Leadership in a school context: how leadership styles are associated with leadership outcomes

Purpose: This paper aims at examining how principal leadership styles are associated with leadership outcomes in an Indonesian school context. Research Methodology: Survey data were collected using MLQ 5X-Short and a demographic questionnaire completed by 475 teachers in 36 junior high schools in the Province of Lampung, Indonesia. With the help of SPSS version 22, the collected data were initially analyzed using descriptive statistics and then Pearson product-moment correlation. Results: Results show that transformational and transactional leadership styles are positively and significantly associated with the leadership outcomes, but the laissez-faire leadership style is negatively and significantly associated with the leadership outcomes. Limitations: This paper used a quantitative research approach and is acknowledged for being limited to using the survey questionnaire so that it could not explain why and how principal leadership styles affect such leadership outcomes. Contribution: This paper theoretically and practically contributes to the body of knowledge, particularly concerning leadership styles and their outcomes. Keywords: School leadership, Leadership outcomes, Principal, teacher


Introduction
The leadership literature in the education sector is enormous, in part, because education is essential for the development of human resources that will lead to economic growth of countries (e.g., Hariri, Monypenny, & Prideaux, 2012;Hourani & Stringer, 2015;Nguni, Sleegers, & Denessen, 2006;Sofo, Fitzgerald, & Jawas, 2012;Storey, 2004). Thus, the various drivers of school principal leadership styles in association with school leadership outcomes become a vital basis of schools and economic growth plans of many governments (Hariri et al., 2012). As shown by Hariri, Monypenny, and Prideaux (2014), for school effectiveness and students' future success, effective school leadership should be in place. The success of schools, as well as students are significantly influenced by school principals. (Gurr, Drysdale, & Mulford, 2005).
The principal capability substantially influences school leadership for adopting more collaborative leadership styles and involving others in making collaborative decision (e.g., Barnett & McCormick, 2003;Ng, Wang, & Liu, 2015;Pashiardis, 1993;Steyn, 2015;Williams, 2008). That is why principalship is said to be a key to economic growth plans of many governments (e.g., Hallinger, 2011;Hallinger & Walker, 2011). To some extent, development of leadership has been much researched and underpinned school policy activities (Mertkan, 2011, p. 79). Thus, principals should be very seriously prepared (Siegrist, 1999). In fact, it is so important that as Thomas and Kearney (2010, p. 10) assert that, until recently, effective principalship has been found to be a key factor in student achievement and school improvement. However, there are few studies about leadership behaviours (Gemeda & Lee, 2020;Little, Gooty, & Williams, 2016;Vullinghs, De Hoogh, Den Hartog, & Boon, 2018), and little is known about studies of school leadership in the Asian context (Raihani, 2008;Wong & Wong, 2005). Hence, this paper helps close a part of the wide gaps of leadership research in the Asian context, in addition to the other three published papers resulted from the research project. This paper, Paper 4, is about Leadership in a School Context: How Leadership Styles Are Associated with Leadership Outcomes. The other three previous papers were published in 2012, 2014, and 2016 respectively. Unlike the three previous papers, Paper 4 includes leadership outcomes (extra effort, effectiveness, and satisfaction), in addition to leadership styles. Paper 1 examined whether teacher job satisfaction could be significantly predicted by decision-making styles (Hariri et al., 2012); whereas, Paper 2 investigated leadership styles in relation to decision-making styles (Hariri et al., 2014), and Paper 3 examined the congruency of data from Indonesian schools with the Anglophile and Western literature concerning principal leadership styles, decision-making styles and teacher job satisfaction (Hariri, Monypenny, & Prideaux, 2016).
The whole project that produced these papers included a significant review of the literature that produced the following framework for the whole project: First, that there is a large but mainly Anglocentric and Western literature on the topic of the project. Second, that within this literature, relationships exist between: (1) laissez-faire, transactional, and transformational leadership styles, (2) avoidant, spontaneous, dependent, intuitive, and rational decision-making styles, (3) nine job satisfaction facets (communication, contingent rewards, co-workers, fringe benefits, nature of work, operation conditions, pay, promotion, and supervision), (4) four demographic variables (sex, teacher certification, school location, and marital status). Third, that these relationships may range from highly positive to highly negative. Fourth, that given the mainly Anglocentric Western literature on the topic of the project, the significant issue for the project was to get good data for Indonesia to underpin the conclusions for each of the papers. Thus, this project breaks new academic ground by analysing the Indonesian context instead of the more Anglocentric focus of other studies.
For the whole research project, to randomly select prospective participants from public junior high schools in Lampung Province, sampling with three stages was employed (Hariri et al., 2016). Stage 1, that is, primary units, random sampling started with geographic districts, where six districts (Lampung Tengah, Lampung Selatan, Pesawaran, Pringsewu and Tanggamus, and Bandar Lampung) out of the 15 districts were chosen. Stage 2, that is, secondary units, where 37 schools were selected in the six districts. In Stage 3, that is, tertiary units, 518 out of 11,401 teachers and 37 principals (Kemdiknas, 2009) was chosen from the schools previously selected. Based on 95% level of confidence, such size of the sample is considered greater than the recommended size of sample (Gray, 2004).
There were four questionnaires used for the whole Project, these were the three standard survey questioners: the Multifactor Leadership Questionnaire (MLQ) Form 5X-Short , the General Decision-Making Style (GDMS) questionnaire (Scott & Bruce, 1995), and the Job Satisfaction Survey (JSS) (Spector, 1985), and a self-designed demographic questionnaire. The demographic questionnaire was particularly developed for the whole project and is employed to dedepict demographics of the participants: sex, marital status, school location, teacher certification, and tenure with current principal. Data collected through these questionnaires are teacher reported data and actions of principals are perceived actions interpreted by teachers.
Within the above project framework, this paper investigates associations between leadership styles exhibited by the principals (laissez-faire, transactional, and transformational) and teacher-perceived leadership outcomes (extra effort, satisfaction, and effectiveness) in the context of Lampung public junior high schools, Indonesia. Three study questions were formulated to guide this paper: (1) What is the association between transformational leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness)? (2) What is the association between transactional leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness), (3) What is the association between laissez-faire leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness) in the context of Lampung schools, Indonesian?
What is meant, in this paper, by the leadership styles as well as leadership outcomes is that the principals exhibit their leadership styles, whereas the teachers perceive the leadership styles exhibited by the principals as being laissez-faire, transactional, or transformational. This manuscript investigates whether leadership styles (exhibited by the principals) as perceived by teachers are associated with the extra effort, satisfaction, and effectiveness. As respondents, the teachers rated the principal leadership styles and leadership outcomes by completing the leadership questionnaire (MLQ).
For this paper, it is assumed that: (1) laissez-faire leadership style is negatively and significantly associated with leadership outcomes (extra effort, satisfaction, effectiveness) in an Indonesian school context. (2) Transactional leadership style is positively and siginificantly associated with leadership outcomes (extra effort, effectiveness, satisfaction) in an Indonesian school context. (3) Transformational leadership style is positively and siginificantly associated with leadership outcomes (extra effort, effectiveness, satisfaction) in an Indonesian school context.

