Transactional and Transformational Leadership Styles as Predictors of Employee Performance During the Covid-19 Crisis and the Mediating Role of Organisational Culture

The purpose of this study was to examine the impact of transformational and transactional leadership styles on employee performance. The mediating role of organisational culture was also examined. This quantitative study utilised a survey approach to collect primary data from 165 respondents engaged in the private sector in Jakarta, Indonesia. The data was collected during the Covid 19 pandemic. Structural equation modelling using Smart Pls was used to analyse the data. The data analysis showed that only transformational leadership behaviour had a strong and significant impact on employees' performance. However, the impact of transactional leadership on employee performance was not significant. In addition, the organisational culture had an indirect effect on employee's performance. The practical implications suggest that leaders should adopt transformational leadership to inspire and motivate employees. Transformational leaders must develop the inner agility to resolve problems by being open-minded, flexible and adopt best practices. This study provided new insights from the theoretical perspective, and the findings were consistent with the transformational leadership model. From an originality perspective, this study found that organisational culture mediates the relationship between leadership styles on employee performance.


Introduction
Indonesia is today the world's fourth most populated nation. It has a total population of about 270 million people. Indonesia is also the largest economy among the nations in Southeast Asia. Indonesia's economy has shown an uptrend, with GDP increasing from USD755 billion in 2010 to USD 1,042 trillion in 2018 (The World Bank, 2018). Since 1990, Indonesia's real GDP has grown at a rate of 5% per year on average. The high price of commodities, favourable demographic patterns, and a sound macroeconomic policy framework all contribute to Indonesia's development and growth. Between 1990 and 2018, the growth resulted in a high six-fold rise in GDP per capita (World Bank Group, 2019). The development strategy implemented by Indonesia, known as the 'Rencana Pembangunan Jangka Menengah Nasional,' aims to strengthen the Indonesian economy by developing human capital and achieving higher global competitiveness levels (The World Bank, 2020). Jakarta is Indonesia's capital, as well as the country's largest city. The city of Jakarta is one of the world's most populated metropolitan areas. It is situated on Java's northwest coast. Strategically situated in the archipelago, Jakarta serves as a gateway to the entire country (Barbour-Lacey, 2015). In 2020, Jakarta had a population of more than 10 million (World Population Review, 2020). Nowadays, Jakarta is considered a global city and one of the world's fastest-growing economies. As an administrative centre in its own right and as an important industrial centre, Jakarta is a control centre for the national economy. Moreover, its location as a port makes it an important trading hub.
The Private Sector in Indonesia comprises the formal sector and the informal sector. The formal private sector comprises major corporate organisations, state-owned corporations, and investors from abroad. It is calculated that 54 per cent of the business that operates in Indonesia's private sector are primarily micro and small enterprises (Asian Development Bank, 2015). The report provided by Asian Development Bank (2015) also highlighted that effective implementation of the policies to improve the private sector is taking longer than was initially anticipated. In addition, there are questions about the efficiency and productivity of Indonesia's employees that have been raised. It is predicted that the next generation in Indonesia would only be 54 per cent as effective as she or he might have been with full health and complete education, according to the World Bank Human Capital Index (The World Bank, 2020). The low levels of human capital limit Indonesia's ability to switch into higher-value-added operations, raise productivity, and enhance household welfare. Leadership effectiveness is one of the essential factors that can play a critical role in enhancing employees' performance and productivity.

Problem Statement
There are several leadership styles today, and the transactional and transformational leadership styles have attracted interest from several scholars and researchers (e.g., (Dumdum et al., 2013;Odumeru and Ogbonna, 2013);. The transformational leadership style has often been touted among the various styles as one of the preferred styles. In this era of accelerating disruption, organisations are facing profound changes (McKinsey Quarterly). The current era needs transformational leaders who possess skills addressing complex problems and influencing followers to achieve performance beyond expectations by transforming their behavior (McKinsey Quarterly). The impact and outcomes of transformational leadership on employee performance have piqued and sparked many researchers and academics in this phenomenon. A transformational leader is expected to be a leader who inspires and drives change in a company and has a clear vision, imagination, and motivation to be highly effective (Mohammad et al., 2011). According to previous studies, transformational leadership positively impacts long-term employee success (Asrar-ul-Haq and Kuchinke, 2016;Jiang et al., 2017). Research has shown that a transformational leader has the skills to stimulate their followers and motivate them to go beyond their normal expectations (Brown et al., 2005) and foster a stronger sense of commitment and unity (Heneman and Judge, 2005) among organisations members to improve performance.
Despite the importance of leadership and its impact on business organisations' growth and sustainability, there is a scarcity of research on the effect of transactional and transformational leadership on the job performance of private-sector employees in Jakarta, Indonesia. Furthermore, organisational culture's role in mediating the relationship between leadership style and job performance has not been empirically investigated. As a result, this study is one of the few empirical studies in Indonesia's private sector to examine transactional and transformational leadership's impact on employee performance. This study will look at the impact of transformational and transactional leadership styles on employee success in their job performance. In addition, this study will also examine the mediating role of organisational culture. This study is expected to improve the understanding and adoption of the appropriate leadership style and behaviour on employees' performance.

