Proactive personality, value congruence, perceived organizational support, and problem preven- tion behavior: A reciprocal moderated mediation model

Article history: Received: May 30, 2020 Received in revised format: May 3


Introduction
The rapid development of society has led to people's increasing demand for health care. This is an opportunity and a challenge for the health sector when facing complex epidemics and a shortage of manpower. Recently, researchers have sought to increase the performance of proactive personality (PP) and factors that affect employee's proactive behavior (PB) (Griffin, Neal, & Parker, 2007;Bindl & Parker, 2011;Helmy & Wiwoho, 2020). Through factors that influence employee's behavior, managers can make optimal decisions and encourage them. From this perspective, an organization may select employees who tend to be willing to act as proactive individuals since these people are likely to control the situation and benefit not only themselves but also the operational efficiency of the organization. In addition, retaining proactive employees with the potential to become highly effective employees is recognized as an essential condition for the organization's survival and development (Shin & Jeung, 2019). A comprehensive study of PP has shown that this factor affects many positive results such as employee performance, organizational commitment, career success (Allen et al., 2012;Christian, Garza, & Slaughter, 2011;Ng & Feldman, 2012). However, Lewin (1964) argues that individual characteristics do not affect behavior as much as personal interactions with the environment. Besides, the studies of Crant (2000) and Bolino, Valcea, and Harvey (2010), show that PP has been shown to be a particularly important premise for PB. However the meta-study of (B. Fuller & Marler, 2009) reveals that PP only has a cumulative effect (β = .32) on PB. (Schneider & Reichers, 1983) pointed out that the effects of appropriate situations can motivate employees to exhibit PB regardless of their inherent personality traits. Therefore, organizations do not simply select employees who are proactive and expect them to exhibit PB, but also need to take into account the organizational environmental factors that motivate employees to exhibit PB. Proactive behavior has captivated the attention of researchers and popular press (Parker & Bindl, 2017) but it still has potential areas for further research. Parker and Collins (2008) summarize the proactive work behaviors of five categories: (1) taking charge, (2) helping, (3) voice, (4) individual innovation, and (5) problem prevention. PB such as help, taking charge, voice and individual innovation have appeared in many studies (Xu, Qin, Dust, & DiRenzo, 2019;Bayayda, 2020), but there are currently only three studies mentioning PrP: Crant (2000); Fay and Frese (2001) and Parker & Collins (2008). However, these three studies have only introduced the concept and characteristics of this behavioral classification, the empirical research on PrP is still limited. This shows that many studies have proven the important role of PB, but the understanding of PrP is still inconsistent. Moreover, today, organizations are not only concerned with the character of their employees to find personnel for specific jobs but also focus on how to harmonize personal traits and values with the organizational environment. In response to constantly changing environments and intense competition, managers are often more concerned with the flexibility of their employees to adapt to changing circumstances and their organizational commitment. Research by Kristof-Brown, Zimmerman, and Johnson (2005) and Chen, Yen, and Tsai (2014) show that POV and PSV motivate employee's behavior and thus remarkably improve employee's job performance. However, the relationship between PP and perceived value congruence as well as the role of value congruence in the connection in which PP leads to PrP has yet to be enlightened by previous researches. On the point of view of Job Demand-Resource (Bakker & Demerouti, 2007), this research creates a moderated mediation model in which POV, PSV play a mediating role in the relationship between PP and problem PB; and POS plays a moderator role in the relationship of POV and PSV mediate the relation between PP and PrP. This model provides a deeper understanding of the interaction mechanisms between personality and environment that lead to PrP and environmental regulatory mechanisms that influence the expression of PrP of PP. In particular, proactive employees are capable of exhibiting PrP if they develop an awareness of their POV and PSV. In addition, this process can be improved if employees proactively see POS. Finally, this study focuses on examining the contextual factors that can enhance or weaken the main relationship in the underlying theoretical model, the POS. The POS element represents the contextual factor that influences the process through which PP exhibits PB. Seibert, Kraimer, and Crant (2001) found that employees with PP are more likely to have career success since they tend to engage in PB. Individuals with positive behavioral tendencies regularly offer to create opportunities, improve the situation through ideas, creative suggestions on their work, and participate in the company's development activities (more or less involved) owing to their needs of attainment (Thompson, 2005) and incentives to learn (Major, Turner, & Fletcher, 2006). Research by Parker, Bindl, and Strauss (2010), and Li, Liang, and Crant (2010) show that PP exhibits PB even without the support of a situation. Therefore, this study expects that when the situation is not conducive to behavior, proactive employees will exhibit PB. PrP is considered an important factor for having a long-term focus on the work, without having to wait for the situation to occur and then address or meet that need (Parker & Collins, 2008). A long-term focus on the job allows employees to consider what is about to happen (such as new needs, new problems, and new opportunities) and opt for a proactive outlook on the situation (Crant, 2000;Fay & Frese, 2001;Parker & Collins, 2008). In other words, PrP has the ability to improve the performance of employees, thereby improving the flexibility and improving the competitiveness of the organization. Many empirical studies have demonstrated the relationship between PP and PB namely verbal action, taking responsibility, creativity, networking, and career plan, searching for response, socialization, learning, and readiness to change (Fuller & Marler, 2009;Ng & Feldman, 2012;Tornau & Frese, 2013;Spitzmuller, Sin, Howe, & Fatimah, 2015). But research on PrP is limited and almost at the level of introduction of a definition and a (Waterwall, 2019) study shows the positive impact of PP and PB (taking charge, task performance, problem prevention) is mediated by regulatory focus. This suggests that research on PrP is still left open. Thus, with a view to filling this research gap, our study concentrates on investigating the connection between PP and PrP. Proactive people have a tendency to work more effectively than submissive ones since these individuals "create situations and environments conducive to effective implementation" (Crant, 1995;Li, Wang, Gao, & You, 2017). On the other hand, according to various studies of the researchers, the PP is related to many motivational factors including psychological empowerment (Hon & Rensvold, 2006), confidence in the role itself (Parker, Williams, & Turner, 2006) and flexible duty direction (Parker, Wall, & Jackson, 1997). Emotional endorsement is a reflection of a positive orientation towards a person's working role (Spreitzer, 1995). The relationship between PP and PB has been demonstrated by studies of (Griffin et al., 2007;Parker & Collins, 2008;Searle, 2011). Therefore, hypothesis H1 is proposed as follows:

