The role of social self-efficacy and psychological capital in IT graduates’ employability and subjective career success

Abstract This study aims to assess the impact of Social Self-Efficacy and Psychological Capital on Employability and Subjective Career Success of IT graduates in Vietnam. An empirical research using a questionnaire was carried out from April to August 2022 in big cities in the North and Center of Vietnam. The sample includes 336 university graduates (within 6 years after graduation) who were undertaking IT-related jobs. Research results show that both Social Self-Efficacy and Psychological Capital have a positive impact on Employability and Subjective Career Success (which includes Job Satisfaction and Career Satisfaction). The study’s findings are significant for both educational institutions that offer IT training and human resource management activities for IT staff in organizations and enterprises in Vietnam.


PUBLIC INTEREST STATEMENT
The labor market in Vietnam is changing quickly and becoming more competitive, especially in the field of IT. The majority of Vietnamese IT enterprises are small and medium sized and are expanding rapidly. In the context of Vietnam, a country characterized by the preservation of community culture, social skills play an important role in maintaining a sustainable career for workers. This study aims to assess the impact of Social Self-Efficacy and Psychological Capital on Employability and Subjective Career Success of IT graduates in Vietnam. Research results show that both Social Self-Efficacy and Psychological Capital have a positive impact on Employability and Subjective Career Success (which includes Job Satisfaction and Career Satisfaction). The study's findings are significant for both educational institutions that offer IT training and human resource management activities for IT staff in organizations and enterprises in Vietnam.

Introduction
As the labor market in Vietnam is changing quickly and becoming more competitive, the topic of graduates' Employability (EMP) has become more significant. For IT human resources, this issue is even more essential due to the specific characteristics of Vietnam's IT industry. Firstly, the majority of Vietnamese IT enterprises are small and medium sized; they perform outsourced work from large international technology companies and corporations and are dependent on these corporations. It follows that the career stability of employees in these companies is low. Secondly, the IT industry has recently experienced rapid expansion and this market has a higher rate of job switching than other industries. The aforementioned characteristics require employees in this industry to have better job-seeking skills and higher adaptability. In addition, it is important to pay attention to this issue since IT is considered as one of the national strategic areas of development in Vietnam in the current and upcoming period.
On the other hand, the objective of modern human resource training is not just to assist students in finding employment after graduation but also to work towards a long-term objective whereby students need to be given the knowledge, abilities, and psychological traits to be able to do the job as demanded by the labor market and to achieve job satisfaction and career success. On a macro level, such a quality workforce will bring benefits not only to the employees and the employer, but also the community and society (Yorke, 2004).
In the context of Vietnam, a country characterized by the preservation of community culture (Phuong-Mai et al., 2005), social skills play an important role in maintaining a sustainable career for workers. Particularly, within the IT industry, this issue is even more important as the characteristics of this industry require both independence and collectiveness (project teams, working with customers, partners…) (Shih et al., 2011). This calls for a closer investigation of the role of Social Self-Efficacy (SSE) in EMP and Subjective Career Success (SCS) of IT graduates. In addition, another important characteristic for IT human resources is an achievement orientation (Wynekoop & Walz, 1998), thus, workers need to be equipped with relevant psychological qualities. One important concept related to the psychological qualities of human resources is psychological capital (PsyCap)  and this study also aims to look into this concept more closely.

