and threat of basic needs), smartphone daily use, and the majority of smartphone functionalities. On the contrary, the other SIS dimensions were expected to show lower associations with such personality traits and psychosocial constructs, as they measure smartphone impacts that should not be directly linked to PSU. Beyond this general prediction, specific hypotheses were advanced on the associations between the SIS dimensions and smartphone functionalities. Dimensions focused on online social relationships (i.e., smartphone-mediated communication and smartphone support to romantic relationships) were hypothesized to correlate especially with instant messaging apps and social networking sites. As a coping strategy, emotion regulation through smartphone usage was expected to correlate with functionalities that allows mind-wandering and distraction from negative life situations (e.g., music and video apps, games, navigation on the Internet, social networking sites). Functionalities that help users in daily and work activities (e.g., calendar and notes apps, e-mail, navigation systems) were hypothesized to correlate with smartphone tasks support. Awareness of negative smartphone impact was expected to show negative or no correlations with the most addictive functionalities, such as instant messaging apps, social networking sites, and games.

Results

Confirmatory Factor Analyses on the SIS-PV

Confirmatory factor analysis (CFA) was conducted on the SIS-PV to test the factor structure that emerged from Study 1 using Mplus version 7 (Muthén & Muthén, 2015). Given that 12 items of the SIS showed an asymmetry higher than |1.0| and that the Mardia test yielded significant multivariate skewness (M = 520.82, SD = 3.29, p < .001) and kurtosis (M = 4608.46, SD = 6.28, p < .001), maximum likelihood with robust standard errors (i.e., MLR) was used as estimator in the following analyses to avoid problems related to data nonnormality. Before running the CFA, a null model in which all variables were uncorrelated to one another was tested to identify the fit indices to be considered in the further analysis. Indeed, a null model with a root mean square error of approximation (RMSEA) smaller than .158 suggests that incremental fit indices (i.e., the comparative fit index [CFI] and the Tucker–Lewis index[TLI]) of any model tested on the same set of variables and sample cannot mathematically reach acceptable values; thus, they are completely uninformative (Kenny, 2015). The null model conducted on the SIS-PV yielded a RMSEA of .120, indicating that only absolute indices should be used to evaluate the fit of the following CFA: the RMSEA, the probability of close fit associated with the RMSEA (Cfit of RMSEA), and the standardized root mean square residual (SRMR). A model adequately explains the data when the RMSEA is lower than .08 (better if lower than .05), the Cfit of RMSEA is nonsignificant (better if higher than .50), the SRMR is lower than .08 (Brown, 2015; Kline, 2015). The CFA showed excellent fit indices (RMSEA = .046, Cfit of RMSEA = 1.00, SRMR = .057), confirming the factor structure found in the previous study. Table 4 reports standardized item loadings that ranged from |.44| to |.89|.

OPEN IN VIEWER

Table 4.

The Results of the CFAs Conducted on the SIS-PV and SIS: Standardized Loadings (λ) Are Reported.

