“A multidimensional approach in examining the role of self-efficacy on innovative work behavior: Evidence from the creative industry”

Innovative work behavior is a significant factor for business success, including in the creative industry. This paper aims to explore the effect of self-efficacy on the innovative work behavior of employees in the creative industry sector. This study uses a quanti- tative research approach. It was conducted on small and medium enterprises in the creative industry sector in Labuhanbatu and South Labuhanbatu regencies, Indonesia. The participants in this study were employees and managers in the creative industry. The sampling technique is non-probability sampling. A total of 250 questionnaires were distributed, and 216 questionnaires were returned. Therefore, the samples in this study were 216 respondents. After the data were collected, they were processed by the Structural Equation Modeling (SEM) method, which uses a multidimensional approach to testing the hypothesis. The results revealed that innovative work behavior was in- fluenced by self-efficacy. Based on the results of the multidimensional analysis, it was shown that the most influential dimension in measuring innovative work behavior was the idea champion. In addition, the strength dimension dominantly influenced self-efficacy as the most influential dimension. The originality/value of this paper is that the analysis using multidimensional analysis shows that self-efficacy can predict innovative work behavior. The SMEs in the creative industry are suggested to give support to en- hance their capability to improve employee self-efficacy and innovative work behavior.


INTRODUCTION
Small business has a vital role in economic growth (Ngek, 2015), especially in the creative industry sector. Chollisni et al. (2022) also said that the creative industry affects the economy and social welfare. However, the COVID-19 pandemic has made economic growth, including in the creative industry sector, tend to decline. COVID-19 has paralyzed the country's economy and society (Chollisni et al., 2022). It has also had a negative impact on economies around the world (Hoke et al., 2022).
To survive, the firms must innovate in running the business. The firms need innovative human resources at work. Jodi et al. (2019) argued that the creative industries require the skills and expertise of workers or business owners in running their businesses. Innovation must be conducted to be competitive. Human resources are essential as sources of innovation in an organization (Getz & Robinson, 2003). The firms must develop and improve employ-The idea of examining the influence of self-efficacy on innovative work behavior is based on Albert Bandura's social theory that said individual beliefs about their ability would affect their behavior at work (Bandura, 1997). Moreover, the theory of planned behavior is known as the most influential and popular conceptual framework for studying human behavior (Ajzen, 2002). Murwani and Caesar (2016) also indicated that being successful at a job requires high self-efficacy.
However, low self-efficacy will make employees or managers not dare to discover and explore creative ideas to enhance business success. This will lead to low innovation, and the firm's competitiveness will be challenging. Therefore, the firm needs employees with high self-efficacy to achieve business goals.

