American Journal of Economic and Management Business p-ISSN: XXXX-XXXX e-ISSN: 2835-5199 Vol. 4 No. 6 June 2025 698 THE INFLUENCE OF EMPLOYEE PERCEPTION OF TOP MANAGEMENT LEADERSHIP STYLE ON WORK COMMITMENT AT X Soni Herianto 1 , Samuel Tarigan 2* Sekolah Tinggi Ilmu Ekonomi Harapan Bangsa, Indonesia 1.2 Email: mm-24012@students.ithb.ac.id, samuel_tarigan@ithb.ac.id Abstract This study analyzes the influence of employee perception of top management leadership style on work commitment at x. Using a quantitative approach with a questionnaire survey, this study examines four leadership styles including transformational, democratic, coaching, and servant as well as work commitment which includes affective, normative, and sustainable aspects. Multiple linear regression analysis showed that all four leadership styles had a positive effect on work commitment, with transformational and servant leadership increasing affective commitment, coaching contributing to individual development, and democratic strengthening normative commitment. Additionally, employees' perceptions of organizational fairness reinforce this relationship. The implications of this study highlight the need to implement inclusive and empowerment-based leadership to increase employee loyalty and engagement in the face of operational challenges. The results of this study also contribute to the theory of leadership and human resource management in the aviation navigation service sector. Keywords: Leadership Style, Employee Perception, Work Commitment, Perum LPPNPI INTRODUCTION An effective leadership style can enhance employees' positive perceptions, which in turn increases their work commitment. Research by Avolio et al. (2023) shows that 65% of employees who have a positive perception of top management’s leadership style report a high level of work commitment. In contrast, only 25% of employees with a negative perception of leadership style report a high level of work commitment. Data from a survey conducted by Deloitte (2023) also shows that 80% of companies are shifting toward a more inclusive and participatory leadership style in an effort to increase employee engagement. Perum LPPNPI (Lembaga Penyelenggara Pelayanan Navigasi Penerbangan Indonesia, or Indonesian Aviation Navigation Service Organizing Institution) plays a critical role in aviation navigation services in Indonesia. Its jurisdiction is divided into two Flight Information Regions (FIR): the Jakarta FIR, covering 2,842,725 km², and the Ujung Pandang FIR, covering 4,936,543 km². Combined, the total airspace managed by the Republic of Indonesia amounts to 7,789,268 km², as illustrated in the following figure. Perum LPPNPI operates 295 aviation navigation service unit offices across the Indonesian archipelago, including in West Kalimantan, where the Pontianak branch—designated as a supervisory branch—oversees several The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 699 offices and subordinate units. These include one sub-branch in Ketapang and four unit offices in Sintang, Nanga Pinoh, Putussibau, and Singkawang. Currently, the Pontianak branch is facing significant challenges in terms of human resource availability. Between 2024 and 2025, the number of employees is projected to drop from 115 to 98. This decline is particularly concerning given the addition of a new subordinate branch—Singkawang Airport—which was officially inaugurated on March 20, 2024, by the President of the Republic of Indonesia. Rather than easing operations, this expansion has increased the workload for existing employees, who are now also required to rotate and be assigned to units lacking staff, including the Putussibau, Singkawang, and Nanga Pinoh offices. This situation presents a major dilemma and challenge for top management at the Pontianak branch. Leadership policies differ regarding the appointment and duration of employee assignments to subordinate branches, ranging from three days to one week or even a full month. The variability is largely due to differing leadership styles among managers. Therefore, this study aims to examine how the leadership style applied by top management influences employee perceptions and, consequently, their work commitment. Employees’ perceptions of leadership can significantly affect their motivation, job satisfaction, and loyalty to the organization. It is thus crucial to understand how these perceptions correlate with employee commitment— especially under challenging circumstances such as those currently faced by Perum LPPNPI Pontianak branch. Between 2020 and 2025, several studies have examined the relationship between leadership style and employee work commitment, particularly in the public service sector. One relevant study by Sari and Prasetyo (2021) found that a participatory leadership style enhances employees’ affective commitment in government agencies. Their findings showed that employees involved in decision-making tend to have higher job satisfaction, which leads to stronger organizational