Proactive personality and Problem prevention
Hypothesis 1: Proactive personality affects problem prevention behavior.

The Mediating Role of POV
Tett and Burnett (2003) utilized trait activation theory to elaborate on how personal characteristics will manifest when there are signals related to the characteristics of a motivational situation that people exhibit behaviors. Individual behavior not only relies on external environmental factors but also depends more on what the individual sees and believes because the individual often thinks and argues before acting. Values create a foundation for knowledge of people's perceptions, attitudes, and motivations, so the value of employees is influenced by the impacts of their surroundings (for example, leadership styles), in turn, influences their behavior (Stern, Dietz, Abel, Guagnano, & Kalof, 1999). The POV is an important factor to create solidarity, performance and welfare of the organization (Vveinhardt & Gulbovaitė, 2016). The study is person-organization fit (P-O fit) framework by Kristof-Brown et al. (2005) and Chen et al. (2014) show that the value of people and the surroundings has a favorable impact on work behavior and efficiency. Chatman (1991) defines value congruence as the compatibility of organization' values and employee's values like orientation to a group or innovation. Research by (Kristof-Brown et al., 2005) reveals that value congruence exerts a constructive effect on satisfaction, commitment, intention to stay with the organization, actual revenue and performance. Edwards and Cable (2009) argues that value congruence is a value compatibility work between individuals and the entire organization, including supervisors, interviewers, colleagues and working groups. Based on attraction-selection-attrition (ASA) framework (Schneider, Smith, Taylor, & Fleenor, 1998), we propose that PP positively affects POV for two reasons. First, proactive people have higher chance to succeed in case they find the right job (Brown, Cober, Kane, Levy, & Shalhoop, 2006). These people tend to seek organizations having higher person-organization value congruence for themselves and when they feel the fit and motivation, the goals of the organization will be easier to achieve (Silpa & ChittiBabu, 2016). Secondly, individuals with positive behavioral tendencies regularly offer to create opportunities, improve the situation through creative work ideas or suggestions, and engage in development activities of the company (can be more or less involved) due to their achievement needs (Thompson, 2005) and their incentives to learn (Major et al., 2006). Moreover, value congruence increases the motivation for feeling more responsible at work. It is depicted in studies that those with a more flexible duty direction have more probability to conduct proactive work behaviors (Parker et al., 2006) and feel responsible for the change. Helpful change (Fuller, Marler, & Hester, 2006;Morrison & Phelps, 1999) is the premise of proactive work behavior. Therefore, the following hypotheses are proposed: Hypothesis 2: POV is mediated in the positive connection between proactive personality and problem prevention behavior.