	SIS-PV λ	SIS λ
Loss of control of smartphone use
I do not realize how much time I spend on my smartphone	.45	—
I have to check my smartphone often, even when I have conversations	.60	—
I can keep track of the amount of time I use the smartphone	−.53	—
Others tell me I spend too much time on the smartphone	.73	.88
People around me often find my use of the smartphone excessive	.73	.91
I have tried several times to reduce the time spent on the smartphone but I did not succeed	.63	—
If I find out my smartphone is affecting my life, I would not have any problems reducing the time using it	−.53	—
When I walk down the street I do not pay attention because of the smartphone	.44	—
I can spend as much time as I want without checking my smartphone	−.56	—
I spend a lot of time using my smartphone	.65	—
When I’m with my friends, I often feel the need to look at the smartphone	.57	—
I cannot stop using the smartphone even when there are more important things to do	.66	—
Even when I think I should stop, I still use my smartphone too much	.76	—
Sometimes I have discussions with those around me about my excessive use of the smartphone	.70	.82
Nomophobia
If my smartphone has a problem (e.g., it does not turn on, it breaks), it is the only thing I can think about	.66	.70
I’m terrified at the idea of losing my smartphone	.69	.70
I would be able to spend a whole day without smartphone	−.46	—
I always carry my smartphone with me	.44	—
I never leave home without the smartphone, not even for short periods	.50	—
If the smartphone turns off, I feel lost	.73	.74
I’m scared to find myself alone in an open natural space (eg in the mountains, in a woods, open country) with an uncharged smartphone	.49	—
I’m very annoyed when there is no signal or I have no connection	.61	—
Losing the smartphone would be worse than losing the wallet	.62	—
I would panic if I realized I had forgotten the smartphone at home after going out to go to school/university/work	.67	.66
Smartphone-mediated communication
I can be more sincere with someone using the smartphone than I can face-to-face	.67	—
I prefer smartphone communications because I can interrupt it at any time, unlike those face-to-face	.60	—
I prefer to talk about my feelings via smartphone than face-to-face	.81	.81
I find it easier to keep virtual relationships than face-to-face relationships	.72	.72
I prefer to talk about my problems via smartphone than face-to-face	.83	.86
I prefer smartphone communications because you can decide if and when to intervene, unlike those face-to-face	.74	.70
Emotion regulation through smartphone usage
When I’m feeling down, using the smartphone helps me feel better	.66	—
If something worries me too much, I use my smartphone to make myself feel better	.73	—
When I’m stressed, the smartphone helps me feel better	.78	—
When I’m angry, using smartphone makes me feel better	.81	.81
When I feel depressed, using my smartphone helps me feel better	.83	—
When I feel pressured, using the smartphone makes me feel better	.83	.82
When I’m very upset, the smartphone helps me get better	.88	—
If I’m very tense, using my smartphone helps me feel better	.88	—
When I say or do something wrong with other people, using the smartphone helps me get on	.77	—
If I’m sad, using the smartphone makes me feel better	.87	.88
When I’m nervous, using my smartphone makes me feel better	.89	.89
When I get bored, the smartphone keeps me entertained	.56	—
When I’m anxious, using my smartphone helps me feel better	.88	—
If I had a hard day, using the smartphone helps me relax	.76	—
If I’m aloof for something, using the smartphone makes me feel better	.87	—
In highly ansiogenic situations, using smartphones helps me feel better	.76	—
Smartphone support to romantic relationships
The relationship with my partner would be affected by the absence of the smartphone	.76	.76
An important part of my relationship with my partner comes from smartphone communication	.83	.83
The smartphone helped me (or helps me) keep my relationship alive	.80	.80
Smartphone tasks support
The smartphone makes me faster and more efficient in work and/or study	.54	—
The smartphone helps me carry out tasks that I would otherwise not know how to complete	.57	—
The smartphone helps me remember what I have to do	.58	.81
My performance in studying/working has improved since I used the smartphone	.65	—
My smartphone helps me perform tasks faster	.71	.50
Without my smartphone I would not be able to remember my appointments	.48	.68
The smartphone helps me in the day-to-day activities	.68	.59
Smartphone anthropomorphization
The smartphone understands me	.61	—
The smartphone perceives the stimuli that I send	.61	—
The smartphone is conscious	.79	—
The smartphone comes with free will	.61	—
The smartphone feels emotions	.66	—
Awareness of smartphone negative impact
When I do not use the smartphone, I feel better	.69	.73
When I use the smartphone too much, I feel dizzy	.49	—
If I do not use my smartphone, I feel more relaxed	.65	—
I felt better when I had a normal mobile phone	.56	.66
The smartphone is an overwhelming device	.57	.51
When I do not use the smartphone, I feel more serene	.72	.72

Note. CFA = confirmatory factor analysis; SIS-PV = preliminary version of the Smartphone Impact Scale; SIS = Smartphone Impact Scale. All the loadings are significant at the .001 level.

Scale Reduction and Dimensionality of the Final SIS

To proceed to the scale reduction, the items were evaluated from both statistical and theoretical standpoints. Specifically, for each of the factors, the items were selected based on the magnitude of factor loadings showed in the former CFA and their representativeness of the construct measured. As a trade-off between reliability and scale brevity, we aimed at selecting four items per construct. This was not possible for smartphone support to romantic relationships, because only three items were derived from Study 1. Moreover, the selection procedure led to retain only the three items of loss of control of smartphone use that were related to one’s perception about how others evaluate his or her own smartphone overuse. Finally, the strong floor effect showed by smartphone anthropomorphization in Study 1 and in the present study (M = 1.28, SD = 0.52), led us to drop this dimension. A total of 26 items (reported in Table 4) were included in the final version of the SIS. The detailed procedure followed for the item selection is reported in the supplementary material (available online) of the present article.