Leadership styles
The MLQ comprises three leadership styles and three leadership outcomes . This paper focuses on the associations between these two components. In this paper, school leadership aspect of the leadership literature in the education sector interests the author; this paper uses leadership styles as defined by Bass (1997) and studies on school principal leadership styles (laissez-faire, transactional, and transformational leadership styles) (Bass, 1997;Northouse, 2007) by means of the MLQ 5X-Short .
Regarding leadership styles as Hariri et al. (2014) asserted, a priori findings of this paper would expect to support Bass' augmentation effect theory (Bass, 1985(Bass, , 1999 as well as Judge and Piccolo's findings (Judge & Piccolo, 2004, p. 755). Hence, when school leaders need to remain effective and to attain leadership outcomes, they need to become more transformational and transactional, but to elude laissezfaire. Bass (1999) claimed that transactional leadership effectiveness is strengthened by transformational leadership. Autonomy and challenging work are nurtured by transformational leadership. Transformational leadership has made followers' satisfaction become increasingly vital. On the contrary, laissez-faire leadership greatly leads to followers' conflict, ineffectiveness, and dissatisfaction. The finding indicates that transformational leadership enhances the effect from transformational leadership better than that from transactional leadership (Judge & Piccolo, 2004, p. 755). A priori, it would be expected that despite being not mutually exclusive, transformational leadership and transactional leadership are distinct (Judge & Piccolo, 2004).