Conceptualisation of Employee Performance
Job performance or employee performance have been used interchangeably, and it is one of the most important outcomes in organisations. Several authors specified the concept of employees' job performance and the associated parameters (Campbell, 1990;Moeheriono, 2010;Robbins, 2005;Viswesvaran and Ones, 2017). A comprehensive definition was stated by Viswesvaran and Ones (2017). Viswesvaran and Ones (2017), referred to job performance as a construct that exhibits employees' behaviours that can be evaluated. The behaviours encompass both visually observable and non-observable behaviours. They added that job performance refers to actions and behaviours and the outcomes employees engage in or produce. These outcomes are associated with and contribute towards the goals of the organisation. Similarly, Moeheriono (2012) referred to job performance as a reflection of the outcome or achievement that is linked with the implementation goals, vision, and mission by an organisation. Moeheriono (2012) defines performance as a working result and encompasses the dimensions of quality, efficiency, and effectiveness. Robbins (2005), referred to job performance as work-related outcomes or results contributed by an employee in carrying out their roles and responsibilities. This shows that job performance is a multiple-dimensional construct.
Researchers have paid attention to the performance dimensions and predictors of job performance (e.g., (Campbell, 1990;Leea and Donohue, 2012;Viswesvaran and Ones, 2017). As stated by Viswesvaran and Ones (2017), the concept of job performance is changing, and new dimensions are added. The dimensions include outcomes and behaviours. Viswesvaran and Ones (2017), argued that individual job performance refers to behaviour, but the outcomes between outcomes and behaviours are not clear. From a psychological perspective, Campbell (1990) stated that there is a clear demarcation between outcomes and behaviours. Campbell (1990), defined job performance in terms of eight dimensions. Job performance is also conceptualised as work-related behaviour and not about employee work outcomes (Aguinis, 2009). Motowidlo et al. (1997) indicated that job performance is the aggregate value of the discrete behavioural episodes that an individual performs over a normal time interval to organise them. A model was proposed by McCLoy et al. (1994), that specifies three performance determinants, namely declarative knowledge, procedural knowledge, and motivation. The model built by Campbell (1990) was evaluated by Leea and Donohue (2012), and the job efficiency model with six components was further amended. Job performance was further categorised into the task and contextual performance by Borman and Motowildo (1997). What constitutes output can be presumed to vary between workers. Therefore, job performance is a multidimensional construct, and a vast number of metrics exist as indicators of employees' performance or success.