commitment. This study will focus on four categories of leadership style. First, according to Usman (2020), transformational leadership is a leadership model rooted in religious, systemic, and cultural values that seeks to inspire innovation and creativity in achieving a shared vision. In contrast, transactional leadership, as explained by Setia Dinata (2023), is a model in which leaders motivate employees to meet goals through a system of rewards based on productivity and performance. Second, Firmansyah & Winarto (2024) describe democratic leadership as an approach that emphasizes active participation from members in decision-making, including offering suggestions and criticism. Leaders using this style value the talents and contributions of their team and provide space for open expression. On the other hand, autocratic leadership, as defined by Sri Wahyuni, Sukatin, Inda Nur Fadilah, and Winda Astri (2022), is The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 700 characterized by centralized control and minimal input from subordinates, with leaders making unilateral decisions. Third, coaching leadership, according to Lamhot Hutapea (2023), involves mentoring and empowering individuals within the organization through constructive feedback and support for their personal and professional development. In contrast, the laissez-faire leadership style—derived from the French phrase meaning ―permission to act‖—is passive. Leaders provide their teams with full autonomy to make decisions and manage tasks, with minimal intervention, as outlined by M. Teguh Saefuddin & Suherman (2024). Fourth, servant leadership, as proposed by Robert Greenleaf (2020), begins with the desire to serve, with leadership emerging as a natural extension of that service-oriented mindset. In contrast, selfish leadership, described by Hlalele Matebese, Zimkhitha F. Juqu, and Normah F. Mutongerwa (2024), focuses on self-interest at the expense of good governance. This style is associated with various negative consequences, such as conflicts of interest, unethical behavior, financial manipulation, and poor organizational asset management. Given this background, the present study aims to analyze how employee perceptions of top management’s leadership style influence their work commitment at Perum LPPNPI. The findings are expected to provide valuable insights for management in formulating strategies to enhance employee commitment, even under suboptimal conditions. Moreover, this research seeks to contribute to the theoretical and practical development of human resource management in the public service sector, particularly within the context of aviation navigation. RESEARCH METHOD This research focuses on employees of x, particularly those in the Pontianak branch and its fostered units, including Ketapang, Sintang, Nanga Pinoh, Putussibau, and Singkawang. The aim of the study is to analyze the impact of employee perceptions of top management leadership styles on work commitment. The research examines transformational, democratic, coaching, and servant leadership styles as the main independent variables, with perceptions of organizational justice acting as a moderating variable, and employee work commitment as the dependent variable. The study is conducted at the Pontianak branch, which serves as the main office, along with its five fostered units. These units—Ketapang, Sintang, Nanga Pinoh, Putussibau, and Singkawang—represent the geographically diverse locations in which the research takes place. The research period began on February 14, 2025, and will continue until May 31, 2025. A quantitative approach is adopted, employing a survey method. The data collected will be analyzed using multiple linear regression and The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 701 Moderated Regression Analysis (MRA) to assess the relationships between leadership styles and employee work commitment. The sample consists of 113 employees, selected through a total sampling technique, as the total population is fewer than 150 individuals. Primary data is collected via questionnaires designed to capture employees’ perceptions of leadership and work commitment, while secondary data is obtained from company reports, academic journals, and other relevant documents. The data analysis process will include exploratory factor analysis (EFA) followed by multiple linear regression to test the hypothesized relationships among the variables. RESULTS AND DISCUSSION Exploratory Factor Analysis (EFA) Transformational Leadership Variables In this section, the Exploratory Factor Analysis (EFA) stage is explained which is carried out to test the unidimensionality of the leadership style variable construct, which consists of four main dimensions, namely transformational, democratic, coaching, and servant. This analysis aims to ensure that each indicator in the variable truly represents the construct in question and has good internal consistency. The stages in the EFA include reliability test, KMO (Kaiser-Meyer-Olkin) test, Bartlett's Test of Sphericity, Measure of Sampling Adequacy (MSA), Total Variance Explained, Communalities, Component Matrix, and factor interpretation. The following are the results of the EFA analysis for the leadership style variables: A Reliability Analysis Test of Transformational Leadership Variables The reliability analysis test aims to measure the level of internal consistency of the statement items in the questionnaire. In this study, the variable of Transformational Leadership (X1) was measured through five statements prepared using the Likert scale with a value range of 1 (Strongly Disagree) to 5 (Strongly Agree). Reliability testing is performed using Cronbach's Alpha method, which is a common statistical technique for assessing the extent to which items in a construct produce consistent measurements. According to Nunnally (1978), an instrument is declared to have good reliability if Cronbach's Alpha value ≥ 0.70. The higher the alpha value, the higher the level of consistency between items in measuring the same construct. The results of the reliability testing of the Transformational Leadership variable can be seen in the following table. Based on the results of the reliability test analysis of the Transformational Leadership variable, a Cronbach's Alpha value of 0.955 was obtained, which indicates a very high level of internal consistency between statement items. This value far exceeds the recommended minimum threshold of 0.70, so it can be concluded that the instrument used is very reliable. Thus, all indicators in the Transformational Leadership variable (X1) are worthy of being maintained because they have met the reliability The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 702 criteria and are considered to be able to consistently measure employees' perceptions of transformational leadership styles within the West Kalimantan Region LPPNPI Region. Based on the Item-Total Statistics table above, all statement items in the Transformational Leadership variable (X1.1 to X1.5) show a Corrected Item-Total Correlation value above 0.834, which means that all five items have a strong and significant correlation to the total construct score. This suggests that each item is consistent in measuring the same aspect of transformational leadership. Cronbach's Alpha if Item Deleted value also indicates that no item if deleted will significantly increase total reliability, as the entire value remains below the total alpha value of 0.955 (i.e. ranges from 0.942 to 0.952). Item X1.3 has the highest correlation (0.893), while item X1.5 has the lowest correlation (0.834), but remains in the very strong category. Thus, all items can be maintained because they are statistically proven to be valid and reliable in representing the Transformational Leadership variable. Based on the Item Statistics table above, all indicators in the Transformational Leadership variable (X1.1 to X1.5) show a fairly high mean value, in the range of 3.76 to 3.96 on the Likert scale of 1–5, which indicates that respondents tend to agree with all statements made in the questionnaire. Items X1.1 and X1.4 had the highest mean value (3.96), while item X1.5 had the lowest mean value (3.76), but still showed positive perception. The standard deviation value ranging from 0.890 to 0.918 indicates that the spread of respondents' answers is moderate and not extreme. This reflects the consistency of respondents' perception of the transformational leadership style applied, and supports the validity of the instrument in measuring the construct of these variables. Results of the Analysis of the SME Test, Barlett's Test, and MSA Transformational Leadership Variables After all indicators on the variables are proven to be reliable, the next step is to test the feasibility of the data to be analyzed through exploratory factor analysis (EFA). This test aims to assess the sampling adequacy and the strength of the correlation between indicators. A strong correlation between items is an important requirement in factor analysis, as it shows that the indicators have the potential to form the same construct. In contrast, non-zero correlation is not expected because it can weaken the structure of the factor. To assess this, the Kaiser-Meyer-Olkin (KMO) test and Bartlett's Test of Sphericity were used, the results of which are shown in the following table. The Kaiser-Meyer-Olkin (KMO) analysis test and Bartlett's Test of Sphericity were conducted to assess the feasibility of data in the application of exploratory factor analysis (EFA). Based on the results of the KMO and Bartlett's Test above, the Kaiser-Meyer-Olkin (KMO) value of 0.908 indicates that the data has an excellent level of sample adequacy for factor analysis, because the KMO value > 0.90 is categorized as "marvelous" The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 703 according to Kaiser. Meanwhile, Bartlett's Test of Sphericity results showed a chi-square value of 587.360 with a degree