Multi-level modeling of principal authenticity and teachers' trust and engagement.

Wang, Chuang ; Bird, James J.

INTRODUCTION

Teachers play a significant role in education (Easton-Brooks & Davis, 2009; Gere, Buehler, Dallavis, & Haviland, 2009; McKown & Weinstein, 2007; Stronge, Ward, Tucker, & Hindman, 2007). Therefore, it is of utmost importance to provide a healthy work condition for the teachers. As the leader of a school, principal's behaviors and management skills have an impact on teachers' job satisfaction, efficacy, trust in the organization, and engagement levels (Bird, Wang, Watson, & Murray, 2009; Bogler, 2001; Dipaola & Tschannen-Moran, 2001; Hipp & Bredeson, 1995; Koh, Steers, & Terborg, 1995). Most of the previous studies on these topics are in the business management field and rarely linked the three constructs of principal authenticity, teacher trust and teacher engagement (Branson, 2007; Connell, Ferres, & Travaglione, 2003; George, Sims, McLean, & Mayer, 2007). Very few studies have considered employee's trust and engagement and the employer's leadership style simultaneously (Yammarino, Dionne, Schriesheim, & Dansereau, 2008). The purpose of this study is to investigate the relationships between principal's authenticity and teacher's trust and engagement using a multi-level approach. The use of a multi-level approach allows the researcher to examine relationships among variables within schools as well as between schools (Goddard, Tschannen-Moran, & Hoy, 2001; Stapleton, 2006).

Authentic Leadership Style

Authentic leadership style is usually understood as being true to oneself (Harter, 2002) and has been researched extensively recently in management literatures (Gardner & Schermerhorn, 2004; George et al., 2007; Luthans & Avolio, 2003). Drawing upon Luthans and Avolio's (2003) initial definition of authentic leadership that focused on self-awareness and self-regulated behaviors as well as Gardner, Avolio, Luthans, May, and Walumbwa's (2005) emphasis on the self-regulation components of authentic leadership (i.e., internalized regulation, balanced processing of information, relational transparency, and authentic behavior), Walumbwa, Avolio, Gardner, Wernsing, and Peterson (2008) defined authentic leadership as "a pattern of leader behavior that draws upon and promotes both positive psychological capacities and a positive ethical climate, fosters greater self-awareness, an internalized moral perspective, balanced processing of information, and relational transparency on the part of leaders working with followers, fostering positive self-development." (p. 6). This concept of authentic leadership in educational settings, however, is relatively new although some scholars have investigated principal's leadership behaviors, teacher's trust in the schools, as well as school climate and their impacts on student learning outcomes (Bird et al., 2009; Begley, 2001; Branson, 2007; Lambert, 2002; Marzano, Waters, & McNulty, 2005).

Trust

Trust is a critical component of school improvement and effectiveness (Tschannen-Moran, & Hoy, 2001). Mayer, Davis, and Schoorman (1995) defined trust as "the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other party will perform a particular action important to the trustor, irrespective of the ability to monitor or control that party" (p. 712). Scholars in the past paid attention to trust as an important factor that influences employees' well-being, retention, and the success of the institution (Connell et al., 2003; Kramer & Tyler, 1996; Mayer & Davis, 1999; Shaw, 1997). In education, Tschannen-Moran and Hoy (2000) reviewed a wide variety of theoretical articles and empirical studies on trust as it related to relationships within schools and concluded that trust included such constructs as willing vulnerability, benevolence, reliability, competence, honesty, and openness. Tschannen-Moran and Hoy (2000) further concluded that trust is required for many of the reforms in American schools and collaborative decision-making and teacher empowerment depend upon trust.

Engagement

The concept of engagement was defined as employees' commitment and the positive emotions they experience (Ostrem & Wheeler, 2006). Levels of engagement were found to be positively related to success, rapport, sense of belonging, and trust in the organization (Buckingham & Coffman, 1999; Fredrickson, 1998; Ostren & Wheeler, 2006). In educational settings, Dipaola and Tschennan-Moran (2001) noted a strong positive relationship between organizational citizenship behavior and school climate and posit that trust in the principals and trust in the schools are essential for teachers to engage in their teaching and to help students improve their academic learning outcomes. Another study in education noted that teachers' participation in the decision-making process will increase teachers' sense of belonging, commitment to the school, and self-efficacy beliefs (Bogler & Somech, 2004). A more recent study by Hoy, Hoy, and Kurz (2008) found that teachers' engagement was associated with their dispositional optimism, classroom management, and efficacy for classroom success.

This study is unique in that it combines previous literature in business management and education fields and examines the three constructs (authenticity, trust, and engagement) simultaneously by considering the feature that teachers' trust and engagement are nested within principal's authentic leadership. Specifically, the research questions addressed by this study are as follows:

* What is the measurement model for each of the surveys used: authenticity, trust, and engagement?

* How well is the principal's self-report of authenticity related to their teacher's report of their principal's authenticity?

* Are there between-school variances of these measurement models? If so, what are the relationships between principals' authenticity and teachers' trust and engagement levels within and between schools?