Leadership outcomes
This paper is about the teacher-perceived school principal leadership outcomes (extra effort, , satisfaction, effectiveness), all are aspects of the leadership literature in the education sector. The term 'leadership outcomes' is a wide topic; for example, the response to leadership outcomes could be by a range of stakeholders, for example by students, by teachers, or by external stakeholders; furthermore, the outcome could be in a range of attributes, for example in student learning, in teacher development, in the school structure, or in community stakeholders. However, the standard questionnaire used in this paper (the MLQ 5X-Short) ) focuses on extra effort, satisfaction, and effectiveness. However, all of the studies that have been found were carried out in a non-school and in a non-Indonesian context (e.g., Alloubani, Abdelhafiz, Abughalyun, Edris, & Almukhtar, 2015; Amirul This paper focuses on teacher-perceived leadership outcomes (extra effort, satisfaction, effectiveness) in an Indonesian school context. A priori, regarding leadership outcomes, it would be expected that the findings suggest significant associations between the three leadership styles and leadership outcomes.

Research methodology
The data for this paper came from the MLQ 5X-Short  plus the demographic questionnaire. The MLQ questionnaire measures principal leadership styles and leadership outcomes; the questionnaire was completed by the teachers. The principals exhibited these variables, but this paper examines these variables according to the perceptions of the teachers (n=475), rather than the perceptions of the principals. In MLQ, there are only the three leadership outcomes. Leadership outcomes are what teachers perceive based on their feeling of the impacts of teacher-perceived principal leaderships styles. Participants rated principals on a five-point Likert rating scale ranging from not at all to frequently, if not always, with scores ranging from 0 to 4 .
Regarding scoring, the scores of the MLQ scale are mean scores for the scale items. The score is derived by calculation as follows: sum the items and divide by the number of items making up the scale. In case of an item being left blank, the score is obtained using the total for that scale which is divided by the number of items answered. The number of scales is as follows: (1) nine scales measuring three leadership styles: laissez-faire (one scale consisting of four items), transactional leadership (three scales consisting of 12 items), and transformational leadership (five scales consisting of 20 items), and (2) three scales measuring outcomes of leadership (three scales consisting of: extra effort with three items, satisfaction with two items, and effectiveness with four items). (Bass & Avolio, 2004, p. 108).
Findings of studies suggest the MLQ Form 5X-Short consistency. For example, Antonakis, Avolio, and Sivasubramaniam (2003) found the nine factors best representing the factor structure underlying the MLQ Form 5X-Short instrument. In similar line, earlier research findings carried out by Avolio, Bass, and Jung (1999), revealed that the questiomnnaire was higly consistent in estimating factor loadings, interrelationships among factors, and reliability. Recently, Haj and Jubran (2016) examined the reliability of the instrument for its internal consistency resulting a Cronbach's α of 0.96.
More notably, Bass and Avolio (2004, p. 79) examined the MLQ nine-factor model which included 2003 nornative samples and used Confirmatory Factor Analysis (CFA). They revealed conclusive findings that the model showed the best fit with a clear consistency pattern across the corresponding results by participant and by region. The nine factor model comprises: 1) II (A) or Idealised Influence (Attribute), 2) II (B) or Idealised Influence (Behaviour), 3) IM or Inspirational Motivation, 4) IS or Intellectual Stimulation, 5) IC or Individualised Consideration, 6) CR or Contingent Reward, 7) MBEA or Management-by-Exception (Active), 8) MBEP or Management-by-Exception (Passive), and 9) LF (Laissez-Faire).
The MLQ had high reliabilities which exceeded standard cut-offs for internal consistency suggested in the literature. Each leadership factor scale and the total items had reliabilities ranging from 0.74 to 0.94. (Bass & Avolio, 2004, p. 48). Alpha (α) scores ranging from 0.60 to 0.70 suggest adequate reliabilities (Yellen, Cella, Webster, Blendowski, & Kaplan, 1997). Muenjohn and Armstrong (2008, pp. 9-10) supported the acceptable reliabilities of the instrument, where they reported the Thai MLQ version with a Cronbach's α of 0.87) and the original one with a Cronbach's α of 0.86 as well as its acceptable construct validity. Their findings indicate successful in sufficiently capturing the factor constructs of the full leadership approach. Consequently, researchers are confident to use the instrument to assess the factors that represent the leadership styles and outcomes.
However, a different evidence is from Japan. Fukushige and Spicer (2011) carried out a comparative study using the MLQ Form 5X-Short (Bass & Avolio, 2000). They compared leadership preferences of British and Japanese followers. The Japanese sample comprised 59 females and 207 males, while the British one comprised 28 females and 97 males. From the 36 items of the full-range leadership, they used 32 items, excluding the four items of laissez-faire leadership style. They removed the four items due to not receiving support in their previous research (see Aya & David, 2007). The MLQ in the Japanese version was addressed to the Japanese repondents, while the English version was addressed to the British respondents They also modified the original Likert rating scale, that is, they employed the scale ranging from 1 (strongly disagree) to 5 (strongly agree).
They used Cronbach's α to explore Bass and Avolio's leadership scales (Bass & Avolio, 2000). Their study for the Japanese sample revealed low reliability for four scales (IIA, CR, MBEA, and MBEP). These factors did not fulfil the adequate scores of Cronbach's α (0.6). This indicates that, for the Japanese case, that the findings do not support Bass' claim of construct universality of the instrument despite being for four scales (Bass, 1996). However, for the British case, the number of the scales which did not meet the acceptable scores of Cronbach's (0.6) was only two scales (IIA and IS) (Price & Mueller, 1986). For comparation, the result of independent samples t-test used to test leadership scales between the two different samples revealed different preferences in followers' leadership between Japan and the UK. In terms of universality, however, unlike transactional leadership, transformational leadership was found to be more effective and active across a variety of situations including in schools due to transformational ledership being found as universal style of leadership (Hamdani, 2018). This paper employed Cronbach's α and item-total correlations to check the internal consistency for the instrument. It met acceptable scores of Cronbach's α (exceeding 0.7) with the values of mostly above 0.3 for the item-total correlations.