of freedom (df) of 10 and a significance value (Sig.) of 0.000. Because the significance value < 0.05, it can be concluded that there is a significant correlation between variables, so the data is suitable for further factor analysis. These findings support the assumption that all indicators in the Transformational Leadership variable are strongly related enough to form the same construct. Based on the results of the Anti-Image Matrices for the Transformational Leadership variable above, it can be seen that the values of Measures of Sampling Adequacy (MSA) on the diagonal part of the Anti- image Correlation (marked with a superscript "a") are as follows: X1.1 = 0.914, X1.2 = 0.888, X1.3 = 0.901, X1.4 = 0.917, and X1.5 = 0.921. All of these MSA values were above the recommended minimum limit of 0.50, and all of them were in the very good category (> 0.80), which indicates that each indicator in this variable has a very adequate sample size for factor analysis. Thus, all indicators of Transformational Leadership variables can be maintained because they have met the statistical requirements for further analysis in the Exploratory Factor Analysis (EFA) stage. Test Analysis Results Total Variance Explained Transformational Leadership Variables Total Variance Explained is used to determine the proportion of total variance of all indicators in one variable which can be explained by factors formed through factor analysis. The test aims to identify a significant number of latent factors from a set of indicators on the Transformational Leadership variable. The results of this test analysis provide an overview of the contribution of each component in explaining the structure of the analyzed data. Based on the results of the Total Variance Explained for the Transformational Leadership variable above, it is known that only one main component is formed that has an eigenvalue greater than 1, which is 4.245. This component is able to explain the total variance of 84.901% of the total indicators analyzed. This shows that the five indicators in the Transformational Leadership variable have a very high consistency and lead to one dimension of the dominant factor. Meanwhile, the following components have eigenvalues below 1 and are not used in factor interpretation. Thus, it can be concluded that the variables of Transformational Leadership are unidimensional, and all of its indicators are declared valid to represent the same construct in the factor analysis model. Test Analysis Results Communalities Transformational Leadership Variables The test of the analysis of communalities aims to find out the extent to which the variance of each indicator can be explained by the main factors formed through the analysis of exploratory factors. The extraction value in The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 704 the communalities column illustrates the contribution of each indicator in forming the transformational leadership variable construct. Based on the results of data processing, the values of communalities for each indicator are presented as follows. Based on the results of the above Communalities, all indicators in the Transformational Leadership variable have high extraction values, which shows the significant contribution of each indicator in explaining the main components of the analyzed construct. The highest extraction value is indicated by the X1.3 indicator of 0.847, followed by X1.2 (0.831), X1.4 (0.825), X1.1 (0.824), and X1.5 of 0.730. All of those extraction values are well above the minimum threshold of 0.50 suggested in factor analysis, which means that each indicator has a strong involvement in shaping the Transformational Leadership construct. Thus, all items can be maintained because they have been statistically proven to be able to adequately explain construct variance and are relevant for further analysis. Test Analysis Results Component Matrix Transformational Leadership Variables Factor Matrix is part of factor analysis that shows the factor loading value of each indicator against factors formed based on the Principal Axis Factoring method. This value represents how much each indicator contributes in forming the latent factor (main component) of a construct. The higher the loading value, the stronger the indicator is in representing the dimensions of the variable being measured. Based on the results of data processing, the factor loading values for each indicator in the Transformational Leadership variable were obtained as follows: Based on the Factor Matrix table above, all indicators in the Transformational Leadership variable have a very high factor loading value , which is between 0.855 to 0.920. The highest score was found in the X1.3 indicator ("My leader encourages innovation and creativity in work") of 0.920, followed by X1.2 ("My leader has a clear vision and can communicate well") of 0.912, and X1.1 and X1.4 of 0.908 respectively. Meanwhile, the X1.5 indicator ("My leader builds my awareness of the organization's