METHOD

Participants

A total of 83 principals and 1240 teachers from 83 public schools in three county school districts of a southeastern state completed the surveys. For the sake of statistical analyses and the concern of representativeness of the target population, all teacher and principal data where there were fewer than five teacher participants were removed from the multi-level modeling process. Due to the limitations of mean imputation method, list-wise deletion was used to remove all cases with missing values after an examination of the assumption that missing is completely at random. This resulted in a sample of 60 principals (72% of the original sample) and 917 teachers (74%). Of the 917 teachers, 798 (87%) were female and 119 (13%) were male. These teachers were predominantly (90%) Caucasian (n = 825) with 55 (6%) African American, 18 (2%) Hispanic, and 18 (2%) reporting other ethnic background. This demographic information matches that of the teacher population at the school districts (80% female and 20% male; 89% Caucasian, 8% African American, 2% Hispanic, and 1% other). Of the 917 teachers who reported their highest education level achieved, 537 (59%) had Bachelor's degrees, 358 (39%) had Master's degrees, 9 (1%) had Educational Specialist degrees, and 7 (1%) had Doctorate degrees. Six teachers did not report this information. Teachers' teaching experience (M = 13.36, SD = 9.21) ranged from 1 to 42 years, number of years working in the current school (M = 6.16, SD = 5.67) ranged from 1 to 32 years, and number of years working under the current principal (M = 3.08, SD = 2.02) ranged from 1 to 17 years. Teacher information of educational background and experience was not available at the school district level, however, the variance of these variables suggest that the respondents were diverse in educational background and experience. Therefore, we had no concern of a possible bias of voluntary selection of a particular group.

Of the 60 school principals, 36 (60%) were female and 24 (40%) were male. These principals were also predominantly (90%) Caucasian (n = 54) with six (10%) being African American. The distribution of these principals' highest education level was 45 (75%) Master's degrees, 10 (17%) Educational Specialist degrees, and 5 (8%) Doctorate degrees. Seven of these principals 11 (18%) were of high schools, 14 (23%) were of middle schools, and 35 (59%) were of elementary schools. Their teaching experience (M = 13.10, SD = 6.91) ranged from 3 to 33 years, number of years in administration (M = 10.86, SD = 6.37) ranged from 3 to 30 years, and number of years working as the principal at the current school (M = 3.97, SD = 4.07) ranged from 0 to 25 years.

Instruments

Authentic leadership questionnaire

To measure authentic leadership style, Walumbwa, Avolio, Gardner, Wernsing, and Peterson (2008) developed a 16-item questionniare (Appendix A) which consists of four subscales: (a) self-awareness, (b) relational transparency, (c) internalized moral reasoning, and (d) balanced processing. The questionnaire has two forms. The first is for the organization's leader to self-report their own levels of authenticity. The second form is for the organization's employees to report their leader's level of authenticity. Self-awareness is measured by four items that reflect the extent to which leaders are aware of their strengths and limitations and how others perceive them. Relational transparency is measured by five items that reflect the extent to which leaders reinforce a level of openness with others. Internalized moral reasoning is measured by four items that reflect the extent to which leaders set high standards for moral and ethical conduct. Balanced processing is measured by three items that reflect the extent to which leaders solicit sufficient opinions and viewpoints of others prior to making important decisions. The original version of the questionnaire was designed for raters to assess their leaders. The internal reliability for each sub-scale is as follows: self-awareness, .92; relational transparency, .87; internalized moral perspective, .76; and, balanced processing, .81. Content validity of the 16 items was established through discussions of faculty members and a group of doctoral students, and the construct validity was confirmed with a sample of 224 participants in the United States and another sample of 212 participants in the People's Republic of China. The comparative fit index ranges was .97 for the U.S. sample and .95 for the Chinese sample. The root mean square error of approximation was .05 for the U.S. sample and .06 for the Chinese sample. Predictive validity of the authentic leadership questionnaire was checked by correlating each of the four sub-scales to variables such as ethical leadership, organizational citizenship behavior, organizational commitment, and satisfaction with supervisor. All reliability and validity information in this paragraph is from the work of Walumbwa et al. (2008). Principal participants in this study completed the self-report version of this questionnaire at a principal's meeting while teacher participants completed the rater version on-line. Participants were asked to rate the frequency of each statement that fits the leadership style using a 5-point scale ranging from 1 (not at all) to 5 (frequently, if not always).

Workplace trust survey

Ferris and Travaglione (2003) developed a 32-item survey (Appendix B) that consists of three subscales: (a) trust of the supervisor (9 items), (b) trust of co-workers (12 items), and (c) trust of the organization (14 items). The reliability for the subscales is .96 for trust of the supervisor, .93 for trust of co-workers, and .95 for trust of the organization (Ferris & Travaglione). Concurrent validity was established by correlating the subscale of trust in the organization with Cook and Wall's (1980) subscale of trust in management (r = .91) and correlating the subscale of trust of co-workers with Cook and Wall's (1980) subscale of trust in peers (r = .93). Trust of the supervisor was also significantly related to trust in co-workers (r = .52) and trust in the organization (r = .65). Discriminant validity was also established by finding no significant relationships between these three subscales and age, gender, position level, and tenure. Teacher participants in this study completed the survey on-line to report their trust in their principals, colleagues, and the schools where they were working. A 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used.