Results of descriptive statistics
Descriptive statistics are used for both sample demographics and variables. Results of descriptive statistics for demographics of the sample of 475 teachers who completed the survey are described here.
In terms of sex balance, 56.4% of respondents were female teachers and 43.6% were male teachers. This is consistent with the reality that female teachers dominate teaching in Indonesia. But today men continue to outnumber women particularly in principal positions similar to the trends across APEC countries (Craig, Kraft, & Du Plessis, 1998, p. 5). The surveyed teachers have been mostly: married (94.9%), between 40 and 49 years old (52.8%), had been awarded an undergraduate degree (76.4%) and working with current principals for 1-2 years. It is assumed that this tenure with the principal is the shortest time for the teachers to be able to effectively identify principals' behaviors. This paper used descriptive statistics to depict the attributes of the variables including mean (M), standard deviation (SD), skewness, and kurtosis (see Table 1.  Table 2 shows the variables used in this paper: principal leadership styles and leadership outcomes (extra effort, satisfaction, and effectiveness) as perceived by teachers with the scores of mean, standard deviation, skewness, and kurtosis. The means of principal leadership styles as perceived by teachers in this paper from the highest to the lowest, ranging from 0 to 4, are as as follows: mean of transformational was the highest (M = 2.51, SD = 0.56), followed by mean of transactional (M = 2.04, SD = 0.49) and mean of laissez-faire (M = 0.82), SD = 0.67) . As teachers percieved, assessed using the MLQ 5X-Short, and analyzed using descriptive statistics, the leadership styles exhibited by the principals were all the three leadership styles. Nevertheless, the principals most frequently demonstrated transformational leadership style, followed by transactional leadership style, and scarcely displayed laissez-faire leadership style.
The descrptive statistics have been presented, then assumptions of Pearson correlation need to be checked before employing the correaltion.

Assumptions for Pearson correlation
Assumptions for Pearson correlation used for the data analysis, as for other parametric tests, are initially examined to ensure the data to be parametric (Field, 2009, p. 132). These examined assumptions as classified by Pallant (2007) are: continuous measures, random sampling, normality, observation independence, and variance homogeneity. All such assumptions were confirmed during the processing of the data for this paper. The assumption of continuous measures was executed by rescaling the qualitative data obtained from the survey into quantitative data, following the method of Bass and Avolio (2004). The random sampling assumption was confirmed throughout the phase of research design. The normality assumption was examined using skewness and kurtosis in which the data were normally distributed (both skewness and kurtosis values ranging from the +1 to -1 range), a much more stringent criterion than the +2 to -2 range criterion as suggested by Garson (2010). Last, the assumption of homogeneity of variances was tested using Lavene's test by doing ANOVA. Table 2 shows equal variances for all the respondents (F (1, 473) ≥ 0.05, NS). The leadership outcomes in this study are three: extra efforts, satisfaction, and effectiveness. Effectiveness had the highest mean (M = 2.61, SD = 0.98). Satisfaction (M = 2.58, SD = 0.95) had a slightly lower mean than effectivness, followed by mean extra efforts (M = 2.48, SD = 0.80). The leadership styles (laissez-faire, transactional, and transformational) had relationships with the leadership outcomes. Once the descriptive analysis was completed, Pearson product-moment correlation by means of SPSS version 22 was applied to answer the research questions as described below.