goals, not just relying on rewards or punishments") has a loading value of 0.855. All of these values show that each indicator has a very strong contribution in forming one main factor, namely transformational leadership style. Therefore, all indicators are considered highly representative and worthy of being maintained in the measurement model. Interpretation of Factors Variable Transformational Leadership Based on the results of factor analysis displayed in the Factor Matrix, all indicators in the Transformational Leadership variable showed a very high factor loading value , ranging from 0.855 to 0.920. These values indicate that each item has a strong contribution in shaping the overall construct of The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 705 Transformational Leadership. The indicator with the highest score was X1.3 ("My leader encourages innovation and creativity at work") of 0.920, which indicates that the innovation aspect is the dimension that best reflects the character of transformational leadership in the context of the organization studied. Furthermore, the X1.2 indicator ("My leader has a clear vision and can communicate well") has a loading value of 0.912, followed by X1.1 ("My leader provides inspiration and motivation to achieve common goals") and X1.4 ("My leader explains the importance of change for shared growth") which recorded a value of 0.908, respectively. The X1.5 indicator ("My leader builds my awareness of the organization's goals, not just relying on rewards or punishments") also has a strong contribution with a value of 0.855. These findings reinforce that the five items in the questionnaire have very high construct validity and consistently describe the essence of transformational leadership. Dimensions such as innovation, vision, motivation, awareness, and the ability to facilitate change have proven to be the main components that form this leadership style within the organizational framework of the West Kalimantan Region Perum LPPNPI. Exploratory Factor Analysis (EFA) Democratic Leadership Variable This section describes the Exploratory Factor Analysis (EFA) process used to test the unidimensionality of the variable construct of leadership style, specifically Democratic Leadership as one of the four main dimensions: transformational, democratic, coaching, and servant. The EFA is performed to ensure that each indicator in the variable truly represents the construct in question and has adequate internal consistency. The analysis stages include reliability test, Kaiser-Meyer-Olkin test (KMO), Bartlett's Test of Sphericity, Measure of Sampling Adequacy (MSA), Total Variance Explained, Communalities, Component Matrix, and factor interpretation. The following are presented the results of the EFA for the Democratic Leadership variable. A Test of Analysis of the Reliability of Democratic Leadership Variables The reliability analysis test aims to assess the internal consistency of the five statements used in measuring the Democratic Leadership variable. The test uses Cronbach's Alpha approach, a commonly used statistical method to measure the extent to which items in a construct show consistency in results. Based on the table above, it is known that Cronbach's Alpha value for the Democratic Leadership variable is 0.954 which indicates a very high level of internal consistency among all statement items in this variable. This value far exceeds the minimum threshold of 0.70 as suggested by experts, which means that the instruments used are very reliable. Thus, all items in the Democratic Leadership variable can be maintained because they have met the The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 706 reliability criteria and are consistently able to measure respondents' perceptions of democratic leadership practices in the work environment of x. Based on the Item-Total Statistics table for the Democratic Leadership variable, all items have a Corrected Item-Total Correlation value above 0.80, indicating that each item has a very strong correlation with the total scale score. The highest score was achieved by the X2.3 item with a correlation of 0.908, which means that this item is most representative in reflecting the overall democratic leadership construct. Additionally, Cronbach's Alpha if Item Deleted values range from 0.937 to 0.952, indicating that there is not a single item that, if deleted, would significantly improve the reliability of the scale. This means that all items have a positive contribution to the instrument's reliability and are worth preserving in subsequent analysis. Based on the Item Statistics table above for the Democratic Leadership variable, it can be seen that all items were measured by 113 respondents. The highest mean was found in the X2.1 item of 4.04, indicating that most respondents tended to strongly agree with the statement that their leaders encourage engagement in decision-making. Meanwhile, the lowest average score is found in item X2.5 which is 3.84, although