The Mediating Role of PSV
Although PB is very important to the organization, it is not frequently valued by supervisors (Grant, Parker, & Collins, 2009). Research by Benson, Hardy, and Eys (2016) shows that although PB has many positive meanings, supervisors do not accumulate their employees to show PB since they think employees should be cautious before making changes and sometimes supervisors are afraid of employees being too proactive might reveal their incompetence or weakness (Fay & Frese, 2001;Grant & Ashford, 2008). Therefore, a deeper understanding of specific PB which is PrP is to assess whether the staff realize that their supervisors consider their PB to be useful or not. Magee and Galinsky (2008) think that person-supervisor relationships have a strong impact on employee behavior because employees are influenced and supervised. In particular, employee's PB depends on the degree of association between employees and supervisors, because the trend of supervisor is to arrange and facilitate employees to promote their strengths. This viewpoint shows that the staff and supervisor respond favorably to each other when they have common traits thanks to the great similarities that enable supervisor to appreciate and understand each other's behavioral trends. And vice versa, this dyadic relationship will make employees feel insecure and obstruct PB. Sun and van Emmerik (2015) suggest that some sensible employees will realize that their PB may not be appropriate for their supervisor. The person-supervisor fit (Kristof-Brown et al., 2005) states that subordinates with similar characteristics to their supervisors may be more satisfied, more committed and more productive (Zhang, Wang, & Shi, 2012;Shin, Kim, Choi, Kim, & Oh, 2017) and when supervisors and subordinates conduct similar actions, they tend to have a more hopeful relationship (Matta, Scott, Koopman, & Conlon, 2014;Qin, Huang, Hu, Schminke, & Ju, 2018). In contrast to previous studies that center employees with PP, this study proposes that PSV is a crucial part in making the staff feel pleasant when showing PB. From the perspective of P-S fit, the research proposes that PSV increases the likelihood of showing PrP of PP.
Hypothesis 3. PSV is mediated in the positive connection between proactive personality and problem prevention behavior.

The Moderating Role of POS
Perceived organizational support describes "employees" faith under a circumstance which the organization assesses their dedication and considers their benefits" (Shore & Tetrick, 1991). According to organizational support theory (Eisenberger, Huntington, Hutchison, & Sowa, 1986;Shore & Shore, 1995), when an organization appreciates employees' dedication and considers their benefits, the staff will take pride in being a member of the organization and continuously make efforts to improve work performance, and to perform behaviors that exceed the organization's requirements such as engaging in PBs. Therefore, we propose POS as an important organizational context element that increases POV and PSV. According to Blumberg and Pringle (1982), if employees are only capable and motivated to work without the support of the organization, it is very difficult to bring high work performance. Research on the interaction effect of proactivity also shows that PB depends on opportunities to be proactive (Fuller, Marler, & Hester, 2012). Proactive individuals may act in a more proactive way providing that they can find support from their organization and/or supervisor or vice versa. POS is a premise for a range of attitudes and behaviors (Jeung, Yoon, & Choi, 2017). POS can be a special moderator that affects the relationship between PP and POV, PSV because employees will become more confident about their own ability in the process of incorporating their own identity into the organization. Indeed, in order to identify themselves as members of the group and be able to devote to it, they must feel comfortable with the organizational values they are representing. While proactive individuals often seek compatibility with the organization, when the POS is low, they will stress at work, feel insecure and defensive, which will hinder the psychology of comfort in incorporating employee and organizational values (Shanock & Eisenberger, 2006). Therefore, POS is necessary to develop POV as well as PSV. This forms the basis for the idea of the research that POS is the guarantee and endorsement that organization and monitoring will support the integration and development of employees. This continues to lead to employees feeling higher PSV. Similarly, (Rineer, Truxillo, Bodner, Hammer, & Kraner, 2017) and Lartey, Amponsah-Tawiah, & Osafo (2019) revealed the moderator role of POS in the connection between employees and the work context. Therefore, the study proposes the hypothesis: Hypothesis 4a: POS will favorably moderate the effects of PP on POV: The relationship is stronger when POS is higher. Hypothesis 4b: POS will favorably moderate the effects of PP on PSV: The relationship is stronger when POS is higher.