The EMP concept
EMP is one of the important concepts attracting the attention of researchers from different fields, especially in training and human resource management, vocational training and higher education. Past studies offer several different definitions of this concept. Hillage & Pollard (1998) defined EMP as "the capability to gain initial employment, maintain employment and obtain new employment if required" (p. 1), the capacity of "getting and keeping fulfilling work" (p. 2), or "the capability to move self-sufficiently within the labor market to realize potential through sustainable employment"(p. 12). Similarly Nabi (2003) defined EMP as the skills, attitudes of graduates and their ability to mobilize these resources in finding and retaining jobs. Yorke (2004) defined EMP as "a set of achievements-skills, understandings and personal attributes-that make graduates more likely to gain employment and be successful in their chosen occupations, which benefits themselves, the workforce, the community and the economy" (p. 410). According to Yorke (2004) EMP is not only about getting a job, keeping a job and finding a new job when needed, but also including success at work and benefiting the community and society. According to Mason et al. (2006) EMP is not only about getting a job, it includes skills, work motivation and problem-solving ability to help businesses operate effectively and achieve goals. For Coetzee (2012) career-related EMP attributes help promote adaptive cognition, behavior and affect and enhance a graduate's suitability for sustained employment. According to the author, EMP is different from graduateness, it includes "the inherent characteristics (transferable meta-skills and personal attributes) of graduates … that differentiate them as responsible, accountable, relevant, ethical and enterprising citizens, and employees of choice in the workplace"(p. 121). For employees, Rothwell and Arnold (2007) described the concept as "the ability to keep the job one has or to get the job one desires" (p. 25). Empirical research results also showed that there are two aspects of EMP, namely internal EMP (self-value and the perceived value of occupation in the actual organization), and external employability (self-value and the perceived value of occupation in the actual market, i.e. outside of the organization).
An important point that should be mentioned in EMP studies concerns the approach and measurement of EMP. Previous studies mentioned various perspectives to approach and measure EMP, including perspectives of universities (Camps & Torres, 2011), graduates (Rahmat et al., 2012;Yusof et al., 2012), employees (Rothwell & Arnold, 2007), and employers .

Relationship between EMP and SCS
Career Success is defined as ' the positive psychological or work-related outcomes or achievements one accumulates as a result of work experiences" (Seibert et al., 1999, p. 417). There are two approaches to study Career Success: Objective Career Success (OCS) and Subjective Career Success (SCS). Objective Career Success is defined as directly observable by others and measurable with criteria such as salary, job level, promotion history, or occupational prestige. SCS, on the other hand, refers to the individuals judgments of achieving personally meaningful career outcomes which are measurable subjectively with criteria such as career satisfaction, growth and development, personal life, and authenticity (Spurk, Hirschi & Dries, 2019). Regarding SCS, researchers often mention two important components including job satisfaction (JOS) and career satisfaction (CAS) (Heslin, 2005;Ng et al., 2005). JOS refers to the extent to which an employee expresses a positive emotional state towards his/her job or job experiences (P. C. Smith et al., 1969). CAS, on the other hand, refers to the satisfaction coming from subjective perception of his/her career success, based on his/her career-related expectations and goals (Judge et al., 1995;Loi & Ngo, 2010). These two components are independent, but also closely related. An individual's lack of satisfaction with his or her current job is often accompanied by a negative perception of career success (Judge et al., 1999).
Regarding the relationship between EMP and SCS, studies show that EMP has positive effects on JOS and CAS (Bozionelos et al., 2016;Niu et al., 2019;Rothwell & Arnold, 2007). Through a sample of 207 IT professionals working in SMEs in three European countries (Greece, Italy, and Poland) Bozionelos et al. (2016) found a positive relationship between EMP and SCS. Rothwell & Arnold (2007) also studied 200 employees working in human resource management in the UK, finding a significant correlation between the two variables. In a study by Niu et al. (2019) on 85 graduates of the workforce education and development (WED) program at a university in Midwestern United States, similar results were found between EMP and SCS. Based on the above results, within the framework of this study, we hypothesized that EMP has a positive effect on two components of SCS, namely JOS and CAS. Specifically: H1a: EMP has a positive effect on JOS H1b: EMP has a positive effect on CAS

Relationship between two components of SCS-JOS and CAS
Although JOS and CAS are the two main components of the SCS concept, some studies believe that because these are two independent components both in terms of concept and scale, the study of the relationship between the two above components is also essential. A study by Bostjancic and Petrovcic (2019) conducted on 1168 individuals working at a university in Slovenia showed that JOS is an important factor in predicting CAS besides other factors such as managerial position, engagement, satisfaction with co-workers, satisfaction with opportunities for promotion and education, and satisfaction with salary. In another study by Murawski et al. (2008) on 108 pharmacists in the US, the results also showed that CAS had a positive impact on the perception of CAS, thereby concluding that JOS is an important factor in the formation of CAS. Similarly Brady (1980) studied 170 graduates from two physician's assistant programs and found a positive effect. Based on the above research results, this study put forward the following hypothesis: H1c: JOS has a positive effect on CAS