it is still in the "Agree" category. The standard deviation of each item ranged from 0.834 to 0.996, which indicates that there is a fairly reasonable variation in respondents' answers and not too spread to the extreme. This data shows that respondents' perception of democratic leadership indicators tends to be positive and relatively consistent. Results of SME Testing Analysis, Barlett's Test, and the Democratic Leadership Variable MSA After all indicators of the Democratic Leadership variables are declared reliable, the next stage is to test the feasibility of the data to be analyzed through Exploratory Factor Analysis (EFA). The purpose of this test is to assess the sampling adequacy and the extent to which the indicators have an adequate correlation with each other. Strong correlations between indicators are important in EFA, as they suggest that the items may measure the same construct. In this case, there is no expected non-zero correlation, as this will weaken the basis of factor analysis. Therefore, the Kaiser-Meyer-Olkin (KMO) test and Bartlett's Test of Sphericity were used to ensure that the data is worthy of further analysis through factor techniques. The test results are presented in the following table. Based on the results of the analysis of KMO and Bartlett's Test above for the Democratic Leadership variable, a Kaiser-Meyer-Olkin (KMO) value of 0.908 was obtained, which is above the minimum threshold of 0.50. This shows that the data has an excellent sample feasibility rate for factor analysis. In addition, the results of Bartlett's Test of Sphericity showed a value of Approx. Chi-Square is 586,810 with a significance of 0.000 (smaller than 0.05), which means that there is a fairly strong and significant correlation The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 707 between items. Thus, the data are eligible for Exploratory Factor Analysis (EFA) and the Democratic Leadership variable deserves further analysis in its factor structure. Based on the Anti-image Matrices table for the Democratic Leadership variable, all Measures of Sampling Adequacy (MSA) values shown on the main diagonal have values above 0.5, which ranges from 0.892 to 0.931. The value indicates that each indicator in this variable has a good sample feasibility for further analysis through Exploratory Factor Analysis (EFA). A high MSA indicates that each item has an adequate correlation with other indicators in one construct, so that all indicators in the Democratic Leadership variable can be maintained because they have met the requirements for statistical validity of the construct. Test Analysis Results Total Variance Explained Variables of Democratic Leadership Total Variance Explained is used to determine the proportion of total variance of all indicators in one variable that can be explained by the factors formed. This analysis aims to identify the number of latent factors that emerge from a series of indicators on the Democratic Leadership variable. The test results provide an idea of the extent to which each component is able to explain the structure of the variables as a whole. Based on the Total Variance Explained table, it can be seen that there is only one main component that has an eigenvalue greater than 1, which is 4,231. This component was able to explain 84.619% of the total data variation from all indicators in the Democratic Leadership variable. This percentage is very high, which shows that one factor alone is enough to represent the entire construct of democratic leadership as a whole. These results reinforce the assumption of unidimensionality, where all indicators come together in one common and consistent component in describing the concept of democratic leadership. Thus, all indicators used in this variable have been shown to have excellent representative power against the single factor that is formed, and are suitable for use in advanced analysis. Test Analysis Results Communalities Variables of Democratic Leadership The test of the analysis of communalities aims to find out the extent to which the variance of each indicator can be explained by the main factors formed through the analysis of exploratory factors. The extraction value in the communalities column describes the contribution of each indicator in forming the construct of the Democratic Leadership variable. Based on the results of data processing, the value of communalities is obtained as shown in the following table. Based on the results of Communalities for the Democratic Leadership variable, all indicators show a high extraction value, which reflects the magnitude of each item's contribution in explaining one key component. The highest extraction value was found in the X2.3 indicator ("My leader receives The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 708 input openly") of 0.876, followed by X2.2 ("My leader considers the opinions of subordinates in decision-making") of 0.847, and X2.4 ("My leader provides a space for discussion before decisions are