The P-E fit theory offers a solid theoretical foundation for studying the impact of POQ on intensive smartphone use. According to the P-E fit theory (Edwards, 2008; van Vianen, 2018), when individuals are compatible (incompatible) with one or more characteristics of the work environment, fit (misfit) will occur and thus produce positive (negative) results for the individuals and organizations. The characteristic of the work environment mainly includes job, organization, vocation, team, and supervisor (Jansen and Kristof-Brown, 2006). Therefore, in terms of content, the P-E fit includes five types (van Vianen, 2018). Among them, person-job fit means that individual abilities and job demands and supplies are supposed to be fit appropriately (Edwards, 2008; Rodrigues et al., 2020). Based on this definition, POQ, which refers to the perception of individuals of their qualifications, such as knowledge, skills, abilities (KSAs), work experience, and education beyond the job requirements (Erdogan and Bauer, 2009; Chen et al., 2021; Zhang et al., 2021c), is a typical person-job misfit (Liu and Wang, 2012). Previous studies have found that POQ could bring several adverse consequences to employees (Harari et al., 2017), such as more cyberloafing (Cheng et al., 2020), high turnover (Wu and Chi, 2020), and less proactive behavior (Luksyte et al., 2020).

However, although the effect of job boredom, an unpleasant state commonly links to insufficient stimulation in the work environment (Mikulas and Vodanovich, 1993; Reijseger et al., 2013), on intensive smartphone use was well documented in the literature (Elhai et al., 2018; Wang et al., 2020), and there are relatively few studies on the association between POQ or other forms of person-job misfit and job boredom (Harju and Hakanen, 2016; Sánchez-Cardona et al., 2020). For the following reasons, we proposed that POQ could affect intensive smartphone use through job boredom.

First, the KSAs of overqualified employees far go beyond the job requirements, which indicates that the stimulation provided by the job is not enough to attract the interests of employees. The misfit between person and job will result in the boredom of employees (Reijseger et al., 2013; Harju and Hakanen, 2016). In fact, prior studies have provided preliminary evidence for the effect of POQ on job boredom (Harju and Hakanen, 2016; Sánchez-Cardona et al., 2020). Second, when individuals feel bored at work, they usually try to deal with it by distracting themselves (e.g., intensive smartphone use) rather than focusing on work tasks (Reijseger et al., 2013). As mentioned above, previous literature has given sufficient evidence on the effect of job boredom on intensive smartphone use (Elhai et al., 2018; Wang et al., 2020). In summary, overqualified individuals are likely to experience job boredom due to their qualifications exceeding the job requirements, and they will more likely to use the smartphone intensively to eliminate the discomfort caused by this misfit and boredom. Accordingly, we proposed the following hypothesis:

According to the P-E fit theory (Edwards, 2008; van Vianen, 2018), among the five types of P-E fit, the person-organization fit is the most studied one (Kristof-Brown et al., 2005; Jansen and Kristof-Brown, 2006) and mainly refers to an individual identifying with the value of the organization and having a sense of dependence on the organization (van Vianen, 2018). Based on this definition, affective commitment, as the core of organizational commitment, means identification of an individual with, participation in, emotional attachment to an organization and willingness to be a member of the organization (Meyer et al., 1993; Lin, 2010; Taylor et al., 2012; Vandenberghe et al., 2017), can be categorized as a type of person-organization fit (Kristof-Brown et al., 2005; Jansen and Kristof-Brown, 2006).

Based on the P-E fit theory (Edwards, 2008; van Vianen, 2018), the impact of person-job fit on the outcomes is most likely to be moderated by person-organization fit (Kristof-Brown et al., 2005; Erdogan et al., 2020). Along with this logic, the consequences of person-job misfit on job boredom and intensive smartphone use may be alleviated by person-organization fit too. Thus, affective commitment, as a type of person-organization fit, is likely to alleviate the job boredom caused by POQ (person-job misfit). For instance, overqualified individuals with a high affective commitment have a strong emotional attachment to the organization (Meyer et al., 1993). They may reduce the job boredom caused by the person-job misfit by shifting their attention to the organization (Harju and Hakanen, 2016). Therefore, we proposed that affective commitment could alleviate the job boredom caused by POQ and also weaken the mediation path from POQ to intensive smartphone use via job boredom. To sum up, we assumed the following hypotheses:

We surveyed data from a large state-owned enterprise in China, which is part of a larger data collection. The enterprise was mainly responsible for maritime traffic and transportation, so all employees were men and belonged to several teams. After contacting the top management of the enterprise, they expressed great willingness to cooperate and appointed a human resources staff to be responsible for the research. Before the questionnaire survey, we told team leaders and employees that the survey was only for academic research and there was no reward. They could freely decide whether to participate in the survey or not. Finally, with the help of the staff in charge, we set up a WeChat (an instant messaging app) group for each team willing to participate in the research and distributed questionnaires via those online groups.