The concept of SSE and its relationship with EMP and SCS
SSE is defined as an individual's "beliefs about their level of ability to successfully perform a given task or behavior in social relationships" (Wright et al., 2013, p. 219). For H. M. Smith and Betz (2000) SSE refers to "the individual's confidence in her/his ability to engage in the social interactional tasks necessary to initiate and maintain interpersonal relationships"(p. 286). In the context of education, researchers found that SSE has a positive relationship with college satisfaction, life satisfaction, self-esteem, and a negative one with loneliness (DeWitz & Walsh, 2002;Hermann & Betz, 2006;Wright & Perrone, 2010). In the context of the workforce, SSE was found to have a negative impact on career indecision (Betz et al., 1999).
With regards to the relationship between SSE and EMP, although many authors are interested in understanding this, most research deals with General Self-Efficacy (GSE), and there has been no research on SSE as an independent component of GSE (Sherer et al., 1982). In a study conducted on 267 undergraduate students at a number of public and private schools in Malaysia, Chow et al. (2019) found that GSE has a strong impact on EMP. Another research by Tentama and Nur (2021) conducted on 132 vocational high school students in Indonesia also showed similar results. Wujema et al. (2022) did research on 264 students at some universities in Nigeria, and found that GSE also had a strong impact on EMP. In this study, GSE acted as an intermediary variable between the 5 components of the CareerEDGE model (Pool & Sewell, 2007) including career development learning; work and life experience; degree subject knowledge skills, and understanding; generic skills; and emotional intelligence to EMP.
Regarding SSE, although no research has mentioned its relationship with EMP, a few studies have been done to evaluate the relationship between a number of variables close to SSE, Social Capital (SOC), and EMP. A typical example is a study by Ngoma & Dithan Ntale (2016). Through a survey conducted on 215 unemployed young people studying in a skills training program in Uganda, Ngoma & Dithan Ntale found that SOC has a strong impact on EMP. In the framework of this research, SOC acts as an intermediate variable in the relationship between PsyCap and Career identity with EMP.
Since SSE is an aspect of GSE (Anderson & Betz, 2001), we hypothesized, on the basis of previous research results, that there existed a positive relationship between SSE and EMP: H2a: SSE has a positive effect on EMP Regarding the relationship between SSE and SCS, Luo et al. (2019) showed that there exists a significant relationship between these two variables. Lin (1999) investigated the impact of SOC-a concept close to SSE-on SCS, and found that SOC has a positive correlation with SCS. It could be explained that when an individual had wide connections and effective relationships, they would have an advantage in accessing information about potential jobs.
Regarding the relationship between SSE and JOS, Luo et al. (2019) conducted a study on 172 employees in China and found that SSE has a positive effect on JOS, and this relationship went through an intermediate variable-Social Status (defined by Anderson & Betz (2001) as "an individual's informal social position in a group that is represented by prominence, respect, and influence in the eyes of other group members").
Regarding the CAS component of SCS, a study by Hochwarter et al. (2004) showed a negative relationship between SSE and CAS. This study was conducted on 222 MBA graduates from Southeastern University. The results showed that individuals with low SSE have high levels of satisfaction with their career, which meant they had a higher level of perceived CAS compared with the other groups.
In summary, results about the relationship between SSE and the components of SCS vary among previous studies. In the case of Vietnam, based on the nature of the SSE variable and the nature of the IT profession, individuals often have to work on their projects, interact with customers, partners and colleagues, especially in the collective culture here. Within the framework of this study, we proposed the following hypotheses about the relationship between SSE and SCS: H2b: SSE has a positive effect on JOS H2c: SSE has a positive effect on CAS  defined PsyCap as "an individual's positive psychological state of development and is characterized by: (1) having confidence (self-efficacy) to take on and put in the necessary effort to succeed at challenging tasks; (2) making a positive attribution (optimism) about succeeding now and in the future; (3) persevering toward goals and, when necessary, redirecting paths to goals (hope) in order to succeed; and (4) when beset by problems and adversity, sustaining and bouncing back and even beyond (resilience) to attain success" (p. 542). With this definition, there were 4 factors forming the PsyCap, which are: self-efficacy, optimism, hope and resilience.