made") of 0.852. Meanwhile, X2.5 ("My leader opens a discussion space free from hierarchical pressure") has a value of 0.760, and X2.1 ("My leader encourages active participation in solving work problems") shows a value of 0.709. All of these extraction values are above the minimum limit of 0.50 suggested in the factor analysis, so it can be concluded that all indicators in the Democratic Leadership variable are worth maintaining because they have a strong contribution in shaping the measured construct. Test Analysis Results Component Matrix Variables of Democratic Leadership The Factor Matrix is part of factor analysis that presents the factor loading value of each indicator against the factors formed based on the Principal Axis Factoring method. This loading value reflects how much each indicator contributes or relates to one latent factor underlying the variable construct. The higher the factor loading value, the stronger the indicator is in representing the dimensions of the construct in question. The following are the results of data processing for the Democratic Leadership variable. Based on the Factor Matrix table above, all indicators in the Democratic Leadership variable have a very high factor loading value , which ranges from 0.842 to 0.936. The highest score was indicated by the X2.3 indicator ("My leader accepts input openly") of 0.936, followed by X2.4 ("My leader provides a space for discussion before decisions are made") of 0.923, and X2.2 ("My leader considers subordinate opinions in decision- making") of 0.920. Meanwhile, the X2.5 indicator ("My leader opens a discussion space free from hierarchical pressure") has a value of 0.872, and X2.1 ("My leader encourages active participation in solving work problems") of 0.842. All of these values are well above the minimum threshold of 0.50, so it can be concluded that the five indicators are very representative in forming the Democratic Leadership construct and are worthy of being maintained in the measurement model. Interpretation of Factors Variable Democratic Leadership Factor Matrix is part of factor analysis that presents the factor loading value of each indicator against the latent factor formed, based on the Principal Axis Factoring method. This value illustrates the strength of each indicator's contribution in forming a single complete construct. Based on the results of the analysis on the Factor Matrix, all indicators in the Democratic Leadership variable showed a very high factor loading value , which was between 0.842 to 0.936. The highest value is indicated by the X2.3 indicator ("My leader accepts input openly") of 0.936, which indicates that openness to input is the most reflective aspect of a democratic leadership style. Furthermore, the X2.4 indicator ("My leader provides a discussion space before the decision is made") has a loading value of 0.923, followed by The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 709 X2.2 ("My leader considers subordinates' opinions in decision-making") of 0.920. Indicator X2.5 ("My leader opens a discussion space free from hierarchical pressure") obtains a value of 0.872, and X2.1 ("My leader encourages active participation in solving work problems") shows a value of 0.842. All of these indicators show a very strong contribution in forming one main factor, namely the construction of democratic leadership. Thus, it can be concluded that the five indicators consistently support each other in representing one dimension of democratic leadership. This strengthens the validity of the construct and confirms that all items are worthy of being maintained in the measurement model to illustrate the democratic leadership style in the context of the West Kalimantan Regional LPPNPI Perum. CONCLUSION The initial results indicated that none of the four leadership styles— transformational, democratic, coaching, and servant—had a significant effect on work commitment when analyzed simultaneously in the regression model. This prompted further analysis using an exploratory approach and the application of a stepwise method, resulting in a more optimal and representative model. The final regression model identified transformational leadership (X1) and coaching leadership (X3) as the two leadership styles that significantly influenced work commitment. These two styles were shown to play a dominant role in shaping employee commitment, offering a simple yet statistically robust model. Transformational leadership was effective in inspiring and motivating employees, while coaching leadership enhanced employees' skills, boosted motivation, and provided personal support in professional tasks. Although democratic and servant leadership styles were not statistically significant, they still demonstrated a generally positive relationship with work commitment. The findings suggest that strengthening transformational and coaching leadership styles is a relevant strategy for improving work commitment at Perum LPPNPI. This study also provides practical implications