LITERATURE REVIEW
The study discusses the concept of self-efficacy and innovative work behavior. Then, the paper explains the relationship between self-efficacy and innovative work behavior. First, a review of the literature on innovative work behavior is done.
Innovative work behavior is defined as employees' overall actions or behavior that lead to the emergence of ideas, recognition, and application of something new (Kleysen & Street, 2001). According to Putri et al. (2019), innovating means introducing new things that have renewal quality. Goller and Paloniemi (2017, p. 145) cited that the construct of innovative work behavior is defined as all work activities of employees, both physical and cognitive, both individually or in social interactions, to produce innovation in work.
According to De Jong and Den Hartog (2010), innovative behavior is all individual behavior direct-ed to produce, introduce, and apply new things which are useful at various levels of the organization. Gaynor (2002, p. 57) defines innovative behavior as individual actions to create and adopt ideas/thoughts or new ways to be applied in the implementation and completion of work. All individual actions are directed at the generation of an idea, processing, and application/implementation of new ideas on how to do things, including new product ideas, technologies, work procedures, or processes to improve the organizational effectiveness and success (Nijenhuis, 2015). Thus, it can be explained that innovative employees tend to have creative ideas to solve any existing problem, and can look for opportunities to achieve the firm's goals.
Messmann and Mulder (2012) identify four dimensions of innovative work behavior: opportunity exploration, idea generation, idea promotion, and idea realization. De Jong and Den Hartog (2010) also found four dimensions of innovative work behavior: idea exploration, idea generation, idea champion, and idea implementation. Idea exploration is a dimension of innovative work behavior where employees can find opportunities and ideas for improving work methods. Idea generation is a dimension of innovative work behavior where employees can develop innovative ideas by creating and suggesting ideas to others. Generally, new ideas emerge when employees succeed in exploring ideas. In this dimension, employees look for new methods and techniques in executing tasks, bringing up original solutions, and having concepts about improving work methods. The next is idea championing: it is active and enthusiastic individuals who promote ideas, build support, overcome obstacles, and believe that ideas can be implemented. In this dimension, employees are encouraged to seek support in realizing innovative ideas that have been generated. They also look for coalitions so that new ideas can be implemented and believe in the success of these ideas. Finally, idea implementation is related to employee behaviors like new product or work processes development, and testing and then modifying them.
The second step discusses the literature review on self-efficacy. Albert Bandura proposed the self-efficacy concept first (Agu, 2015). Self-efficacy, or Albert Bandura's social theory, defines an individual's beliefs about the ability to organize and execute a task and do the action required to manage the situations and affect their behavior at work (Bandura, 1997). He also argued that interpreting the changes in one's behavior affects change in oneself and change in the environment, which in turn affects future behavior changes (Newman et al., 2018).
Efficacy is also defined as an individual's ability to produce the desired result (Kar et al., 2017). Self-efficacy reflects an individual's confidence to manage his/her skills required in producing even relatively routine behaviors (Rodgers et al., 2008). Self-confidence is one of the characteristics related to entrepreneurship (Naushad & Malik, 2018). Furthermore, Gibson et al. (2012, p. 159) suggest that self-efficacy is the belief that someone can do enough in a particular situation. A person with high self-efficacy believes there are chances to accomplish a specific task (Kreiner & Angelo, 2010, p. 128); thus, he/she can perform the task successfully (Schermerhorn, 2010, p. 359).
Individuals with high self-efficacy believe in their ability to complete a task more effectively (Campbell & Nelson, 2013, p. 178). Individuals with high self-efficacy can feel more confident, consider difficult tasks as challenges, set high goals, and try harder to overcome the challenges themselves (Hsu et al., 2011). Saleem et al. (2012) claimed that self-efficacy is the name of pride, self-confidence, and trust in one's capabilities so an efficacious employee can succeed. Thus, an employee with high self-efficacy will tend to have confidence in own abilities to work and solve problems at work. Bandura (1997) highlighted three dimensions of self-efficacy, i.e., magnitude, strength, and generality. The level or magnitude dimension is related to the task's degree of difficulty when individuals feel able to do it. Suppose an individual is confronted with tasks arranged according to the level of difficulty. In that case, the individual's self-efficacy may be limited to tasks that are easy, moderate, or even include the most difficult tasks, according to the limits of perceived ability to meet the demands of behavior that are needed at each level. This dimension has implications for the choice of behavior to be tried or avoided. Rodgers et al. (2008) found that the level of task/ job demands represents varying challenges to successful performance. The strength dimension emphasizes the level of strength or stability of the individual against beliefs, even though faced with an unpleasant situation. Self-efficacy reflected that actions taken by individuals would give results that follow the expectations of individuals (Bandura, 1997). The generality dimension is related to the beliefs of individuals to assess their confidence in completing an activity and situation, or a series of activities, that can be done and think to avoid failure in various fields. Individuals can declare themselves to have self-efficacy in broad activities or limited to certain domain functions (Bandura, 1997).
Furthermore, discussing the relationship between self-efficacy and innovative work behavior is vital. Bandura (1997) claimed that individuals' beliefs about their ability would affect their behavior at work. Murwani and Caesar (2016) also noticed that being successful at a job requires high self-efficacy. Self-efficacy has a positive influence on innovative work behavior (Hsiao et al., 2011). The higher the level of self-efficacy, the higher the innovative behavior that will be generated (Kumar & Uzkurt, 2010). Therefore, this study aims to analyze the effect of self-efficacy on innovative work behavior in the creative industry. This study uses a more comprehensive analysis by using multidimensional approach (second-order constacts) to examine the effect of self-efficacy on innovative work behavior. Thus, the following hypothesis is analyzed: H1: Self-efficacy positively influences innovative work behavior.