Transactional Leadership and Job Performance
Transactional leadership is considered suitable to the moral values of employees. It generates the employee"s energy and resources and further increases their consciousness about ethical issues (Burns, 1978, as cited by Yukl and Gardner (2020). The employees feel motivated because the transactional leadership is appealing to their selfinterest and offer rewards and benefits. Transactional leadership is an exchange process that focuses on the exchanges between a leader and the employees (Northouse, 2013;Bass, 1985, as cited by Yukl andGardner (2020). In the process, the employees who follow the leader"s requests are rewarded. Therefore, transactional leadership styles and behaviours are focused on employees' contractual obligations and the associated rewards (Burns, 1978). House (1996) also stated that transactional leadership is dependent on the reward and performance of a contractual obligation by employees. As stated by Bass (1985, as cited by Yukl andGardner (2020), the types of transactional behaviours encompass contingent reward, passive management by exception. Another transactional behaviour referred to as active management by exception was added by Bass and Avolio (1990 as cited by Yukl and Gardner (2020). However, as stated by Bass (1985, as cited by Yukl andGardner (2020), transactional leadership behaviour does not generate commitment by employees. House (1996) further suggested that transformational leadership enhances employees' performance and should replace transactional leadership.
Transactional leadership is expected to influence employee performance based on the exchange process between the leader and the follower (Burns, 1978;House, 1996). Past studies have shown both positive and negative relationships between transactional leadership behaviours and employees' performance (Singh K., 2015;Sundi, 2013;Widayanti and Putranto, 2015). A study that was done by Singh K. (2015) among employees in the banking sector revealed that transactional leadership behaviours had a positive and significant impact on employee"s performance and productivity. The work environment is an essential antecedent of employee performance. The study by Widayanti and Putranto (2015) also revealed that transactional leadership had a significant and positive relationship with employee performance. Breevaart et al. (2014) explained that contingent rewards positively and directly impact employee performance by providing a safe and good working environment, autonomy, and support by leaders. Yang and Yang (2019) further added that rivalry and competition could influence the effectiveness of leadership. In environments where competition is moderate, transactional leadership will have a greater influence on employees' innovation and performance. On the contrary, a study by Lor and Hassan (2017) found that transactional leadership was not a positive and significant predictor of employees" performance. In addition, scholars and researchers have stated that there is no single leadership style that is effective in all situations (Lim and Ployhart, 2004). As suggested by Mahdinezhad and Suandi (2013), a leader's leadership style should fit the setting or situation in which the leader and the employees interact. Based on the literature review, it is posited that: H1: Transactional leadership exerts a positive influence on the performance of employees in Indonesia.

Transformational Leadership and Employee Performance
As Burns (1978, as cited in Yukl andGardner (2020) said, transformative leadership appeals to the employee's moral ideals and seeks to unlock their energy levels and resources. Bass (1985, as cited in Yukl andGardner (2020) later added that leaders demonstrate transformative leadership behaviours to get the employees' confidence, loyalty, and respect. This leadership style contributes to higher enthusiasm levels, and workers are motivated to move beyond what is expected. Employees become conscious of the value of task outcomes, sacrifice their self-interest for the company's sake, and activate their higher-order needs through transformative leadership behaviours. (Bass, 1985, as cited in Yukl and Gardner (2020). Northouse (2013) also added that transformational leadership contributes to employee success by going further and deeper than expected, unlike transactional leadership resulting in expected results. Bass (1985) stated that the three categories of transformational leadership behaviour cover idealised influence, intellectual stimulation, and individualised consideration. As the fourth aspect of transformational leadership behaviour. Bass and Avolio later introduced inspirational motivation.
Transformational leadership has gained prominence as the effective leadership style to be practised by leaders (Jiang et al., 2017;Wilkes et al., 2015). This leadership behaviour is expected to motivate and inspire employees to move beyond the expected goals and change their behaviour and beliefs . Past studies have found a positive and significant relationship between transformational leadership and employee performance (e.g., (Jiang et al., 2017;Singh R. and Rani, 2017;Wang et al., 2011). Wang et al. (2011), in their study, discovered that transformational leadership was a positive and essential predictor of contextual performance and task performance of employees. Singh R. and Rani (2017) found that all four dimensions of transformational leadership, which include inspirational motivation, intellectual stimulation, and individualised consideration and idealised influence, were positive and important predictors of employees' contextual performance and success of employees. Nemanich and Keller (2007) found similar findings when they looked at the effect of transformational leadership style on employee performance in multinationals. Studies by other researchers have also generally revealed a positive and significant relationship between transformational leadership style and employee performance (Asrar-ul-Haq and Kuchinke, 2016;Boehm et al., 2015;Jiang et al., 2017). The study by Boehm et al. (2015) revealed that both the transformational leadership behaviour and the leader"s charisma increased the organisational identity strength, and this subsequently leads to better job performance. However, some different results were also found. The study by Masa'deh et al. (2016) revealed a significant and positive impact of transformational and transactional leadership behaviours on employee performance. The positive impact of transactional leadership on employee performance was augmented or further strengthened by transformational leadership, according to a study by Rowold and Heinitz (2007). However, when it came to forecasting group or organisational efficiency, Banks et al. (2016) discovered that authentic leadership outperformed transformational leadership. The type of leadership behaviour may also be influenced by the situation that normally encompasses employees, the environment, and the leader's amount of authority. For instance, transformational leadership had a significant impact on results only in normal circumstances, according to Geier (2016), and transactional leadership was more influential in extreme situations. Geier (2016) further added that leaders' behaviour normally would adapt or change between normal contexts and extreme events. It is therefore hypothesised that: H2: Transformational leadership exerts a positive influence on the performance of employees in Indonesia.