for developing evidence-based leadership training policies and programs aimed at fostering a more productive, professional, and safety-oriented work environment, thereby enhancing overall organizational performance. REFERENCE Arfandi, Y., & Kasran, M. (2023). The influence of leadership style on organizational performance. Journal of Management and Business, 12(1),45–60. The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 710 https://download.garuda.kemdikbud.go.id/article.php?article=1111999 &title=Pengaruh+Gaya+Kepemimpinan+Terhadap+Kinerja+Organisas i&val=16775 Avolio, B. J., Walumbwa, F. O., & Weber, T. J. (2023). Leadership: Current theories, research, and future directions. Annual Review of Organizational Psychology and Organizational Behavior, 10, 1–25. https://europepmc.org/article/MED/18651820 Bass, B. M., & Riggio, R. E. (2006). Transformational leadership (2nd ed.). Lawrence Erlbaum Associates. https://www.taylorfrancis.com/books/mono/10.4324/9781410617095/tr ansformational-leadership-bernard-bass-ronald-riggio Cherry, K. (2022). The democratic leadership style. Verywell Mind. https://www.verywellmind.com/democratic-leadership-2795654 Deloitte. (2023). Global human capital trends 2023: The future of work. https://www2.deloitte.com/us/en/insights/focus/human-capital- trends/2023.html Hadi, S., Fitriana, H., Kirana, K. C., Subekti, N. B., & Ogwu, I. J. (2023). The impact of transformational leadership on innovation performance: A self-efficacy mediating analysis. Journal of Leadership in Organizations, 5(2), 132–154. https://jurnal.ugm.ac.id/leadership/article/view/86695 Indahsari, R., & Manafe, A. (2022). Employee perception of leadership style and work commitment. Journal of Psychology and Education, 9(2), 123–135. https://jiss.publikasiindonesia.id/index.php/jiss/article/download/459/1 157 Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755–768. https://www.researchgate.net/publication/8210096 Meyer, J. P., & Allen, N. J. (2020). A three-component model of organizational commitment: Revisited. Journal of Applied Psychology, 105(5), 87–103. https://www.wilmarschaufeli.nl/publications/Schaufeli/259.pdf Northouse, P. G. (2018). Leadership: Theory and practice (8th ed.). Sage Publications. https://books.google.com/books/about/Leadership.html?id=Gx5zDwA AQBAJ Salanova, M., Bakker, A. B., & Llorens, S. (2022). Employee engagement and commitment: The role of transformational leadership. Journal of Organizational Behavior, 43(4), 129–145. https://link.springer.com/article/10.1007/s10902-005-8854-8 https://download.garuda.kemdikbud.go.id/article.php?article=1111999&title=Pengaruh+Gaya+Kepemimpinan+Terhadap+Kinerja+Organisasi&val=16775 https://download.garuda.kemdikbud.go.id/article.php?article=1111999&title=Pengaruh+Gaya+Kepemimpinan+Terhadap+Kinerja+Organisasi&val=16775 https://download.garuda.kemdikbud.go.id/article.php?article=1111999&title=Pengaruh+Gaya+Kepemimpinan+Terhadap+Kinerja+Organisasi&val=16775 https://europepmc.org/article/MED/18651820 https://www.taylorfrancis.com/books/mono/10.4324/9781410617095/transformational-leadership-bernard-bass-ronald-riggio https://www.taylorfrancis.com/books/mono/10.4324/9781410617095/transformational-leadership-bernard-bass-ronald-riggio https://www.verywellmind.com/democratic-leadership-2795654 https://www2.deloitte.com/us/en/insights/focus/human-capital-trends/2023.html https://www2.deloitte.com/us/en/insights/focus/human-capital-trends/2023.html https://jiss.publikasiindonesia.id/index.php/jiss/article/download/459/1157 https://jiss.publikasiindonesia.id/index.php/jiss/article/download/459/1157 https://www.researchgate.net/publication/8210096 https://www.wilmarschaufeli.nl/publications/Schaufeli/259.pdf https://books.google.com/books/about/Leadership.html?id=Gx5zDwAAQBAJ https://books.google.com/books/about/Leadership.html?id=Gx5zDwAAQBAJ https://link.springer.com/article/10.1007/s10902-005-8854-8 The Influence of Employee Perception of Top Management Leadership Style on Work Commitment At X 711 Sari, D. P., & Prasetyo, A. (2021). Participatory leadership style and employee commitment in government agencies. Journal of Public Administration, 8(3), 201–215. https://ejournal.unsrat.ac.id/index.php/jab/article/download/47635/4230 1 Ting, H., Wan, S., & Zhang, Y. (2021). The impact of transformational leadership on employee commitment and innovation. International Journal of Management Studies, 8(3), 45–67. https://journal.ugm.ac.id/leadership/article/view/86213 Yukl, G. (2013). Leadership in organizations (8th ed.). Pearson. https://archive.org/details/leadershipinorga0000yukl_k0v0 Copyright holders: Soni Herianto, Samuel Tarigan (2025) First publication right: AJEMB – American Journal of Economic and Management Business https://ejournal.unsrat.ac.id/index.php/jab/article/download/47635/42301 https://ejournal.unsrat.ac.id/index.php/jab/article/download/47635/42301 https://journal.ugm.ac.id/leadership/article/view/86213 https://archive.org/details/leadershipinorga0000yukl_k0v0