Dimensionality and Psychometric Properties of the SIS

As for the previous CFA on the SIS-PV, an initial Mardia test and a null model were run. Whereas the Mardia test confirmed the need to use maximum likelihood with robust standard errors (multivariate skewness: M = 32.54, SD = 0.88, p < .001; multivariate kurtosis: M = 725.85, SD = 3.14, p < .001), the null model yielded a RMSEA of .177, indicating that incremental fit indices (i.e., CFI and TLI, respectively) must be considered to evaluate model fit. Values higher than .95 of both CFI and TLI indicate optimal fit (Brown, 2015). As expected, the CFA yielded excellent fit indices (CFI = .95, TLI = .94, RMSEA = .043, Cfit of RMSEA = .99, SRMR = .043), confirming the hypothesized factor structure. Standardized item loadings ranged from .50 to .91 and are reported in Table 4. Generally, loadings were higher than those resulted from the CFA on the SIS-PV, especially those of loss of control of smartphone use, indicating the higher degree of homogeneity of this latent construct compared with its former version.

Factors reliability was estimated using McDonalds’ (1999) Omega to overcome Cronbach’s alpha limitations, specifically the violation of tau-equivalence (Dunn, Baguley, & Brunsden, 2013). Descriptive statistics, Omega (ω), and correlations among the SIS dimensions are reported in Table 5. Omega ranged from .74 to .91, indicating good to excellent internal consistency of the factors. The highest mean was observed for smartphone tasks support, and the lowest was for smartphone mediated-communication. Most correlation coefficients were significant and positive, revealing weak to moderate associations among the SIS dimensions. The highest correlation coefficient was detected between nomophobia and emotion regulation through smartphone usage (r = .50), followed by that between nomophobia and loss of control of smartphone use (r = .40). The only exception to the positive associations among factors was represented by awareness of smartphone negative impact, which did not significantly correlate with nomophobia, smartphone support to romantic relationships, and smartphone tasks support, and it was negatively associated with the remaining dimensions.

OPEN IN VIEWER

Table 5.

Descriptive Statistics, McDonalds’ Omega (ω) and Pearson Correlations Among the Dimensions of the SIS.

	M (SD)	ω	(1)	(2)	(3)	(4)	(5)	(6)
(1) Loss of control of smartphone use	2.01 (1.15)	.91	1
(2) Nomophobia	2.85 (1.04)	.79	.40**	1
(3) Smartphone-mediated communication	1.85 (0.95)	.85	.25**	.33**	1
(4) Emotion regulation through smartphone usage	2.22 (1.07)	.92	.29**	.50**	.39**	1
(5) Smartphone support to romantic relationships	2.36 (1.13)	.84	.24**	.26**	.36**	.25**	1
(6) Smartphone tasks support	3.41 (0.88)	.75	.15*	.36**	.12*	.20**	.21**	1
(7) Awareness of smartphone negative impact	3.03 (0.85)	.74	−.16*	−.11	−.14*	−.20*	−.05	−.08

Note. SIS = Smartphone Impact Scale.

*

p < .05. **p < .01.

Associations With Psychosocial Constructs and Smartphone Usage

Table 6 reports correlations of the SIS with psychosocial constructs and smartphone usage, computed using SPSS, version 24.0 (IBM Corp., 2016). Given the multiple correlations estimated, the Bonferroni correction was applied to the p value, dividing the ordinary cutoff of .05 for the number of coefficients estimated (i.e., 161), thus lowering the level of significance at p < .00031.

OPEN IN VIEWER

Table 6.

Pearson Correlations Between the Dimensions of the SIS (In Columns) and the Psychosocial Constructs and Smartphone Usage (In Rows).