Organization Culture as a Mediator
Culture is a multi-dimensional construct, and according to Schein (2010), culture is considered as a complex phenomenon that surrounds peoples at all times, and it is constantly generated and enacted by their experiences with other people and influenced by leadership behaviour. It is a collection of frameworks, structures, rituals, laws, and norms that direct and guides the behaviour of people. Schein (2010) further argued that culture defines leaders in situations where a group or team is successful, and the assumptions imposed by leaders are taken for granted. However, in difficult situations and where assumptions are no longer valid due to environmental changes, leadership needs to change due to the evolutionary change process. Therefore, as Schein (2010) stated, leaders need to understand the culture in which they are embedded in. This indicates that leaders in organisations can be the most powerful determinant of organisational culture. The dominant influence on individuals is the leadership style and culture. Therefore, leadership plays a critical role because the organisational culture is subject to changes and needs to be effectively managed and controlled (Kilmann et al., 1985).
The relationship between leadership styles, organisational culture and employee performance has attached interest from researchers, scholars, and practitioners (e.g., (Abbasi and Zamani-Miandashti, 2013;Ogbonna and Harris, 2000). Past studies have looked at the impact of leadership styles on organisational culture study and the effects and practices of leadership on organisational culture (e.g., (Hood, 2003;Kouzes and Posner, 2002). Scholars have highlighted the crucial role and association between leadership and organisational culture (Karada, 2009;Shiva and Suar, 2012). Shiva and Suar (2012) asserted that transformational leadership influences organisational culture, which further leads to organisations' effectiveness. Similarly, another study by Abbasi and Zamani-Miandashti (2013) revealed a significant and positive relationship between transformational leadership and organisational culture. Ilyas and Abdullah (2016) further added that organisational culture is directly related to employees' job satisfaction, and job satisfaction subsequently affects employees' performance. However, a study by Jung and Takeuchi (2010) revealed that community culture within the firm influences" leadership and subsequently, leadership leads to higher organisational performance. The study by Shahzad et al. (2012) also supported the positive impact of organisational culture on several employees and performance. Based on the support provided by past studies, it is believed that there is a relationship between leadership styles, organisational culture and employee performance.
Studies have also shown the mediating role of organisational culture between constructs (e.g., (Gorondutse and Abdullah, 2016;Xenikou and Simosi, 2006). A study by Chow (2012) supported the mediating role of organisational culture in the HR-performance relationship. Similarly, another study by Gorondutse and Abdullah (2016) revealed that organisational culture mediated the relationship between perceived ethics and performance. Organisational culture can also be a mediator between leadership styles and employee performance (Xenikou and Simosi, 2006). The study by Ogbonna and Harris (2000) found empirical evidence to support the relationship between leadership style and performance that was mediated by the existing organisational culture. Similarly, Xenikou and Simosi (2006) found that organisational culture mediates the relationship between transformational leadership and performance. Based on the review, it is believed that organisational culture will mediate the relationship between leadership styles and employee performance. It is therefore hypothesised that: H3: Organisational culture mediates the relationship between transformational leadership performance of employees in Indonesia.
H4: Organisational culture mediates the relationship between transformational leadership performance of employees in Indonesia.

Research Design
The explanatory study aimed to evaluate the correlation between transactional leadership, transformative leadership, and employee performance. Organisational culture as a mediator was included in this research. In this research, positivism was a more fitting philosophy since it relates to the natural scientist's philosophical position and measurements for the constructs were available to evaluate the hypothesised relationships of cause effects (Saunders et al., 2016). A deductive strategy was necessary, and quantitative analysis was carried out. This study used a crosssectional research and survey technique to gather numerical data using self-administered questionnaires. Bougie and Sekaran (2016) stated that perceptions could be measured using questionnaires. In addition, cross-sectional surveys are helpful in studying a phenomenon and relationships between the variables (Saunders et al., 2016). For data analysis, the Smart PLS tool was used together with the SPSS tool.