An Integrative Moderated Mediation Model
This study develops the theoretical background for the intermediate effect of POV and PSV and the regulating effect of POS. It is POV that mediates the relationship of PP and PrP (H2), PSV mediates the relationship of PP and PrP (H3). POS regulates the favorable connection between PP and POV (Hypothesis 4a), and the favorable connection between PP and POV (Hypothesis 4b). The theory behind the above hypotheses also recommends an integrative moderated mediation model. Particularly, POS can moderate indirect effects of PP through POV and PSV. Theoretical reasons behind the assumptions 2, 3, 4a and 4b show that through the increase or decrease of the connection between PP and POV, POS affects the extent to which PP affects employees' PrP. Similarly, POS, due to their moderating effects on the relationship between PP affecting PrP (Hypothesis 4b), may also have the possibility to change the indirect impact of PP through POV, PSV. Combining these anticipations together, we propose two integrative moderated mediation: Hypothesis 5a: POS indirectly moderates positive effect of PP for PrP through POV: The favorable indirect impact is stronger when the POS is higher (compared to lower). Hypothesis 5b: POS indirectly moderates positive effect of PP for PrP through PSV: The favorable indirect impact is stronger when the POS is higher (compared to lower).
The hypotheses are displayed in Fig. 1.

Participants and procedure
In a complex and demanding hospital environment, mistakes are inevitable. Nowadays, whenever mistakes and incidents occur, people often target individuals. This approach focuses on the individual (doctor, surgeon, nurse, pharmacist, etc.) and thinks that the mistake is due to a deviation in the thought process, such as forgetting, inattention, and lack of motivation, carelessness, negligence, and recklessness. From a new perspective, medical error is a consequence (rather than cause), not entirely from the person but from the organizational system. Therefore, according to this new perspective, our research examines the connection between PP, POV, PSV and the importance of PB as PrP because PrP is considered as an important behavior for the organization in preventing reintegration of errors and barriers to work in the hospital environment. Since then the study proposes administrative implications to help organize staff recruitment and improve the performance of employees and organizations. With the support of leaders and human resources departments of hospitals in Vietnam, the study was conducted through a questionnaire survey of nurses. The questionnaire was sent directly to the volunteers who joined the survey. The questionnaire sets out the research purposes, voluntary feature of the survey and ensures the confidentiality for the participants and the consent of survey participants. Our data is collected from December 2019 to February 2020.

Measure of constructs
The scale of the research is established from the theory of PP (Bateman & Crant, 1993), POV and PSV (Nguyen, 2013), and PrP ( Parker & Collins, 2008). In order to ensure the effectiveness when translating into Vietnamese, the study has conducted double translation and double-blind way. We then discussed and edited the translation with human resources specialists and medical staff to suit the situation in Vietnam. The observed variables were surveyed on a Likert 7 scale ranging from 1 (strongly disagree) to 7 (strongly agree).
PP: The measure of PP concept is measured by 16 observed variables (Bateman & Crant, 1993). The reliability of this scale was .947. POV: The scale of person-organization value congruence uses the scale of Nguyen (2013) which is measured by 06 observed variables. The reliability of this scale was .883. PSV: The scale of person-supervisor value congruence uses the scale of Nguyen (2013) which is measured by 06 observed variables. The reliability of this scale was .863. PrP: The scale of problem prevention behavior is measured by 03 observed variables (Parker & Collins, 2008). The reliability of this scale was .860. POS: The scale of perceived-organizational support uses the scale of (Eisenberger et al., 1986). The reliability of this scale was .959.

