The dimensions of competency on worker performance mediated by work commitment

Abstract Weaving workers in Bali are inherited from generation to generation. The ability to weave is obtained without special education/training. The ability to work weaving is obtained independently of the environment. In addition, the difficulty of finding other jobs causes workers to keep working. These two things are important factors in determining the performance of weaving workers in Bali. This study aims to analyze the effect of work competence (knowledge, skills, and attitudes) on employee commitment and performance, the effect of a work commitment on employee performance, and the role of work commitment in mediating the relationship of competence with employee performance. This research was conducted on 168 weaving workers in Bali. Data collection was carried out through distributing questionnaires to research respondents. The data collected was analyzed using the SmartPLS 3.0 program. The results found competencies consisting of knowledge, skills, and attitudes showed a significant positive effect on employee commitment and performance, work commitment also had a significant positive effect on performance. Work commitments also act as mediators of the relationship of competence with employee performance.


Introduction
Weaving craftsmen in Bali are hereditary works carried out by several Balinese people since the time of the kingdom, and until now some people still carry out this activity as one of their livelihoods, in addition to agriculture and trade. One of the woven products in Bali which is still very well known locally, nationally and even internationally, is endek woven fabric. To further strengthen the existence of endek fabrics, as well as to motivate woven workers, the Bali regional government issued a policy that government employees and private employees on certain days must use clothing made from endek cloth. The existence of regulations that require all civil servants and private employees to wear traditional Balinese clothing, makes endek weaving craftsmen more eager to innovate and creativity in designing endek cloth ornaments.
In an effort to create products that are always in demand by consumers, craftsmen must have the ability to make a variety of endek fabric products with various attractive ornaments. In order to be able to make endek fabrics with various attractive ornaments, every craftsman must have knowledge about various ornamental designs that are liked by the public. In addition, craftsmen must have the ability (skill) to apply designs/ornaments to work on weaving. What cannot be realized is that every worker must have a positive attitude (attitude) in the development of design changes that are in the public interest. Knowledge, skills and attitudes are included in competencies (Hanafi & Ibrahim, 2018;Spencer & Spencer, 1993) In addition to competence, the consistency of weaving workers in processing this work is also very important. Aside from being part of efforts to get income to meet the family economy, this activity is also an effort to preserve cultural heritage. Worker commitment is important in efforts to maintain the fabric of endek which is a permanent Balinese cultural heritage and to be part of the income that can sustain the lives of workers.
The competence and commitment of weaving workers is an important part in relation to performance. Weaving worker competencies are not obtained from formal or informal educational institutions. The ability to have it go down. The ability to possess is obtained through frequent seeing, trying, and continuing to try, so that it can be adjusted and skilled in doing work. Likewise, knowledge about various equipment and weaving processes is obtained by self-taught. The only ability they have causes workers not to switch to another job and continue to do this work as one of the family's livelihoods. Because workers have never received formal or informal education, their knowledge and skills are relatively low, which results in sub-optimal performance. A number of references mention that knowledge and skills are important factors in relation to performance (Almusaddar et al., 2018;Hanafi & Ibrahim, 2018).
Based on this background, this study aims to determine the effect of competency dimensions (knowledge, skills, attitude), and commitment on performance of the weaving workers in Bali, Indonesia.