The concept of PsyCap and its relationship with EMP and SCS
PsyCap was found to have positive relationship with various work and career-related constructs such as job satisfaction, organizational commitment, job performance, organizational citizenship behaviors, absenteeism, career development skills and outcomes, expected career performance and satisfaction (Conklin et al., 2013;Daswati et al., 2022;Luthans, Youssef, et al., 2007;. The relationship between PsyCap and EMP has also been of interest to some previous studies. Ngoma & Dithan Ntale (2016) found a positive correlation between PsyCap and EMP through the intermediary variable Social Capital. In a study by Ayala Calvo & Manzano García (2021) with a sample of 326 final year undergraduate students majoring in Business Administration at three public schools in Northern Spain, the research results showed that PsyCap had both direct and indirect effects (through employability skills) on EMP.
The relationship between PsyCap and the JOS component of SCS has also been studied in several previous studies. Huynh & Hua (2020) conducted a study on 800 employees in small and medium enterprises in three large cities of Vietnam (Hanoi, Da Nang and Ho Chi Minh City) and found that PsyCap positively affects JOS, and both variables had a positive impact on the organizational commitment of employees. Another study by Zhang et al. (2021) on 310 Chinese doctors in hospitals showed that there exists a positive relationship between PsyCap and JOS, and this relationship is both direct and indirect (through Organizational Identification). These relationships were also moderated by income levels. Ganji & Johnson (2020)

Measurement of EMP
EMP has been an important concept in the research on higher education, in human resources, in management and in other fields. There are different approaches to measure this concept in the literature: Competence Based Measurement of EMP by Van Der Heijden & Van Der Heijden (2006); Dispositional Measure of EMP by Fugate et al. (2004) and Self-Perceived EMP Scale by Rothwell & Arnold (2007). In this research, we adopted the conceptual framework by Berntson (2008) according to that "Perceived employability is an important construct because feeling employable potentially provides the individual with a feeling of security and a feeling of independence towards environmental circumstances" (p. 19). In line with this, Rothwell and Arnold (2007) developed a scale with 16 items, considering that EMP refers to "an individuals' discernment of their skills and abilities and how they perceived an organization would react to them as individuals with varying characteristics and attributes" (p. 12). These 16 items were distinct and concerned individuals' perception of employability inside and outside of their organization, their confidence in their personal skills, knowledge and work experiences. In Rothwell & Arnold (2007)'s study, a revised version of the scale (11 out of 16 items) was used and the reported results showed a good internal consistency of the scale (Cronbach's alpha was 0.83).

Measurement of SCS
There are two main perspectives in assessing individuals' career success: objective and subjective. The former perspective is often measured with objective and observable indicators, such as salary, work position, or attained promotions (Dries, 2019;Frederiksen & Kato, 2018). The latter perspective, on the other hand, is assessed by the individual themselves (Gattiker & Larwood, 1986). This judgment is based on their subjective perception of career achievements up to the time of survey and prospects of future attainments, both within and outside their organization.
By focusing on the subjective perception of individuals about their career success, i.e. SCS, we adopted the Career satisfaction (CAS) scale developed by Greenhaus et al. (1990) with five items. Based on a sample of 828 managers participating in their study, Greenhaus et al. (1990)'s study indicated that the 5-item scale measuring CAS had satisfactory levels of internal consistency reliability (Alpha Cronbach = 0.88). To measure JOS, a six-item scale derived from research of Curry et al. (1986) was adopted. In this study, with a sample of 508 participants, the results showed that the scale had a good internal consistency reliability (Alpha Cronbach = 0.86).

Measurement of SSE
The measurement scale developed by Wright et al. (2013) was used for assessing graduates' SSE. This scale was composed of two components, namely "Social Efficacy Expectations" (12 items) and "Social Outcome Expectations" (6 items). However, in this research, we focused mainly on individual's confidence in their abilities to successfully complete relational tasks (initiating, maintaining and developing social relationships and affairs), so only 12 items for "Social Efficacy Expectations" were used. In their study, with a sample of 486 participants from a university in the Rocky Mountain region, North America, the results indicated a satisfactory level of internal consistency reliability for this subscale (Alpha Cronbach = 0.96) and good CFA model fit indices of the scale (N = 355; CFA fit index: χ2 = 448.454, df = 134, RMSEA =.059, SRMR =.043, CFI =.698, NNFI =.963; all factor loadings are 0.79 and above for SSE subscale).