Descriptive Statistics and Correlations
With 715 usable responses from 800 questionnaires (rate 98.38%). Demographic characteristics are listed in Table 1, showing that the majority of nurses participating in the survey are female (71.75%), the main ages are from 30 to 40 years (42.10%) and their working experience is mainly from 5-10 years. The majority of nurses have intermediate and college degrees, only few of them have undergraduate and graduate degrees. In general, the income of nurses is quite low.  Table 2 shows the Cronbach's α, mean, standard deviation of variables in the model (means, standard deviations, of the latent constructs in the measurement model). All correlations are less than .5 indicating no multi-collinearity. Pearson's test in Table  2 shows that most independent variables correlate with dependent variables at the 99% significance level, the correlation coefficients are > .3. The results of estimating the relationship between the research concepts show that all factors have an impact relationship with each other at the statistical significance level of 5%.  Table 3 shows the results of the confirmatory factor analyses (CFA). It can be seen that the existing five-factor model is more relevant to the data than all substitute models, showing support for the differentiation of the structures in this research. Moreover, the CFA results of the four-factor model show that all indicators show suitability of the model fits. This result reveals that the measurement model suits well with the data. Table 3 indicates that the standard weights of the scale are all high, higher than .5 and the p value of each pair is lower than .05. Moreover, the correlation coefficients with standard deviations depict that they are different from 1. Therefore, the modified research model attains convergent and discriminant values. Since the model has no correlation between measurement errors of factor groups, it becomes monad. Note: x 2 = Chi square, df = degree of freedom ratio, NNFI = Non Normed Fit Index, CFI = Comparative Fit Index, RMSEA= Root Mean Square Error of Approximation.

Mediation and Moderation Effects
With data collected from the individual level, the study utilized OLS regression to analyze and check hypothesis 1. Here, we used sample-based bootstrapping to conduct product of coefficient test so that we could check the two set of indirect effects proposed by H2 and H3, Results Table 4 shows that PP has a positive impact on PrP (β = .487, p = .000, Model 2), so that H1 is accepted. Regarding the test of indirect impacts, we conducted product of coefficient test using the PROCESS, we used 5,000 bootstrapping samples and then reported bias-corrected confidence intervals for each indirect impact. The results in Table 5 show that the indirect impacts of PP on PrP through POV and PSV are significant. Thus, hypotheses H2 and H3 are accepted. Table 4 shows the results of testing the impact of POS moderator variables in the connection between PP and POV (β = .299, p = .000, Model 4) and PP and PSV (β = .237, p = .000, Model 6). In addition, we arranged the interaction using the plotting procedures. Fig. 2 and 3 indicate this interaction graphically at two (i.e., high versus low) levels of POS (i.e., +1 and −1 standard deviation). As forecasted, the connection between PP and PrP was more favorable when POS was higher. The relationship between PP and POV is stronger for people with high POS (simple slope = .881) than those with low POS (simple slope = .342) (Figure 2). Similarly, the relationship between PP and PSV is stronger for people with high POS (simple slope = .752) than those with low POS (simple slope = .326). Therefore, hypotheses H4a and H4b are supported (Figure 3).

Moderated Mediation Test
This study uses macro PROCESS for SPSS to check hypotheses 5a and 5b on the role of moderates of POS in indirect effects of PP on PrP through POV, PSV. The results in Table 6 reveal that the POV was a remarkable mediator between PP and PrP regardless of the level of POS, as zero was not included in the confidence intervals (Table 5) (lower limit confidence interval and upper limit confidence interval). Similar, PSV also had a significant mediator between PP and PrP regardless of the level of POS at all three standard deviation levels: (1) less than mean, 95% CI [.099; .257]; (2) equal to the mean, 95% CI [.100; .247]; (3) above the mean, 95% CI [.098; .248]). This shows that the moderating impacts of POS on the relationship between PP, POV, PSV, PrP are significant as reported in Table 6. .248 Note: Bootstrap sample size: 5,000; LLCI: lower limit confidence interval; ULCI: upper limit confidence interval.
The results of the causality test among the research factors indicate that all the connections in the research model are statistically significant or the hypotheses are accepted through standard regression weights. The correlation between positive factors proves that the relationships among these factors have a positive influence on each other. In summary, the results of testing the theoretical model depict that the model is suitable for the survey data set. The standardized results of the causal relationship between the concepts in the theoretical model show that the relationship between the concepts in the theoretical model is statistically significant (p <.05), which gives see the scales of the concepts in the research model reach theoretical relational values.