Work performance
According to the behavior approach in management, employee performance is the quantity or quality of something produced or services provided by someone who does the work of Luthans (2006). Mathis and Jackson (2006) states that performance is what someone does and produces. There are several indicators to state one's performance, namely task performance, adaptive performance, contextual performance (Rubindra & Lalatendu, 2017;Tabiu et al., 2016), organizational citizenship behavior and counterproductive behavior (Rotundo & Sackett, 2002;Viswesvaran & Deniz, 2000), Quantity, Quality, knowledge, cooperative, dependability, initiative, personal qualities (Gomes, 2003).
There are several factors that can affect a worker's work performance, such as work competence, and the level of one's commitment to work. The more competent someone is, the better the performance (Sujana, 2012). This is evidenced by a number of studies conducted by Setyaningtyas et al. (2013), Lotunani et al. (2014), Renyut et al. (2017), and Martini et al. (2018). However, in research conducted by Dubey and Sadia (2012), Sari (2013), and Shodikin (2013) found competence had no significant effect on employee performance. Likewise, with work commitments, the higher one's commitment to work, the better the performance (Harinoto & Bogetriatmanto, 2018). A number of research results have proven this statement, such as Lotunani et al. (2014), Bandula and Jayatilake (2016), Gunawan et al. (2017), and Metin and Asli (2018). But in a study conducted by Hidayat (2010), Sarmawa et al. (2015), and Raveendran and Gamage (2019) found commitment had no significant effect on performance. Spencer and Spencer (1993), Mulyasa (2004); Hutapea (2008) state that work competence is a basic characteristic of a person consisting of knowledge, skills, and attitude that have a causal relationship with extraordinary work performance or work effectiveness. Wibowo (2007) argues that: work competence is an ability to carry out or do a job or task based on skills and knowledge and is supported by the work attitude demanded by the job.

Work competence
Work competence is an individual characteristic that contributes to one's success in completing work (Sjahrazad et al., 2009). Whereas according to Boulter et al. (2003) competence is a characteristic that enables a person to perform excellently. Spencer & Spencer (1993) also suggested competency as a characteristic of a person who directly impacted his performance at work. More explicitly said by Sujana (2012), the higher the competency the higher the performance. A number of research results prove that there is a significant positive effect of work competence on individual performance, namely Setyaningtyas Rabo (2018), and Martini et al. (2018). Based on the results of this study, the research hypothesis is stated as follows: Hypothesis 1: Work competence has a significant positive effect on work performance Noe (2009) (in Hsiao & Lin, 2018), defines the work competence of a person's ability to do his work with his skills, knowledge, attitudes, and personality. The same thing was stated by Goffin and Woycheshin (2006). Vindyani and Desiana (2018) in her research found that knowledge (one of the dimensions of work competence) had a significant positive effect on work commitment. It was also said, that someone who has work competence in a job is less likely to leave the job he is doing. The results of other studies in line with this are Setyaningtyas et al. (2013), Katawneh and Osman (2014), Kuhuparuw and Ferdinandus (2014), Lotunani et al. (2014), Suriadi et al. (2018), Zhang et al. (2018), and Martini et al. (2018). Based on the results of this study, the research hypothesis was built, namely: Hypothesis 2: Work competence has a significant positive effect on work commitment

Work commitment
Some researchers cite work commitment or employee commitment as organizational commitment (Allen & Meyer, 1990;Colquit et al., 2009). Work commitment is a psychological state that characterizes worker relations with work (Carmeli & Freund, 2004;Dan Metiboba, 2012). Bratton and Gold (2007) state that work commitment is a sense of individual commitment to work.
Workers who have good work commitments are more stable at work (Mowday et al., 1982). Nguyen and Aida (2014), someone with a strong work commitment will not leave work or will always pursue the job. A person will have better work results if based on better work commitments, this is stated in research (Harinoto & Bogetriatmanto, 2018).
A number of other research findings also found that work commitment had a significant positive effect on employee performance, namely Setyaningtyas et al. (2013), Lotunani et al. (2014), Bandula and Jayatilake (2016), Gunawan et al. (2017), Renyut et al. (2017), Sugihartono (2018), Paramita et al. (2018), and Hidayah and Tobing (2018), Metin and Asli (2018). Based on the results of the research, the research hypothesis is proposed, namely: Hypothesis 3: Work commitment has a significant positive effect on work performance The results of a study of a number of studies found that work competence influences employee work commitment (Lotunani et al., 2014;Setyaningtyas et al., 2013;Suriadi et al., 2018). On the other hand, work commitment also influences job performance (Setyaningtyas et al., 2013;Lotunani et al., 2014;Bandula & Jayatilake, 2016;Gunawan et al., 2017;Renyut et al., 2017;Sugihartono, 2018;Paramita et al., 2018;Hidayah & Tobing, 2018;Metin & Original, 2018). A number of studies also found that work commitment significant as a mediator in the relationship of work competence to work performance (Bandula & Jayatilake, 2016;Renyut et al., 2017;Sugihartono, 2018). Based on this, the research hypothesis is proposed, namely: Hypothesis 4: Work commitment plays a significant role as a mediator in the relationship between competence and performance