METHODOLOGY
The population unit in this study is employees and managers in the creative industry sector in Labuhanbatu and South Labuhanbatu regencies. This study used a non-probability sampling technique. This technique was chosen considering the number of populations that were not defined or very large. According to Hair et al. (2013), using CB-SEM, the minimum assumption of data adequacy is 200 data or more. Thus, 250 questionnaires were distributed, and 216 questionnaires were returned. Therefore, the samples in this study were 216 respondents.
This study examines the effect of self-efficacy and innovative work behavior in this study. Each variable was measured using indicators adopted from several existing studies. Innovative work behavior uses four dimensions adopted from De_Jong and Den Hartog (2010) and Messmann and Mulder (2012), i.e., idea exploration, idea generation, idea champion, and idea implementation. Idea exploration uses two items, i.e., looking for the opportunity to innovate and looking for ways to improve current methods. Idea generation is measured with three items, i.e., original solutions to identified problems, a new approach to execute a task, and a new concept for developing ideas. Idea champion uses two items, i.e., making important organizational members enthusiastic and trying to convince people to support the idea. Finally, idea implementation uses two items, i.e., applying innovative ideas in daily work activities and contributing to the implementation of new ideas.
Self-efficacy is measured using three dimensions: magnitude, strength, and generality (Bandura, 1997). The magnitude dimension uses two items, i.e., never avoid the difficult task and directly deal with a difficult task. The strength dimension uses three items, i.e., the belief one can survive and never give up doing a complicated task, the stability of confidence in doing a difficult task, and high expectation of success. Finally, the generality dimension uses three dimensions, i.e., the belief one can be successful in various fields, the belief one can be more successful than colleagues, and the belief one can complete a variety of tasks.
This paper is quantitative research. There are several steps in this study, i.e., normality testing, measurement models, and hypothesis testing performed using structural equation modeling with the help of Amos Version 23 software.

Measurement model
The measurement model in this study was performed using Confirmatory Factor Analysis (CFA) processed by AMOS software version 23. The recommended value of the loading factor must be greater than 0.50. The higher loading factor indicates that the indicator is better at measuring latent variables or more valid for measuring the latent variable. The reliability test was performed by Construct Reliability (CR) and Average Variance Extract (AVE). According to Hair et al. (2013), the value of Construct Reliability (CR) must exceed 0.70, and the value of Variance Extracted (AVE) must exceed 0.50.
The measurement model, as presented in Table  2, shows that the result of factor loadings in second-order constructs and first-order constructs has a value exceeding 0.5. The composite reliability (CR) value also has a value exceeding 0.70. The variance extracted (AVE) results also show the value exceeding 0.5. These results showed that all indicators and dimensions used to measure self-efficacy (magnitude, strength, and generality) are of good validity and reliability to explain or measure self-efficacy.  Table 3. The results of the goodness of fit show that the RMSEA value is smaller than 0.08 (RMSEA 0.000 < 0.08), which means that the RMSEA value is in the good fit criteria. The value of AGFI, GFI, and CFI also exceeds the value of 0.900, so the AGFI, GFI, and CFI are in good fit criteria. The model in the study is fit and suitable, and the study can proceed with the estimation of the model.