Results of Pearson Correlation
Relationship between the principal leadership styles and leadership outcomes are examined in this paper. It is essential to highlight that, based on the descriptive statistics analysis previously described, the school leaders displayed all the three leadership styles, where transformational leadership style was most frequently displyed, folloewed by transactional leadership style, but laissez-faire leadership style was hardly displayed. These results are consistent with previous studies, see for example Hariri et al. (2014). Table 3 shows the relationships resulted from the data analysis using Pearson product-moment correlation by means of SPSS version 22. Laissez-faire leadership -0.327*** -0.228*** 1 4 Extra effort 0.782*** 0.670**** -0.296*** 1 5 Effectiveness 0.702*** 0.649*** -0.304*** 0.773*** 1 6 Satisfaction 0.704*** 0.643*** -0.344*** 0.756*** 0.742*** 1 *p < 0.05, **p < 0.01, ***p < 0.001 Source: Data analysis of this paper The results shown in Table 3 are the answers to the study questions investigated in this paper. Below are the results and discussion of each research question (RQ) in detail.

Result of RQ 1
RQ1 is "What is the association between transformational leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness)?" Table 3 shows the answer to the first research question.
Result shows that transformational leadership style was positively and significantly associated with the three leadership outcomes: extra effort (r = 0.782, p < 0.001), satisfaction (r = 0.704, p < 0.001), and effectiveness (r = 0.702, p < 0.001). These results suggest that the principal transformational leadership style has the highest association with extra effort, followed by teacher satisfaction with leader, and leader effectiveness respectively.

Result of RQ 2
RQ 2 is "What is the association between transactional leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness)?" Table 3 shows results as the answer to the second research question. Result shows that transactional leadership style was positively and significantly associated with the three leadership outcomes: extra effort (r = 0.670, p < 0.001), satisfaction (r =0.643, p < 0.001), and effectiveness (r = 0.649, p < 0.001). These results suggest that the principal transactional leadership style has the highest relationship with teacher extra effort, followed by teacher effectiveness, and teacher satisfaction respectively.

Result of RQ 3
RQ 3 is "What is the association between laissez-faire leadership style and leadership outcomes (extra effort, satisfaction, and effectiveness) in the context of Lampung schools, Indonesian?" Table 3 shows results as the answer to the third research question. Result shows that laissez-faire leadership style was negatively and significantly associated with with the three leadership outcomes: extra effort (r = -0.296, p < 0.001), satisfaction (r = -0.344, p < 0.001), and effectiveness (r = -0.304, p < 0.001). These results suggest that the principal laissez-faire leadership styles have the highest but negative relationship with teacher extra effort of the teachers, followed by teacher effectiveness, and teacher satisfaction respectively.