Measurement of PsyCap
In this study, we adopted the scale developed by Luthans, Youssef, et al. (2007) to measure PsyCap. The original scale was composed of 24 items measuring four components including selfefficacy, optimism, hope and resilience (each component is measured by 6 items). These components are first-order constructs of the PsyCap, which is measured as a second-order construct . The reliability and validity of the scale was examined by  and Imran and Shahnawaz (2020) and both research indicated an adequate reliability and validity of the scale. , with a CFA analysis on a sample of 745 management students from different universities in the United States, showed that the second-order structure of the scale (with 24 items) presented an adequate validity (model fit: SRMR =.051, RMSEA =.046, CFI =.934). The scale's reliability was also assessed using Cronbach alphas and it was shown that the Cronbach alpha for each of the four first-order factors and for the overall PsyCap was 0.66 and above. For Imran and Shahnawaz (2020), based on the results of a CFA analysis on the sample of 225 employees in private sector from various industries in Indian setting, they also indicated that both the four-factor structure and the second-order structure of the scale (with 22/24 items retained) presented acceptable model fit indices (χ2 = 337.451, df = 183; CFI = 0.938; TLI = 0.928 and SRMR = 0.048 for the four-factor structure, and χ2 = 355.533, df = 185; CFI = 0.931; TLI = 0.922 and SRMR = 0.064 for the second-order structure). The research also revealed a good level of internal consistency reliability of the scale (0.94). For the present research, to measure three components of PsyCap that were retained (Hope, Optimism, Resiliency), 18 items were adopted.
The final questionnaire consisted of 67 items measuring five variables EMP, JOS, CAS, SSE and PsyCap. The items were rated on a 5-point scale from 1 (Completely false/disagree) to 5 (Completely true/agree). In addition, the questionnaire also contained questions related to personal information such as: Gender (Male/Female); Graduate school; Graduate field/domain (IT related/Non-IT related); Number of years after graduation (undergraduate/from 1 to less than 3 years/from 3 to 6 years/over 6 years); Current job (IT-related/Non-IT related). In order to minimize the number of invalid/incomplete responses, in the introduction of the questionnaire, emphasis was placed on the target group of fresh graduates within 6 years in IT or related fields who were working at organizations, enterprises and undertaking IT or related jobs. The participants were also briefed about the research objectives and 5 levels of the rating scale and assessment methods; and they were asked to answer objectively and truthfully. Survey participants were not paid, awarded gifts or had other benefits. Items were translated into Vietnamese using the translate/ back-translation technique (Prince & Werner, 1967).

Sampling, data collection and data cleansing
Numerous universities that offer IT training may be found in northern Vietnam, particularly in Hanoi, as well as in the central region. Many graduates choose to work in these locations after graduation. Therefore, the survey was conducted in a number of major cities in these areas including Hanoi, Thai Nguyen, Phu Tho, Vinh, and Da Nang.
The convenience sampling method was adopted. 500 alumni who graduated from IT programs of different universities in five major cities were chosen to send the questionnaire (with help from programs' coordinators of chosen universities). To ensure that respondents have from 1 to 6 years of work experience after their graduation, only students who graduated from 2015 to 2021 were chosen. The number of responses received was 392, of which 56 were invalid, which accounted for 14.3%. Invalid responses belonged to those who graduated more than 6 years or not yet graduated; did not attend university in IT fields; or were actually not doing IT-related jobs. These invalid data were filtered out, resulting in 336 valid responses (accounting for 85.7%) that were used for data analysis.
The data collection method for this study is a questionnaire designed with Google Forms and sent via email and other digital channels to research subjects. The data was collected during the period from April to August 2022. The use of Google Forms questionnaire (with a unique value choice for each question and participants were required to answer all the questions in the questionnaire before submitting it) helped prevent missing responses, overlapping selections, or outliers (that fall outside the scale from 1 to 5). Participants with abnormal answers (same value with no distinction for all questions; most values were high, low, or mid) were discarded. The values of items having opposite meaning to their constructs were inversed before being used for analysis.