Hypothesis testing
Next, the paper tests the hypothesis that has been proposed based on the existing theoretical review and previous studies. The analysis of the Structural Equation Modelling (SEM) was done to test the hypothesis. The proposed structural equation modeling (SEM) is shown in Figure 1. The hypothesis result was tested using the critical ratio (C.R.), or t-value, and probability value (P-Va lue). The critical ratio must be > 1.96±, and the value of probability (P-value) does not exceed 0.05 (Byrne, 2010). The hypothesis test results can be seen in Table 4.
The result proves that self-efficacy positively influences the innovative work behavior of employees  in the creative industry in Indonesia. The result of estimate value (factor loading) is 0.690, critical ratio (t-value) is 8.030, and a significance value is 0.000. Statistically, it can be explained that the effect of self-efficacy on innovative work behavior has a significant effect because the value of the t-value is greater than the t-table (t-value 8.030 ≥ t-table 1.96), and the significance value is less than 0.05 (0.000 < 0.05). This finding suggests that self-efficacy needs to enhance innovative work behavior at the workplace. It further provides evidence that the result supports the hypothesis. This means that the better self-efficacy, the more innovative the employee. Thus, this finding proves that self-efficacy is very important to improving innovative work behavior at work.

DISCUSSION
The focus of this study is to examine the relationship between self-efficacy and innovative work behavior in the context of employees in the creative industry in Indonesia. The results of the hypothesis testing prove that self-efficacy positively and significantly influences the innovative work behavior of an employee in the creative industry. It means that the employee who has high self-efficacy tends to be more innovative at work. Conversely, if an employee has a low self-efficacy, he/she will have a low level of innovative work behavior. This finding was supported by Bandura (1997), who claimed that self-efficacy  would make individuals spend the effort needed to be creative and innovative at work. Bandura's social theory also said that individual beliefs would affect work behavior (Bandura, 1997).
The finding is consistent with Kumar and Uzkurt (2010) regarding the influence of self-efficacy on innovative work behavior. They found that individuals with high self-efficacy would enhance their innovative behavior. Hsiao et al. (2011) also found that self-efficacy has a positive effect on innovative work behavior. Moreover, the higher the self-efficacy, the higher the level of innovative behavior that will be generated ( Based on the results of the multidimensional analysis, it was shown that the most influential dimension to measure the innovative work behavior was the idea champion (with a loading factor of 0.98). De Jong and Den Hartog (2010) found that employees with idea champions are active and enthusiastic about promoting ideas. In addition, employees are encouraged to seek support in realizing innovative ideas that have been generated. Next follows the idea generation and idea implementation dimensions with the same loading factor (0.97).
Furthermore, idea generation has a loading factor of 0.91. That is, the findings explain that the idea champion dimension more dominantly forms the factors that shape innovative work behavior. The higher idea champion was formed by the "I practice the innovative idea in achieving the firm's goals (IWB_8)" indicator with a loading factor of 0.78. For self-efficacy, it was dominantly influenced by the strength dimension with a loading factor of 0.97. The more substantial loading factor of the statement item to form the strength dimension was "I have a strong belief I will be successful in doing a difficult task in this firm (SE_4)" with a loading factor of 0.79. It was followed by the statement "I have a high expectation of being successful in this firm (SE_5)" with a loading factor of 0.75, and "I believe in overcoming the problems and all difficult tasks in this firm (SE_3)" with a loading factor of 0.73.

CONCLUSION
This study aims to examine the effect of self-efficacy on employee innovative work behavior in the creative industry. The results of this study prove that self-efficacy positively influences innovative work behavior. It means that self-efficacy has a role as a valid predictor of innovative work behavior. In other words, self-efficacy will enhance the employee's innovative work behavior. Therefore, an employee with a high level of self-efficacy will be more innovative at work, and it will be easier to achieve the firm's goals. Furthermore, based on the multidimensional analysis, the result showed that the idea champion has a higher loading factor in measuring the innovative work behavior, and the strength dimension has a higher loading factor in measuring self-efficacy.
The firms should foster and develop employees' self-efficacy to be more innovative and achieve organizational goals. They can also train employees to enhance their capability to show high self-efficacy at work.

FUTURE RESEARCH
This study has several limitations; therefore, this study offers some future research suggestions. First, it is worth adding the new dependent variable, namely firm performance. Second, this study only examines the creative industry sector. Thus, future research can consider other sectors, for example, the banking sector, the manufacturing sector, and others. Third, this paper uses a quantitative approach; future research may use mixed methods, namely quantitative and qualitative.