All variables were self-reported, which might lead to common method bias (CMB; Podsakoff et al., 2003). For the sake of reducing the impact of CMB on the research results, data collection was carried out in four stages, each with an interval of 2 weeks. Specifically, POQ, age, job tenure, and education were measured at Time 1; affective commitment was measured at Time 2; job boredom was measured at Time 3; and intensive smartphone use was measured at Time 4. All participating employees were full-time men, and we ensured them that their responses were anonymous and confidential.

When we initially contacted all 530 employees (67 teams) of the organization to ask for participation, we got an 89.8% response rate. Considering the sample fitting, only the responses of all employees collected from the four waves of data were included in the final analysis. The final sample was 450 employees (62 teams), and the response rate was 84.9%. The average age of all participants was 35.36 years (SD = 7.57), and the average tenure was 13.00. Moreover, 47.1% of the participants had a bachelor’s degree or above.

According to the translation-back-translation standard, we translated all the English items into Chinese (Brislin, 1980). All the items described below were surveyed with the seven-point Likert scale unless specifically noted.

We measured POQ with nine items using a scale adapted from Maynard et al. (2006), which was commonly used in the POQ literature (Erdogan et al., 2020; Wu and Chi, 2020). A sample item was “I have job skills that are not required for this job” (α = 0.82).

We measured affective commitment using a six-item scale developed by Meyer et al. (1993). An example of the item was “This organization has a great deal of personal meaning for me” (α = 0.77).

We measured job boredom using a six-item scale (Reijseger et al., 2013). A sample item was “At my work, there is not so much to do” (α = 0.92).

A three-item scale developed by Spagnoli et al. (2019) was used to gauge intensive smartphone use. A sample item was “At my work, there is not so much to do” (α = 0.86).

According to the literature on POQ and smartphone use, we controlled age, education level, and job tenure of employees (Kwon et al., 2016; Zhang et al., 2016; Barnes et al., 2019; Busch and McCarthy, 2021).

The descriptive statistics and correlations between variables are exhibited in Table 1. As shown in Table 1, POQ was positively related to job boredom (r = 0.21, p < 0.01) and intensive smartphone use (r = 0.13, p < 0.01). Job boredom was positively related to intensive smartphone use (r = 0.17, p < 0.01).

Since all variables were self-reported by employees (i.e., POQ, affective commitment, job boredom, and intensive smartphone use), confirmatory factor analyses (CFA) were performed using Mplus 8.3 to test the discrimination validity of variables. Considering the number of observed indicators, we adopted the random packing method to pack the items, which avoids non-convergence issues and improves the reliability of indicators (Hall et al., 1999; Nasser and Wisenbaker, 2003). POQ, affective commitment, and job boredom were all randomly grouped into three items. As demonstrated in Table 2, the hypothesized four-factor model fitted better to data [χ²(48) = 83.05, p < 0.001, CFI = 0.99, TLI = 0.98, RMSEA = 0.04, SRMR = 0.04] than alternative models. The results of CFA demonstrated that our variables were well differentiated.

Considering the nested nature of the data, we conducted a one-way random ANOVA of the outcome variable (intensive smartphone use) to determine whether the multilevel analysis is required. The results revealed that variance in intensive smartphone use was significant [F(61, 388) = 1.59, p < 0.01], and the intra-class correlation (ICC 1) was 0.08, which manifested a clear nest structure. Therefore, we performed multilevel analysis using Mplus 8.3 to estimate the research model.

First, we tested the simple mediation model (Hypothesis 1) by specifying intensive smartphone use as the dependent variable. When testing the mediating effect, we used the original data directly to test the main effect of Level 1 predictors, which was considered reasonable (Hofmann, 1997). Second, in testing the moderation model (Hypotheses 2 and 3), as POQ and affective commitment were both at Level 1, they were group-mean centered to analyze the moderation effect (Enders and Tofighi, 2007). The moderation effect was further elaborated through simple slope analyses (Aiken and West, 1991). To test the moderated mediation effect (Hypothesis 3), we calculated the indirect effect of POQ on intensive smartphone use through job boredom when the affective commitment was above and below 1 SD, respectively (Edwards and Lambert, 2007).

For the test of Hypothesis 1, we analyzed the simple mediation model, and all results are shown in Table 3. POQ was significantly and positively associated with job boredom (B = 0.24, p < 0.01), which in turns lead to more intensive smartphone use (B = 0.12, p < 0.01). The indirect impact of POQ on intensive smartphone use via job boredom was 0.03, 95% CI = [0.01, 0.05]. Thus, Hypothesis 1 was supported.

Hypothesis 2 proposed that affective commitment weakened the positive impact of POQ on job boredom. The interaction effect (Model 2 in Table 3) of POQ and affective commitment on job boredom were negative (B = –0.16, p < 0.01). Moreover, we plotted the moderating effect of affective commitment in Figure 2 (Aiken and West, 1991). Figure 2 displays that POQ had a weaker positive effect on job boredom when affective commitment was high (+ 1 SD; B = –0.03, p = 0.30) than low (–1 SD; B = 0.27, p < 0.01), thus supporting Hypothesis 2.