Data analysis methods
With regards to the assessment of the reliability and validity of the scales, a Confirmatory Factor Analysis (CFA) was performed for all variables. Based on the CFA results, the CR and AVE indicators of each scale and the values of inter-variables correlations were calculated. These statistics were used to assess the reliability, convergent and discriminant validity of measurement scales (Fornell & Larcker, 1981;Hair et al., 2009). According to Hair et al. (2009) the reliability of a construct is assured if CR value is of 0.7 and above, and its convergent validity is granted if AVE value and all standardized factor loadings are 0.5 and above. For the discriminant validity of a contract, it was assumed that the root square of AVE value is higher than the correlation coefficient of the construct with other ones (Fornell & Larcker, 1981).
In order to test the relationships among variables in the research model, Structural Equation Modeling (SEM) was adopted. This method is suitable for quantitative data collected in this study (5-level rating scale) and is considered optimal in testing research models including independent variables, mediating variables, and dependent variables. These techniques were performed with AMOS software, version 18.

Description of sample
Validated data was composed of 336 IT fresh graduates (239 men − 71,1% and 97 women − 28,9%) from universities in the North and Central region of Vietnam. There were 209 graduates with 1 to 3 years of work experience (62,2%) and 127 individuals with 3 to 6 years of work experience (37,8%). The participants graduated in IT and related majors. They had a university degree or higher and worked in several industries in major cities in Vietnam.
The distribution of the data was tested using Kurtosis and Skewness coefficients. The coefficients' values for all items, ranging from −1.96 to +1.96, indicated a normal distribution of the data.

Reliability and validity of measurement scales
In order to assess the reliability and validity of used scales, a CFA analysis with all the variables in the model was performed. Based on the results of this analysis, the CR and AVE indicators were calculated. Table 1 below reports the main indicators for the assessment of reliability and validity of the scales: For the PsyCap measure, a second-order CFA was run. Three components (i.e. three first-order factors) were loaded to the second-order factor with high coefficients (0.93, 0.98 and 0.96 for Optimism, Hope and Resilience respectively, with p = 0.00). The 14 retained items were loaded to their respective first-order factor with significant coefficients ranging from 0.71 to 0.86. Main model fit indicators are: CMIN = 260.505, P = 0.00, DF = 74, CMIN/DF = 3.52, CFI = 0.947, RMR = 0.031. These statistics indicate that the quality of the PsyCap measurement scale is adequate as a second-order model. Table 1 shows that the CR values of all variables are greater than 0.7, showing that the variables have satisfactory reliability. The AVE values of the variables are all greater than 0.5 and the standardized factor loadings of the respective items of each variable are above the 0.5 threshold, indicating that the variables have satisfactory convergent validity. Finally, the square root of AVE values of each variable (the values in bold shown in the diagonal) are greater than the intercorrelation of that variable with the other variables (the remaining values in respective rows and columns), showing that discriminant validity of the research variables is guaranteed.

Relationship model and hypothesis testing
In order to test the relationship between variables, a SEM analysis was performed. The statistics include: CMIN = 2907.8, DF = 1305, p = 0.00, CMIN/DF = 2.2, CFI = 0.906, RMR = 0.03. With regards to these statistics, it can be seen that the relationship model indicates a sufficient level of fit between the data and the model.
The relationship coefficients between variables are shown in Figure 1 below: Based on this relationship model, the Table 2 below provides the results of hypothesis testing:

Findings and discussions
The results showed that EMP has a strong and positive impact on both JOS and CAS-the two components of SCS. In other words, individuals with high confidence in their employability expressed a high level of satisfaction with their job and their career, thus a high SCS. These findings supported the result from past research (Bozionelos et al., 2016;Niu et al., 2019;Rothwell & Arnold, 2007). As regards the studies of Rothwell and Arnold (2007) and Niu et al. (2019), although the SCS variable scale used is a one-component scale, which was different from the SCS scale adopted in the current study (with two components JOS and CAS), it showed that the similarity between the results of the two previous studies with that of the present one is high, although the research subjects and social contexts were different. In other words, with this result, it can be seen that the relationship between EMP and SCS is stable in different cultural contexts and in different professions, including the IT sector.
Concerning the relationship between SSE and EMP, this research revealed a positive and direct impact of SSE on EMP. It could be seen that SSE was an important factor affecting IT fresh graduates' perception of EMP. In other words, individuals with a positive perception of SSE also  had a positive perception of EMP. Given that past studies had not been interested in studying the relationship between SSE and EMP, this result bridged this gap in literature. It is a meaningful finding because SSE is an important factor forming an individual's EMP, specifically for fresh IT graduates in the context of Vietnam-a collectivist culture (Phuong-Mai et al., 2005) with a high emphasis on social image (Pham & Richards, 2015).
With regards to the impact of SSE on CAS, the results showed that CAS was significantly influenced by SSE. This meant that the higher the level of perceived SSE, the higher the level of career satisfaction IT graduates could feel. This finding supported the results of the research of Luo et al. (2019). Concerning the connection between SSE and JOS, the present research revealed no significant relationship, showing that fresh IT graduates' JOS was independent of their perception of SSE. Combined with past studies that have given different results in the relationship between SSE and CAS (for example, Hochwarter et al. (2004) found a negative relationship while Luo et al. (2019) found a positive relationship between these two variables), it can be seen that the relationship between SSE and JOS is unstable in different contexts.   The research also revealed strong and positive impacts of PsyCap on EMP. This finding provided evidence that positive psychological attributes (hope, optimism and resilience) contribute to the development of fresh IT graduates' EMP. This result was also supported by previous research (Ayala Calvo & Manzano García, 2021;Ngoma & Dithan Ntale, 2016).
Regarding the relationship between PsyCap and components of SCS, the results showed that PsyCap had a positive effect on JOS, while there was no impact on CAS. The positive impact of PsyCap on JOS meant that graduates with a high perception of PsyCap were more satisfied with their current job than others. This result was consistent with the findings of Huynh & Hua (2020), Zhang et al. (2021) and Ganji & Johnson (2020). The absence of a significant effect between PsyCap and CAS showed that there was no significant difference in the perception of CAS between individuals having different self-assessment of their PsyCap. This result was different from the ones of Baluku et al. (2021) and Al-Ghazali et al. (2021) according to that PsyCap had a positive effect on CAS. The difference could be explained by the fact that in the present research PsyCap only included Hope, Optimism and Resilience components, and did not include Self-Efficacy, while for the other studies PsyCap was composed of all these four components. In other words, the Hope, Optimism, and Resilience components of PsyCap had no direct influence on CAS, and the impact seemed to come from Self-Efficacy.
A strong and positive impact of JOS on CAS was also stated in this research. This meant that the more satisfied fresh IT graduates felt about their job at the time, the higher the level of satisfaction they had for their career. This result was consistent with the findings of previous research such as Bostjancic and Petrovcic (2019) and Murawski et al. (2008). Given that these studies had diverse subjects and different socio-economic settings, the relationship between JOS and CAS was further confirmed in a new social setting and with new research subjects-IT graduates in the present research.
Another conclusion that can be made from the present research is that, when comparing the impact level of SSE with PsyCap in relation to EMP and SCS, we found that PsyCap had a stronger impact (coefficients are 0.66, 0.56 and non-significant for EMP, JOS and CAS respectively) compared to SSE (coefficients are 0.17, non-significant and 0.18 for EMP, JOS and CAS respectively). These figures demonstrated that the components of PsyCap (Hope, Optimism and Resilience) were more important factors than SSE in determining the perception of EMP, JOS and CAS.