Research method
This research was designed with a quantitative approach, to find influence between competency variables (independent variables), work commitment (mediating variables), toward work performance (dependent variable). The research population is endek weaving craftsmen, of which 256 craftsmen are spread across 7 districts/cities in Bali. The number of samples was determined through the Slovin formula with a precision of 5% (n = N/(1 + N.e 2 ) (Yusuf, 2014, p. 170), so the number of samples became 168 weaving workers. Determination of the number of samples and respondents in each district/city was carried out proportionally sampling method. Population distribution and research samples in each district/city are shown in Table 1.
Data collection is carried out by distributing questionnaires for weaving workers. Questionnaire answers are prepared in the form of a Likert scale with a range of 1-5. Furthermore, data processing is done with SmartPLS 3.0, through three stages namely the measurement model evaluation stage (measurement model evaluation), structural model evaluation stage (structural model evaluation), and hypothesis testing stage. The evaluation phase of the structural model consists of evaluating validity and reliability through convergent validity and Composite Reliability.

Validity and reliability
Validity and reliability testing aims to determine the level of validity and reliability of indicators. The validity test results of the research indicators indicate that there are a number of invalid indicators because they have an outer loading <0.50, so they must be released from the research model, and repeated validity is performed. The results of the validity and reliability after reconstruction are shown in Table 2. Testing the validity of the data is based on convergent validity measured by external loading values. Table 2 shows that valid and reliable indicators as supporting models have an external loading value> 0.50 and Composite Reliability> 0.70. Based on testing the validity and reliability as shown in Table 2, where the outer loading value> 0.50 and composite reliability value> 0.70, then all indicators are declared valid and reliable.

Evaluation fit model
Testing the accuracy of the model is done through the R-square test (R 2 ), Q-Square predictive relevance (Q 2 ), and Goodness of Fit (GoF). R 2 coefficient of influence of competence on work commitment shows a value of 0.612 which means 61.2% work commitment is influenced by competence. Similarly, the R 2 of the influence of competence and work commitment on work performance is 0.752. This means that work performance 75.2% is influenced by competence and work commitment, the rest are other factors outside the research model. Q Square Predictive Relevance evaluation results are calculated based on formula Q 2 = 1-(1-R 2 1 ) (1-R 2 2 ), showing the results of 0.9038, meaning that the model is able to provide predictive results of 90.38%. If seen from the results of the evaluation of Goodness of Fit calculated by the formulation of GoF = √ (AVE x R 2 ), it produces a value of 0.6364 (large). Based on the results of the evaluation model through R 2 , Q 2 , and GoF, the model is classified as a good model, so the next process can be continued.