Discussion
Moderated mediation model test shows the intermediate impacts of POV, PSV on the connection between PP and PrP use were regulated by POS. In particular, the intermediate impacts of POV, PSV are stronger for employees with higher POS than employees with lower POS. The study also shows that POS moderated the links between PP and POV, PSV. In particular, employees with high POS have higher POV, PSV than those with low POS. This reveals that when employees feel that they are working harder than others but being given lower rewards or less recognized, they have higher chances to be unhappy and show less proactive behaviors. This finding is compatible with research of (Özarallı, 2015) that employees being supported with knowledge, resources, opportunities, and meaningful job experience will lead to their high internal motivation and better understanding about the role and value of the organization. In summary, based on the research results, we propose some important management implications as follows. Firstly, research results show that PP, POV, PSV have a positive influence on PrP, so for employees to show more PrP, organizations need to attract and recruit employees with PP as good seeds for the sustainable development of the organization. Secondly, we find that PP is effective in promoting employees' PrP through provoking POV, PSV. And for employees to show more PrP, leaders should set out policies and procedures for employees to have higher POV, PSV. Finally, the research results also show that when PP employees have high POS, they have higher POV and PSV than PP those with low POS. Therefore, with organizations providing a good work environment and a supervisor respecting and caring about employees' needs, proactive individuals will complete their tasks in the best way and promote all of their abilities. In addition, in Vietnam, the collective working environment is very high (Hofstede, 2015), so apart from employees doing their job well, the connection between supervisors and employees is also a deciding factor.

Conclusions
This study answers the call of McCormick, Guay, Colbert, & Stewart (2019) regarding understanding the factors that make employees more proactive by exploring the moderating factors of the PP. The results of testing the mediating role of POV, VOS in PP and PrP relationship are significant and consistent with the research of Vveinhardt & Gulbovaitė (2016;. Its findings are also compatible with the opinion of Cangiano, Parker, & Yeo (2019) recognizing that supervisor's reactions play a vital role in deciding whether proactive people are the asset of the organization. PP has been found to have a significant impact on proactive behavior and organizational results. Therefore, this study complements previous research as it explores beyond the effects of workplace conditions to explain how PP leads to PB. First, our study incorporated PP into the study framework for the first time, confirming that PP is a driving factor in PrP. Second, by examining the effects of VOC and VOS in line with the role of the mediator of the relationship between PP and PrP, current research contributes to appropriate literature. Specifically, we have established VOC and VOS as potential intermediaries through which PP transmits its effects to PrP. In particular, we contribute to the theory of value congruence by proving the connection of PP with not only impacting on POV but also PSV. Most previous experimental studies have looked at the effect of POV Pierro, Sheveland, Livi, & Kruglanski (2015), very few studies have regarded both POV and PSV. Therefore, we have integrated them into a coherent model, in which POV and PSV act as a mission-driven mechanism and perceive the support of the organization as a mechanism to navigate this relationship. Finally, the findings show that employees with more PB will show better congruence and suitability with the organization and supervisors compared to others. This adjustment can be done by helping colleagues see problems and ways of working together, and when preventing problems, they can help find problems and adjust behavior to be more appropriate. Besides, POV also has an intermediary impact but this level of expression is quite low. It also suits with theory and empirical studies when staff feel they are responsible for bringing about positive change (Fuller et al., 2006).

Limitations and Future Research
Apart from the aforementioned theoretical and realistic contributions, this research still has the following shortcomings. Firstly, because the data was collected from only one source who were staff at hospitals in Vietnam, this could cause potential general variance in methods. To reinforce the generality of our findings, replication studies ought to be conducted under various circumstances. Besides, with a great sample size, detecting important relationships is quite facile, so our experimental discovery requires more replicated research to bring more theoretical significance. Second, although we strictly control the testing process, all data comes from employee self-reporting, indicating that there may be general method deviations. In the future, PB should be evaluated by other reporters such as leaders and colleagues. Finally, our results are limited to a PB (PrP). Further studies may inspect a wider range of work behavior, particularly performance-related consequences like in-role and out-role performance.