Conclusion
In this study, two key findings could be drawn. Firstly, SSE and PsyCap played a significant role in enhancing the EMP of IT graduates. Secondly, the above two factors, especially PsyCap, and EMP also directly affected the two important components of SCS (JOS and CAS), in the work setting of Vietnamese IT graduates.
Based on the results of the present research, a number of solutions are proposed for educational institutions with IT training in Vietnam to help improve the EMP of students, as well as for businesses to help improve the JOS and CAS of employees working in IT positions.
In terms of universities, the development of curriculum and extra-curricular activities can help students gain necessary knowledge to achieve desirable psychological qualities. The factors that constitute PsyCap include Optimism, Hope and Resilience, and confidence in social competence (SSE). These include skills and qualities such as confidence in the ability to create and maintain social relationships, confidence in communication, sharing, expressing opinions, and persuasiveness in social communication. These psychological traits support IT students in developing a high EMP as well as having a fulfilling work and career. As analyzed above, the impact of PsyCap and SSE on EMP, and that of EMP on JOS and CAS have been stated in different studies in fields other than IT, and also in different economic and social settings. This implies that the above solutions are not only suitable for IT training but also for other training disciplines.
In particular, some specific solutions to help students form a high SSE were proposed by the researchers as follows: improving the quality of the relationship and interaction between teachers and students (Carmeli et al., 2021); building charitable activities and programs that encourage students to show their kindness to the community and others (Layous et al., 2012;Peterson & Seligman, 2004); creating an emotional climate and positive class social environments that reduce stress and increase the quantity and quality of student interactions (Hong et al., 2021); sports activities (Koparan et al., 2009); group work/project and collaborative learning (Dunbar et al., 2018). In particular, according to Bandura (1997) self-efficacy is composed of four main sources: past proficient experiences, observations from other people's experiences (substitution experiences), verbal persuasion and the learners' biological and emotional states. Some specific learning activities that help build up the 4 resources include: creating opportunities for social interaction and experiences through academic and extra-curricular activities; giving guidance for group work for students to learn from each other; creating opportunities for students to present, communicate and persuade others; and forming a social and friendly learning environment (social classroom climates) (Bandura, 1997).
In order to help students improve their PsyCap, educators need to coordinate with public institutions and businesses to offer training and coaching programs to help learners improve their communication, adaptiveness, communication skills, as well as understand and master their own strengths and weaknesses, thereby improving their confidence in their abilities (Ayala Calvo & Manzano García, 2021). In addition, some useful programs for them include: entrepreneurial teaching programs (Karlsson & Moberg, 2013) and academic-based leadership programs (Goertzen & Whitaker, 2015). Besides, specific learning interventions such as persuasion, mastery and vicarious experiences concerning goal setting (Luthans et al., 2004) need to be created. There should also be exercises to help solve real-life challenges (Goertzen & Whitaker, 2015) and opportunities for experiential learning (Kolb, 1984). In addition, Goertzen and Whitaker (2015) also emphasized that educators need to design training programs with clear learning outcomes, and forming psychological qualities including Hope, Resilience and Optimism must be one of the important learning outcomes.
Regarding enterprises, SCS (including JOS and CAS) was an important factor affecting organizational productivity (Pachulicz et al., 2008), organizational emotion (Rasdi et al., 2011) andemployees' feelings (Abele &Spurk, 2009). According to the results of the present study, three important factors determining job satisfaction of new IT graduates were PsyCap, SSE and EMP. Therefore, some recommendations for businesses are as follows: more emphasis on the above qualities in the recruitment process; creation of a friendly working environment with increased group interaction and mutual support; appropriate training programs to develop PsyCap, SSE and EMP for employees. It needs to be reiterated that while psychological qualities have a "trait-like" nature (with high stability, and less affected by external factors and little change with learning and accumulation of experience), the other two variables (SSE and EMP) are "state-like" (with low stability, and easily influenced and changed by context with learning and accumulation of experience). SSE and EMP can surely be improved with appropriate activities in a favorable environment. Organizations need to create favorable conditions and strategies (working environment, company culture, management methods, employee training, etc.) to help employees improve SSE and EMP, thereby enhancing their JOS and CAS.
In addition to the above significant findings, it is crucial to highlight certain limitations in this study. Firstly, the study only focused on IT graduates to evaluate the relationship between the variables, thus the ability to apply these results to other fields of work is limited. In order for the results to be more universal and applicable to other fields, future research needs to examine the relationships between these variables in other fields, especially those requiring a high SSE, and with subjects having longer work experience. Second, according to the present study, the relationship between SSE and JOS was not confirmed, and the impact of PsyCap on CAS was not found, which was different from the past studies. More research is needed to investigate and explain these relationships.