	LC	NP	SC	ER	SR	TS	AN
Psychosocial constructs
Extraversion	.06	−.01	−.26*	−.06	−.07	.12	−.01
Agreeableness	−.03	−.04	−.10	−.11	−.04	−.04	.06
Conscientiousness	−.18*	−.06	−.18*	−.14	−.19*	.02	−.07
Neuroticism	.26*	.27*	.20*	.21*	.10	.01	.11
Openness	−.09	−.06	−.09	−.07	.02	.17*	.11
Impulsivity	.33*	.31*	.18*	.26*	.10	.05	.04
Depression	.12	.11	.21*	.10	.08	−.04	.19*
Anxiety	.22*	.24*	.15*	.19*	.17*	.03	.12
Stress	.19*	.18*	.14	.15	.10	.04	.16*
Loneliness	.13	.16*	.29*	.14	.17*	−.05	.05
Perceived social support	.03	.01	−.23*	−.10	−.11	−.12	.03
Threat of basic needs	.15*	.16*	.27*	.13	.12	−.04	.10
Smartphone functionalities
Phone calls and SMS	.01	.02	−.16*	−.03	−.05	.16*	−.03
Calendar and notes	.08	.18*	−.03	.05	.04	.54*	−.02
Camera	.19*	.23*	.06	.15*	.05	.17*	−.01
E-mail	.07	−.01	.02	−.04	.10	.22*	.11
Instant messaging	.21*	.21*	.13	.16*	.25*	.20*	.01
Social networks	.31*	.24*	.24*	.28*	.23*	.10	−.03
Navigation systems	.14	.14	.08	.05	.07	.19*	.07
Internet browsers	.16*	.12	.18*	.09	.17*	.16*	.00
Music and videos apps	.22*	.16*	.16*	.20*	.17*	.09	−.04
Games	.17*	.11	.21*	.22*	.10	.01	−.17*
Smartphone daily usage	.27*	.23*	.18*	.20*	.20*	.07	.00

Note. LC = loss of control of smartphone use; NP = nomophobia; SC = smartphone-mediated communication; ER = emotion regulation through smartphone usage; SR = smartphone support to romantic relationships; TS = smartphone tasks support; AN = awareness of smartphone negative impact.

*

p < .00031.

Concerning psychosocial constructs, loss of control of smartphone use, nomophobia, smartphone-mediated communication, and emotion regulation through smartphone usage showed a similar pattern of correlations characterized by positive associations with neuroticism, impulsivity, and anxiety. In addition, each of these dimensions showed specific associations with other psychosocial constructs. Loss of control of smartphone use was negatively associated with conscientiousness and positively with stress and threat of basic needs. Nomophobia showed positive correlations with stress, loneliness, and threat of basic needs. Smartphone-mediated communication was positively associated with loneliness and threat of basic psychological needs, and negatively with extraversion and perceived social support. Emotion regulation through smartphone usage did not show any other significant correlation in addition to those with impulsivity, neuroticism, and anxiety. As hypothesized, compared with these four dimensions, the other SIS factors revealed fewer and lower associations with psychosocial constructs. Smartphone support to romantic relationships was negatively linked to conscientiousness and positively to anxiety and loneliness. A positive correlation was found between smartphone tasks support and openness, whereas awareness of smartphone negative impact positively correlated with depression and stress.

Daily time spent on the device was positively associated with all the SIS dimensions, except smartphone tasks support and awareness of smartphone negative impact. Loss of control of smartphone use and nomophobia showed similar patterns of correlations with smartphone specific use. In particular, both these dimensions correlated with instant messaging apps, social networking sites, smartphone camera, and music and video apps. Moreover, loss of control of smartphone use correlated with the use of Internet browsers, whereas nomophobia with calendar and notes apps. Smartphone-mediated communication showed a negative correlation with phone calls and SMS and positive correlations with social networking sites, games, Internet browsers, and music and video apps. Emotion regulation through smartphone usage was especially associated with social networking sites and showed lower, but still significant, correlations with games, music and video apps, instant messaging apps, and smartphone camera. Smartphone support to romantic relationships was primarily associated not only with instant messaging apps and social networking sites but also with Internet browsers and music and video apps. Smartphone tasks support revealed significant and positive associations with the majority of the smartphone functionalities and an especially high correlation with calendar and notes apps. On the contrary, awareness of smartphone negative impact revealed only a negative association with games.