Target Population, Sampling, and Sample Size
The qualified individuals, activities, or items of interest that the researcher wants to examine is referred to as the target population . In a large-scale study, it would be impossible to obtain data from the entire population for this analysis. As a result, a representative sample of the target population was chosen to provide the required data for this study (Saunders et al., 2016). There are two types of sampling techniques: probability sampling (also known as representative sampling) and non-probability sampling (Saunders et al., 2016). Since a list of potential respondents was not readily available, convenience sampling was used instead. The sample size was calculated using Tabachnick and Fidell (2013). The required sample size was 82 that was calculated based on the formula "50 + 8m", where "m" is the number of factors. Hair et al. (2010) recommended that the hypothesised relationship testing based on structural equation modelling (SEM), the sample size should be between 100 to 500. The sample size would grow depending on the number of indicators, model complexity and measurement characteristics. For testing a model using SEM, Kline (2010) recommended sample size of 200. Bacon (1997) highlighted that SEM applications typically use 200 to 400 sample size to fit a model with a minimum of 10 to15 observed variables. The target sample size set for this study was set as 150 respondents.

Instrumentation
This questionnaire was divided into two parts. The first part consists of demographic questions on the respondents' background, particularly gender, marital status, and working experience. A filter question was inserted, and the respondent is asked if they work in a company in the private sector in Indonesia as a full-time employee. Section B of the questionnaire was designed to measure the constructs in this study. The questions were adapted from past studies. For the constructs, transactional leadership and transformational leadership, this research used the items from Ismail et al. (2010) to measure the transactional leadership among the employees from companies in Jakarta, Indonesia, by using five dimensions of the Likert scale. This research used the items adapted from the study by Abid Alvi et al. (2014) to measure organisational culture. For employee performance, this research adapted the questions from the study by Inuwa (2016). The interval scale (Likert type) was used to measure the respondents' response as it is easy to administer, and it is easy for the respondents to understand and answer the questions (Malhotra, 2009). The Likert type measurement was anchored by 1 = strongly disagree, 2 = disagree, 3 = neither, 4 = agree, 5 = highly agree.

Data Collection
There are three types of survey strategies that can be used to collect quantitative data: observation, structured interview, and self-administered questionnaire (Saunders et al., 2016). For this study, self-administered questionnaires were used because they allow data to be collected more quickly from a larger sample size and a wider geographical area (Saunders et al., 2016). Therefore, a self-administered questionnaire was the most appropriate technique for this research due to the restrictions and time constraints. A hybrid technique was used to maximise the response where the questionnaires were sent electronically and delivered by hand. The electronic distribution turned out to be the most appropriate strategy during the COVID-19 pandemic, as around 80% of the responses were received electronically. The first wave of questionnaires was collected within one month. Follow-up was done, and the second wave of questionnaires was collected with the next two months. A total of 169 questionnaires were received, and four questionnaires were rejected due to missing data. A total of 165 good questionnaires were used for further data analysis.

Data Analyses
Data from the questionnaire was edited and transferred to an Excel file. Thereafter, the IBM SPSS and Smart Pls statistical software were used to analyse the data. The IBM SPSS software was used to perform the descriptive analysis. The respondents' demographic profiles were obtained by using the IBM SPSS software that provided the frequency, percentage, and graphical presentations. The Smart PLS software tool was used to generate the inferential statistics, reliability testing and validating testing.
For reliability, validity, and hypothesis testing, the Smart Pls tool was used in this study. The degree of consistency of data as reflected in the measurement model in estimating the proposed latent construct is referred to as reliability (Awang, 2015). According to Hair et al. (2019), the Composite Reliability Index can be used to assess the reliability of a measurement model (CR). As a Rule of Thumb, the value obtained for the construct reliability must be at least 0.70 or greater (Hair et al., 2019). The Smart Pls method can be further used to assess convergent and discriminant reliability. According to Awang (2015), convergent validity is not violated if all of the measurement model's items are statistically important. The convergent validity was based on Average Variance Extracted (AVE) for each construct, and as a rule of thumb, AVE ought to be 0.5 or higher (Hair et al., 2019). Hair et al. (2019) recommended that a low but significant loading of 0.50 and below should be considered for deletion. Items with low factor loadings need to be removed to improve the validity, as retaining the low factor loading items causes the construct to fail convergent validity. The discriminant validity refers to the measurement model in which a construct is free from redundant items. The discriminant validity is the extent to which a construct is empirically distinct from other structural models' constructs. The Fornell and Larcker (1981) criteria were used to test discriminant validity. The Smart PLS software provided correlation coefficients and the significance level of the hypothesised structural relationships among the constructs based on the structural model. In this investigation, there were two direct hypotheses and two mediation hypotheses. Hypothesis H1 and H2 were tested for the cause-effect relationship, and hypothesis H3 and H4 were tested for mediation impact.