Hypothesis 3 posited the moderated mediation effect, in which the indirect effect of POQ on intensive smartphone use via job boredom was moderated by affective commitment. As presented in Table 4, compared with low affective commitment (indirect effect = 0.04, p < 0.01), when affective commitment was high (indirect effect = 0.00, p = 0.30), the indirect effect of POQ on intensive smartphone use was weaker. The difference of indirect effect between high and low affective commitment was –0.04, 95% CI = [–0.09, 0.00]. Thus, Hypothesis 3 was supported.

This research contributes to the literature in the following two ways. First, we investigated the formation process of intensive use from a novel lens. A recent review (Busch and McCarthy, 2021) pointed out that previous studies on the antecedents of intensive smartphone use either focused on personal factors or environmental factors separately, ignoring the possible critical role of the P-E fit. The investigation was not comprehensive because individual behaviors are also affected by continuous interactions between individuals and environments (van Vianen, 2018). In addition, previous literature on the antecedents of intensive smartphone use has not conducted in-depth research on the formation mechanism of why individuals use smartphones frequently (Horwood and Anglim, 2019; Porter et al., 2020; Scott et al., 2020). Based on the P-E fit theory (Kristof-Brown et al., 2005; Edwards, 2008; van Vianen, 2018), the examination of POQ and the effect of job boredom on intensive smartphone use enrich the research on the antecedents and formation mechanism of intensive smartphone use from a more interactive perspective.

Second, we further contributed to the literature by answering when POQ would have an impact on job boredom and intensive smartphone use. Previous studies on the antecedents of intensive smartphone use have not reached a consistent conclusion (Busch and McCarthy, 2021), which may be due to the lack of exploration of boundary conditions (Servidio, 2019; Taylor et al., 2020; Wang et al., 2020). Research has shown that job crafting or meaningful work could alleviate the impact of POQ on job boredom (Sánchez-Cardona et al., 2020) while ignoring the impact of the interaction of different fits/misfits on employees, which is urgently called for an investigation (Kristof-Brown et al., 2002; Jansen and Kristof-Brown, 2006). Based on the P-E fit theory (Kristof-Brown et al., 2005; van Vianen, 2018), this study explores the complementary effect of person-organization fit (affective commitment) and person-job misfit (POQ) on job boredom and intensive smartphone use, which also enriches our understanding of the joint effect among different (mis) fits within the P-E fit framework.

This study also has two important practical implications. First, considering the harm of the intensive use of smartphones, some scholars put forward suggestions to prohibit the use of smartphones in the workplace (Busch and McCarthy, 2021). This research reveals that overqualified individuals frequently use smartphones due to job boredom. Therefore, reducing the boredom of employees or even preventing them from being bored is the key to the problem of the overuse of smartphones. Otherwise, even if the use of smartphones is prohibited, employees will kill boredom in other ways (e.g., cyberloafing, doodling, and fidgeting). Given overqualified individuals are more likely to experience job boredom (Kim et al., 2021), organizations should select qualified employees rather than overqualified ones. For existing overqualified employees, the organization should assign them more challenging tasks (Kim et al., 2021) to improve their person-job fit and to decrease their job boredom. In addition, previous studies have shown that having more job resources could reduce the job boredom of employees (Harju et al., 2016, 2018; Toscanelli et al., 2021). For instance, giving job autonomy for overqualified employees (Toscanelli et al., 2021) or spaces for job crafting (Harju et al., 2016) could provide them with needed job resources (van Hooff and van Hooft, 2014) and reduce their job boredom, and, finally, lower their frequency of using smartphones.

Second, this study shows that affective commitment reduced the intensive use of the smartphone for overqualified employees. Thus, for those overqualified employees to lower their job boredom and intensive use of smartphones, the organization should carry out strategies to improve their affective commitment, such as providing organizational support (Stinglhamber and Vandenberghe, 2003), reducing their work-family conflict (Wayne et al., 2013), and supervisory mentoring (Lapointe and Vandenberghe, 2017).

Although our research has several contributions, there are still some deficiencies that need to be further improved in the future. First, although we identified the mechanism by which POQ affects intensive smartphone use from the perspective of the P-E fit, other mechanisms may also exist (e.g., self-efficacy and relative deprivation), which require further examinations. Meanwhile, this study regards affective commitment as a situational factor to alleviate job boredom of employees. However, we did not further explore the mechanism by which affective commitment affects job boredom. Previous studies have shown that employees with high affective commitment could carry out more job crafting (Naeem et al., 2021), which, in turn, reduces job boredom (Harju et al., 2016). Future research could further explore the mediating mechanism between affective commitment and job boredom. Moreover, as we conceptualized affective commitment as a person-organization fit in this research, future studies could also consider the impact of the other three types of fit on the effect of POQ (person-job misfit).

Second, despite that our data were nested, we analyzed the data at the individual level while controlling the variation at the team level. It is worth noting that the use of smartphones may also be related to the characteristics of the work team or even the organization. Future research could perform multilevel analysis using nested data to explore the impact of the team or organizational factors on intensive smartphone use. Moreover, we collected data over multiple periods, but reverse causality is still possible. Thus, future research is supposed to apply a longitudinal design or experimental design to get a stronger causal relationship between POQ and intensive smartphone use.