Hypothesis testing
The next process is to examine the research hypothesis which consists of the influence of competence on work commitment and work performance, the effect of a work commitment on work performance, and the role of mediation work commitment on the relationship of competence to work performance. To find out this can be shown in Figure 2 and Table 3. Based on Table 3 and Figure 2, it can be seen that the value of the work competence path to work commitment is 0.782 with a p-value of 0.000 < 0.05. This means that the effect of work competence on work commitment is significantly positive. It also means that hypothesis 1 which states that work competence has a significant positive effect on work commitment can be accepted. The results of this study illustrate that increasing work competence (work knowledge, work skills, and work attitudes) can increase work commitment. This is in line with the results of research conducted by a number of researchers namely Setyaningtyas et al. (2013), Katawneh and Osman (2014), Kuhuparuw and Ferdinandus (2014), Lotunani et al. (2014), Suriadi et al. (2018), Zhang et al. (2018), and Martini et al. (2018).
The effect of work competence on employee performance is shown by the path value of 0.310 with a p-value of 0,000. That is, work competence has a significant positive effect on employee  performance. Thus, hypothesis 2 which states that work competence has a significant positive effect on employee performance can be accepted. Improving work competence can significantly improve employee performance. The results of this study are in line with the results of previous studies, namely Setyaningtyas et al. (2013), Lotunani et al. (2014), Renyut et al. (2017), Mukhtar (2018), Mahmood et al. (2018), Sari and Lestari (2018), Keerthy and Biyu (2018), Suriadi et al. (2018), Suhardi and Sari (2018), Zhang et al. (2018), Berliana et al. (2018), Rabo (2018), Martini et al. (2018), and Gunawan and Gunawan (2019).
The effect of a work commitment on work performance is shown by the path value of 0.603 with a p-value of 0,000. The results of this study indicate that work commitment has a significant positive effect on work performance, so hypothesis 3 which states that work commitment (affective, normative, and continuing) is acceptable. The better work commitment can also affect the improvement of work performance, this is supported by the results of previous studies, namely Setyaningtyas et al. (2013), Lotunani et al. (2014), Bandula and Jayatilake (2016), Gunawan et al. (2017), Renyut et al. (2017), Sugihartono (2018), Paramita et al. (2018), Hidayah and Tobing (2018), Metin and Asli (2018), and Sungu et al. (2019).
Testing the role of mediating work commitment in the relationship of work competence with work performance is shown by a path value of 0.472 and a p-value of 0,000. This means that work commitment acts as a mediator in the relationship of work competence with performance. Because the direct relationship that occurs in competence to work commitment and work performance is significant, so also the direct relationship with work commitment to work performance is also significant, so work commitment acts as a partial mediation. Statistically, it is clearly seen that an increase in the path value from 0, 310 from the direct influence of work competence to performance, becomes 0.472 (the indirect effect of work competence on performance through work commitment). That is, work commitment can improve the relationship of work competence to work performance for the better. This means work commitments that increase work competence on the performance of weaving workers in Bali. These results are in line with the results of research conducted by Bandula and Jayatilake (2016), Renyut et al. (2017), and Sugihartono (2018).

Conclusion, recommendation, and contribution of study
This study concludes that competencies that are proxied through knowledge, skills, and attitudes are positively significantly capable of influencing work commitments (affective, normative, and continuing). Work competence is also able to significantly positively influence the performance of weaving workers in Bali. This means that a significant increase in work knowledge, work skills, and work attitudes can drive work commitment and performance of weaving workers in Bali. Work knowledge has the biggest contribution among other competency elements. This means that increasing work knowledge has the greatest impact compared to other elements of competence. Therefore, in an effort to improve the performance of workers, especially weaving workers in Bali, it is necessary to prioritize increasing work knowledge, in addition to increasing other competency elements. Other competency elements (work skills and work attitudes) also have a major contribution to work performance.
Work commitments also have a strong contribution in relation to the performance of weaving workers in Bali. Affective commitment as a picture that workers always want to continue to pursue their current work, has a strong contribution to the relationship with employee performance. This is understandable because this ability is the only one that has been passed down from generation to generation, making it difficult to let go and find another job. Strengthening effective commitment (can be done by giving proper appreciation) will be able to improve employee performance. Besides that, paying attention to other elements of work commitment (normative commitment and continuing commitment) are also very important for business sustainability. Integration of work competencies and work commitments have a greater impact on workers' performance, therefore considering these two factors together is very important for better performance improvement.
The results of this study provide information that workers' competencies and commitments are very important in relation to the performance of weaving workers in Bali. The results of this study are able to strengthen the results of previous studies, which found that work commitment of workers has an important role in the relationship between work competence and performance.

Limitation and future research
This study still has a number of limitations that require more in-depth research, especially in the scope of research is still limited to the variable work competence and work commitment, in relation to employee performance. Theoretically, empirically, and logically, a worker's performance is influenced by very complex factors. Therefore, future research is expected to expand this research by adding other relevant variables. To strengthen the findings of this study, this research model can be applied to other research objects or locations. Thus, the results of research on factors that affect workers' performance become more complete and varied.