Respondents Demographics
There were 165 respondents in this study. The table shows the characteristics of the respondents. In this study, most were male respondents. There were 89 (53.9%) male respondents and 76 (46.1%) female respondents. In terms of age distribution, most of the respondents were in the 22 years to 34 years group. The first age group of employees was less than 21 years old, with 41 (24.8%) respondents. The second group of employees were between 22 and 34 and had 106 (64.2%) respondents. The last group was between 35 and older, and this group had 18 (10.9%) respondents. This indicates that most of the employees were young. Lastly, the length of the employees" working tenure in the private sector was divided into four groups. The first group consisted of employees that had working experience less than one year, and this group had 81 (49.1%) respondents. The second group consisted of employees with one to five years of working experience in the private sector, and this group had 73 (44.2%) respondents. The third group consisted of employees who had between six to ten years of experience, and there were 6 (5.5%) respondents in this group. Lastly, the fourth group consisted of employees who had more than ten years of working experience in the private sector, and there were 2 (1.2%) respondents in this group. The working experience distribution also shows that most of the respondents were young and had between one to five years of experience in the private sector.

Reliability
The results of reliability testing are presented below. The reliability in this study was the degree to which a variable is consistent with what it is supposed to measure (Hair et al., 2010). The reliability score based on established and proven measurements should be at least 0.70. (Hair et al., 2019). This is further supported by Sekaran and Bougie (2010), who stated that reliability is considered poor if the value is less than 0.6, appropriate if the value is 0.7, and excellent if the value is 0.8 or higher. The minimum suggested value by Sarstedt et al. (2014) for internal consistency reliability must have a value of at least 0.6 (Sarstedt et al., 2014). The value of rho for all constructs in this analysis was greater than 0.8. The Cronbach Alpha and composite reliability values were both above 0.8. This is good reliability and indicates that the reliability of data collected in this study was good.

Convergent Validity
In this study, convergent validity was established based on results obtained from the Smart Pls system. In this study, convergent validity refers to the extent to which the construct converges to explain the variance of its items (Hair et al., 2019). In this study, the Average Variance Extracted or commonly referred to as "AVE', was used to check the construct"s convergent validity. In this study, the mean value was computed based on each indicator's square loading on a construct. The Smart Pls system results showed the AVE value was 0.5 and above, and this is above the threshold value of 0.5 that was stated by Hair et al. (2019). All the factor loadings leadings were also above 0.5, and therefore none of the items was deleted. The results confirmed the convergent validity of the items in this study.

Discriminant Validity
In this study, the discriminant validity was tested. The discriminant validity showed the degree to which a construct is empirically distinct from other constructs. The objective of establishing discriminant validity is to make sure that the study's construct has the strongest relationship with the indicators of the construct. Fornell and Larcker (1981) suggested a traditional metric in which the AVE of each construct in the model is compared to the squared inter-construct correlation of the same construct and all other reflectively calculated constructs (Hair et al., 2019).
The findings of this analysis revealed that all of the indicators had the highest loadings on each of the constructs. As a result, discriminant validity was determined. The Fornell Larcker metric has been used in several studies to establish discriminant validity. However, according to Hair et al. (2019), it is not a satisfactory indicator of discriminant validity. The Heterotrait-Monotrait Ratio or commonly referred to as 'HTMT', is an alternative measurement that has been proposed. The mean value of item correlations across constructs compared to the mean of average correlations for items measuring the same construct is referred to as HTMT. In order to establish the discriminant validity in this study, the values should be less than 0.9 (Henseler et al., 2015). Since all of the HTMT values in this analysis were less than 0.9, discriminant validity was established.

Multicollinearity
As Hair et al. (2010) explained, multicollinearity refers to the extent to which the other variables in the analysis can explain a variable. A high value of multicollinearity can complicate the interpretation of the variate. The value of variance inflation factor (VIF) is used to test whether multicollinearity exists. Hair et al. (2019) stated that the VIF values of five and above indicate the existence of the multicollinearity issue. As a rule of thumb, VIF values should close to 3 and lower. In this study, all the VIF values were close to 3 or lower. This indicates the non-existence of multicollinearity issues.