General Discussion

The present research aimed at broadening the view of human–smartphone interaction, developing a comprehensive and psychometrically valid self-report scale to address smartphone impact in everyday life, the SIS. To achieve this goal, we reviewed the psychological literature on smartphones, specifically focusing on smartphone cognitive, affective, social, and behavioral impacts that still lacked a proper measurement. In Study 1, we administered an item pool to a sample of adult smartphone users. We obtained a 67-item scale that measured eight factors that demonstrated good external validity with the SAS. In Study 2, this preliminary scale was reduced to obtain the SIS, a 26-item scale that measures the following seven dimensions of smartphone impact: loss of control of smartphone use, nomophobia, smartphone-mediated communication, emotion regulation through smartphone usage, smartphone support to romantic relationships, smartphone tasks support, and awareness of smartphone negative impact. The analysis of the SIS dimensionality yielded excellent fit indices and its subscales demonstrated good to excellent reliability. Moreover, its dimensions showed meaningful associations with a series of psychosocial constructs related to PSU and human–smartphone interaction, as well as with self-reported use of smartphone and its primary functionalities.

Conclusions

The present research broadened the view of human–smartphone interaction, developing a brief, comprehensive self-report scale that measures behavioral, social, affective, and cognitive impacts of smartphone in everyday life—the SIS. The SIS improves the existing measures of PSU by introducing new related factors and extends beyond the debated construct of smartphone addiction, addressing new, underinvestigated impacts that lack proper measurements. The improvements in the measurement of human–smartphone interactions introduced by the SIS can be summarized by three main points. First, we extended the concept of PSU by considering the role of smartphone in coping with negative emotions and addressing preference for online versus offline communications. Second, we accounted for the positive impacts of smartphone (i.e., supporting romantic relationships and daily activities). Third, we introduced the measurement of a cognitive dimension that could play a role in the treatment and prevention of PSU, namely awareness about the negative impacts of the device.

Overall, the SIS is potentially useful both for theoretical and applied purposes. Researchers could benefit from this comprehensive measure of smartphone impacts by considering the antecedents (e.g., personality traits, situational factors) and consequences (positive and negative psychosocial outcomes) of smartphone usage. From the applied point of view, this scale could be adopted in school settings to explore the potentials (supporting technology-mediated communication for special educational needs) and risks (assessing pervasiveness of the device in students’ life) of smartphone in developmental populations. In organizational settings, where smartphone can serve as a working instrument, the SIS might be useful to weight potentials (e.g., task support, improving communication within working groups) and negative impacts (e.g., coworkers’ metaperception of smartphone overuse) of this device. Thus, the theoretical and applied usefulness of the SIS lies in the inclusion of both problematic and potentially adaptive impacts of smartphone use.

The primary limitation of the present research is the lack of an objective measurement of smartphone use. Future research should investigate the real, unbiased associations between the SIS subscales and each of the functions/applications conveyed by smartphone as well as actual daily use, providing strongest evidence of convergent validity. Second, a complete validation of the SIS will require repeated administrations of the scale to evaluate test–retest reliability and assess the stability of the subscales’ scores over time. Third, given that our sample was self-selected and population representativeness was not assessed, future research is needed to extend the generalizability of our findings.

Beyond addressing these issues, a number of future directions were hypothesized. First, studies are needed to test whether the SIS performs differently in different populations. Specifically, the SIS dimensionality (as well as its associations with psychosocial constructs and smartphone use) should be replicated in samples with different age (e.g., digital natives vs. digital immigrants), culture (e.g., developed vs. developing countries), and levels of smartphone use. Second, longitudinal studies are needed to investigate whether and how the SIS dimensions change over time and whether these changes are predicted by specific psychosocial constructs and predict differences in smartphone use. Finally, the SIS dimensions can be used to identify clusters of individuals characterized by different profiles of smartphone impacts, detecting the typologies of users at risk to develop PSU. These profiles could be then compared with profiles derived from the two main typologies of smartphone usage, namely process (i.e., content-based consumption of media) and social (i.e., functionalities that involve interactions among users) use, to deepen the relationship between smartphone impacts and its actual use and to identify more specific categories of smartphone users (Contractor et al., 2019; Elhai & Contractor, 2018; Elhai, Hall, et al., 2017).

Overall, despite the need for future research to address the aforementioned issues and to test its applied usefulness in specific contexts, the SIS was demonstrated to be a valid and reliable self-report scale that can account for the various impacts of smartphones in everyday life over and beyond existing measures of PSU.

Supplemental Material