Discussion
In recent years, the studies of leadership outcomes have explored a range of attributes connected to outcomes and to a range of stakeholders' perceptions of the impact of leadership actions and policy (see, for example, Alloubani et al., 2015;Amirul & Daud, 2012;Bennett, 2009;Nordin, 2013). However, all of the studies were found to be carried out in organizations other than schools and in countries other than an Indonesian context. Therefore, this paper focuses on teacher-perceived leadership outcomes (extra effort, satisfaction, and effectiveness) in relation to leadership styles as measured using the MLQ 5X-Short developed by Bass and Avolio (2004) in Lampung schools in Indonesia.
As early outlined, this paper studied three research questions. Findings show that: (1) transformational leadership style was positively and significantly associated with the three leadership outcomes: extra effort, satisfaction, and effectiveness, (2) transactional leadership style was positively and significantly associated with the three leadership outcomes: extra effort , satisfaction, and effectiveness, but (3) laissez-faire leadership style was negatively and significantly associated with with the three leadership outcomes: extra effort, satisfaction, and effectiveness.
These findings essentially contribute to the literature as well as to the development of the school system of Indonesia. More specifically, these findings provide empirical and theoretical contributions to the body of knowledge concerning leadership theories in ways, particularly, by determining the extent to which leadership outcomes are influenced by leadership styles. There are previous research results relevant to this paper. These research results describe these three leadership styles: (1) laissez-faire, (2) transactional, and (3) transformational in relation to leadership outcomes (extra effort, satisfaction, and effectiveness).
In general, the findings of this paper support those of earlier studies (e.g., Alloubani et al., 2015;Amirul & Daud, 2012;Casida & Parker, 2011;Nordin, 2013) focusing on how leadership styles are associated with leadership outcomes measured by means of the MLQ 5X-Short. It could be said that the findings are conclusive that is, either transformational leadership or transactional leadership is positively associated with such leadership outcomes as extra effort, satisfaction, and effectiveness; nevertheless, laissez-faire leadership is negatively associated with the three leadership outcomes. Thus, this ype of leadership needs to be avoided due to its negative impacts on the ledership outcomes. It has an attribute of "passive-avoidant behaviour toward addressing challenges" (Ritchie et al., 2020, p. 8).
In particular, Casida and Parker (2011, p. 481) found that, in addition to being positively and significantly associated with the leadership outcomes (extra effort, effectiveness, and satisfaction), transformational leadership style is also the best predictor for these outcome variables. Moreover, transformational leadership style best predicts teacher job satisfaction, out of the three leadership styles (Hariri et al., 2016). Besides, transformational leadership style has the strongest correlation amongst all leadership styles with leadership outcomes (Alloubani et al., 2015, p. 294). The findings of this paper are consistent with the earlier ones. In recent research, it was found that the leadership outcomes (satisfaction, extra-effort, and effectiveness) are also associated with emotional intelligence in transformational leadership (Biswas & Rahman, 2017). In particular, Jenner (2019) revealed that followers' satsfaction was greater with the leaders they perceived as transformational leaders rather than transactional leaders. Transformational leaders including principals also can make followers enthusiastic and optimistic (Jyoti & Bhau, 2016) so that the followers can be highly motivated to do extra efforts, to work effectively, and to get satisfaction.
In terms of the associations between the other two leadership styles (transactional as well as laissezfaire) and the three leadership outcomes, little is known in earlier studies, particularly in the education context. Therefore, these findings provide empirical and theoretical contribution to the body of knowledge, particulalrly, concerning leadership theories by determining the extent to which leadership styles influence leadership outcomes.
It is finally expected that school leadership leads to effective leadership because effective leadership will positively contibute to making organizations including schools to become and remain competitive (Nandedkar & Brown, 2018). Thus, it is recommended that principals exhibit transformational leadership to make them able to lead schools in rapid changes and encourage teachers to be creative and innovative (Albugami, 2020). In addition, Litz and Scott (2017) support this thought by saying that increased pressure in schools for changes through school reform needs principals to be transfomational.

Conclusion
Findings for an Indonesian school context are that the three leadership styles, as perceived by teachers, are significantly associated with the leadership outcomes (extra effort, satisfaction, and effectiveness). Either transformational or transactional leaderhip style is positively and significantly associated with the leadership outcomes, but laissez-faire leadership style is negatively and significantly associated with the leadership outcomes. Therefore, principals are recommended to exhibit transformational leadership style as well as transactional leadership style but to avoid laissez-faire leadership style when working with teachers.
These findings are congruent with what was expected from the literature review of the mainly Anglocentric Western literature on leadership styles and leadership outcomes. However, little is known in prior studies, particularly in the education context, concerning relationships between transactional as well as laissez-faire leadership styles and the three leadership outcomes.
Thus, the study of these associations examined in this paper will provide an important contribution to the body of knowledge, particulalry, school leadership including leadership outcomes and particularly to the development of the school system in Indonesia. .

Limitation and study forward
This paper used quantitative research approach and is acknowledged for being limited to using the survey questionnaire so that it could not explain why and how principal leadership styles affect such leadership outcomes as extra effort, satisfaction, and effectiveness. However, the limitation is an opportunity for researchers to include in-depth interview for further research.