Coefficient of Determination (R-square)
In this study, the measurement model was satisfactory. Collinearity was also not an issue. The next step is the assessment of the structural model. This can be based on the value of the coefficient of determination (R-square). The R2 value related to the first endogenous construct (organisation culture) was 0.74, and this indicates that 74% of the variance, which is explained in the endogenous construct, namely organisational culture. The r-square value of 0.41 indicates that 41% of the variance is explained in the endogenous construct, namely performance.

Significance and Relevance of Path Coefficients
Bootstrapping with a resample of 5000 was done to evaluate the significance of path coefficients and determine the mediator's mediation effects. To evaluate the significance, the t-value was examined. Hair et al. (2019) stated that the standard beta (β) values and the associated t-values that were generated after running the bootstrapping procedure need to be examined. A t-value that is higher than 1.96 or a p-value lower than 0.05 indicates a significant relationship. The first hypothesis, H1, examined the relationship between transactional leadership and employee performance. The standardised regression value of .116, t-value of 1.159 and a p-value of 0.246 indicate the impact is not significant. Therefore, hypothesis H1 was not supported. The second hypothesis, H2, examined the relationship between transformational leadership and employee performance. The standardised regression value of .551, t-value of 6.227 and p-value of 0.000 indicate the impact is significant. Therefore, hypothesis H2 was supported.
Next, the hypothesis H3 and H4 were tested. Hypothesis H3 was to find out whether organisational culture mediates the relationship between transactional leadership and employee performance. In step 1, the indirect effect (a×b) must be significant. The effect of transactional leadership on organization culture was positive and significant (β = 0.184 and p-value = 0.004). The effect of organisational culture on employee performance was also positive and significant ((β = 0.492 and p-value = 0.000). The total effect (a x b) was 0.090. In this case, X's direct effect on Y was reduced (β = 0.116 to 0.024), and the indirect effect (through M) was insignificant before and after mediation. This indicates only indirect mediation. Hypothesis H4 was to find out whether organisational culture mediates the relationship between transformational leadership and employee performance. In step 1, the indirect effect (a×b) must be significant. The effect of transformational leadership on organization culture was positive and significant (β = 0.551 and p-value = 0.000). The effect of organisational culture on employee performance was also positive and significant ((β = 0.492 and p-value = 0.000). The total effect (a x b) was 0.271. In this case, X's direct effect on Y was reduced after mediation (β = 0.551 to 0.197). The direct effect that was significant before mediation (p = 0.000) becomes insignificant after mediation (p = 0.118). The type of mediation here is called a "complete mediation" since the direct effect of X1 on Y is not significant after X2 entered the model. Instead, the indirect effect is significant. Thus, transformational leadership has an indirect effect on Y through the mediator variable organisational culture (Awang, 2012).

Discussion
The first hypothesis was to investigate the impact of the transactional relationship on employee performance in Indonesia's private sector. The direct effect of transactional leadership on employee performance was not significant (β = 0.116, t-value = 1.159 and p-value = 0.246). The results of this study deviated from past studies. The results are similar to a study by Lor and Hassan (2017) that also found that transactional leadership was not a positive and significant predictor of employees" performance. One explanation of this can be the respondents of this student were generally young, and the study was done during the Covid-19 pandemic. Therefore, as Mahdinezhad and Suandi (2013) suggested, the practised leadership style should fit the situation. During the Covid-19 crisis, transformational leadership may be the preferred leadership behaviour that is expected to motivate and inspire employees to move beyond the expected goals and change their behaviour and beliefs . Another explanation can be the environment. As revealed in the study by Wei et al. (2010), transactional leadership behaviour is positively related to employee"s creative performance in teams with a higher empowerment climate and vice versa. Yang and Yang (2019) also stated that interaction between transformational leadership and competitive intensity is positively related to firm performance in highly competitive environments. The private sector is expected to be highly competitive.
The second hypothesis was to examine the impact of transformational leadership on employee performance in Indonesia's private sector. The results showed that there was a strong and significant relationship between transformational leadership and employees" performance (β = 0.551, t-value = 6.227 and p-value = 0.000). Therefore, the second hypothesis was supported. This study's results were similar to the significant relationships shown in other studies (Jiang et al., 2017;Singh R. and Rani, 2017;Wang et al., 2011). In past studies, transformational leadership was a positive and significant predictor of employees' task and contextual performance (Wang et al., 2011). In past studies, the dimensions of transformational leadership also predict employee performance (Manesh et al., 2018). This shows that transformational leadership behaviour is essential in motivating employees and increasing their performance. Transformational leadership will lead to higher enthusiasm levels, and workers will be motivated to move beyond what is expected. This study's results are consistent with Northouse (2013), who stated that transformational leadership contributes to employee success by going further and deeper than what is expected, unlike transactional leadership that only leads to expected results. Transformational leadership can be effective during times of crisis and during normal times (Zhang et al., 2012). Therefore, leaders in the private sector should focus on transformational leadership behaviour to improve employee performance. Forward-thinking leaders should practice transformational leadership to increase the performance of both on-site and remote employees.
In this study, the third hypothesis (H3) was to examine whether organisational culture mediated the relationship between transactional leadership relationship and employee performance. The fourth hypothesis (H4) further examined the role of organisational culture as a mediating factor in the relationship between transformational leadership and employee performance. The findings revealed that organisational culture has a positive and significant indirect impact on the relationship between transformational leadership and employee success. Similarly, organisational culture had a significant and positive indirect impact on employee performance. The essential association or relationship between leadership behaviour and employee performance was highlighted in this study. These similar findings were also highlighted in other studies (Karada, 2009;Shiva and Suar, 2012). Abbasi and Zamani-Miandashti (2013) also found a significant and positive relationship between transformational leadership and organisational culture. Shahzad et al. (2012) also supported the positive impact of organisational culture on several employee"s performances. There is a close association between organisational culture and leadership, and as stated by Yukl and Gardner (2020), an organisation's culture is a situational influence on leaders. However, over time, leaders can also influence culture. The organisational culture that encompasses cultural values can improve the organisation's performance (Yukl and Gardner, 2020).

Implications
This study provided some good practical implications for leaders in the private sector, human resource managers and other organisations on the crucial role of transformational leadership behaviour and the essential role of organisational culture. This is consistent with McKinsey's report that stated that the current disruptive times need transformational leaders who possess the ability to address complex problems. Adapting the leadership behaviour in uncertain and rapidly changing environments requires innovation and strong leadership in sustain and remain competitive. Transformational leaders must develop the inner agility to resolve problems by being open-minded, flexible and adopt best practices (McKinsey Quarterly). The adoption of transformational leadership behaviour will result in higher leadership effectiveness and higher job performance levels. The adoption of transformational leadership by leaders in the private sector will further motivate employees to embrace change and work towards the achievement of the organisation's mission and goals. The transformational leader must develop a healthy and 'safe' work environment where employees can be innovative and productive. Organisations should support and invest in transformational leadership development programs to further inspire and motivate their employees to go further than expected. In this study, it was also found that organisational culture mediates the relationship between leadership styles and employee performance. It can also be construed that leadership is associated with organisational culture. Organisations must build a culture that encourages employees to change their attitude and mindset that is focused on going beyond expectations and working for the common good. Therefore, organisations need to focus on organisational culture to ensure that transformational leadership behaviour positively impacts employees' performance.
This research provided some new insights and added to the current body of knowledge from the theoretical perspective. Firstly, the impact of transactional and transformational leadership behaviours on employee performance was examined. The study found that transactional leadership has no significant impact on the performance of employees. Transformational leadership was found to have a significant and strong impact on employee performance. The findings support the transformational leadership model that was initially developed by Burns (1978) and later expanded by Bass (1985). Although the model was developed some time ago, this study shows that the model is still an effective indicator of today's leaders' behaviours or styles. The situation is also relevant, and this study shows that it is also vital in a crisis situation where the leadership style can make or break an organisation. This study also found the mediating effects of organisational culture were positive and strong. More specifically, organisational culture's mediating role was consistent with past studies (Gorondutse and Abdullah, 2016;Ogbonna and Harris, 2000).

Limitations and Future Research
There were some limitations associated with this study. Firstly, this study only examined transformational leadership and transactional leadership. There are other contemporary leadership styles such as ethical leadership, servant leadership and authentic leadership styles. Future studies should focus on other leadership styles. There are several culture topologies. For instance, the Organizational Culture Profile (OCP) is represented by seven distinct values (O'Reilly et al., 1991). This study did not look at the different dimensions of culture topologies such as aggressive culture. The inclusion of culture topologies in future studies can provide more in-depth information. The employee performance construct was measured based on the in-role perception and honesty of the respondents. It is recommended that future studies consider different sources of in-role performance, such as collecting data from supervisors. This study only looked at the performance of employees. It is recommended that future studies include other employee-related outcomes such as engagement and creativity.