Gallup organization's Q12 survey

Buckingham and Coffman (1999) developed 12 items (Appendix C) to measure employee engagement as a result of thousands of focus groups and interviews. The reliability for this survey was .88. Concorrent validity of the 12 items were established by correlating the sum of the 12 items with customer satisfaction, profitablity, and productivity, respectivley (Harter, Schmidt, & Keyes, 2002). Teacher participants in this study completed the survey on-line to report their engagement levels. A 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used.

Procedures

Box-plots were used to check outliers for each variable, and residuals of the multivariate models were checked as well. No outlier was found in the data set. Descriptive statistics and internal consistency of the instruments were examined to explore and examine the constructs measured before inferential statistical procedures. Construct validity of the instruments was checked with confirmatory factor analysis (CFA). Both hierarchial linear modeling (HLM) and stucture equation modeling (SEM) were used to examine the relationships between teachers' trust and engagement levels with their ratings of principals' authentic leadership skills.

RESULTS

All the instruments used in this study were reliable: the internal consistency ranged from .76 to .93 for principal's self-report of authenticity; from .81 to .96 for teacher's report of their principal's authenticity; from .93 to .97 for teacher's trust, and .86 for teacher's engagement (Tables 1 and 2). Means and standard deviations of all the first-order and second-order constructs were also reported in Tables 1 and 2, respectively. The means for all the constructs were close to 4 out of 5, which suggested that, on average, the principal and teachers rated favorably on their authenticity, trust, and engagement.

Comparisons of the mean scores and standard deviations of the first-order and second-order constructs of authenticity between principal's self-report and teachers' ratings also revealed that principals rated themselves significantly higher than what their teachers rated them, F (5, 54) = 4.19, p =.001; [[eta].sup.2] = .14; and that the distribution of principal's self-ratings had much higher values of leptokurtosis than teacher ratings, which means that the variance for the principal's self-ratings was much smaller than that for the teachers' ratings. On average, principals rated themselves 4.64 whereas teachers rated principals 4.02 out of 5. Follow-up multiple comparisons with Tukey's Honestly Significant Difference (HSD) method showed that principal's rated themselves significantly higher than teachers rated them on all first-order and second-order constructs of authenticity (p values less than .01).

The correlation coefficients among all first-order constructs of principal authenticity were highly correlated to each other with a minimum value of .63 for principal's self-report and .78 for teacher's rating of their principal's authenticity. Moreover, all first-order constructs of authenticity were highly related to the second-order construct of authenticity (the minimum correlation coefficient was .82 for principal's self-report and .91 for teachers' rating of their principal's authenticity). These high coefficient values suggest that the measurement model for authenticity is second order with all first-order constructs loaded to the second-order construct of authenticity. Similar patterns were found in Table 2 about the relationships among the first-order and second-order constructs of trust.

Table 1 also showed that the relationships between teachers' rating of their principal's authenticity was not statistically significantly related to the principal's self-report of authenticity for the general construct of authenticity (r = .12), self-awareness (r = .11), relational transparency (r = .08), internalized moral reasoning (r = .21), or balanced processing (r = .07). A further analysis with multi-level structure equation modeling would be needed to understand this phenomenon.

The goodness-of-fit indices of the measurement models for principal's self-report of authenticity, teachers' ratings of their principal's authenticity, engagement, and trust were presented in Table 3.

We evaluated the goodness of the fit of the models using various fit indices reported in previous studies, including the [chi square] statistic, normed fit index (NFI), non-normed fit index (NNFI), comparative fit index (CFI) goodness of fit index (GFI); Standardized Root Mean Square Residual (SRMR); and Root Mean Square Error of Approximation (RMSEA). Although Hu and Bentler's (1999) joint criteria for absolute and incremental models (i.e., NNFI [greater than or equal to] .96, CFI [greater than or equal to] .96, and SRMR [less than or equal to] .09) have been widely used to judge model fit, they have been criticized for too restrictive assumptions and rejecting adequately fit models (Marsh, Hau, & Wen, 2004). In addition, Fan and Sivo (2005) questioned the validity of the two-index strategy in model fit assessment presented by Hu and Bentler (1999). Therefore, Hu and Bentler's (1999) criteria were used with caution in this study. The suggestions provided by LISREL to add paths from observable variables to latent variables or to add error covariance between observable variables were not followed because of the concern of mechanically fitting the model (MacCallum, Roznowski, & Necowitz, 1992).

The measurement models for teacher's trust (Figure 1) and engagement (Figure 2) were confirmed. The authenticity measurement model was confirmed at the teacher-level (Figure 3; teachers' ratings of their principal's authenticity) but not at the principal level (principal's self-report of authenticity). Although the ratio of the chi-square value to the degree of freedom for the principal's self-report of authenticity was relatively lower than those for other models, the sample size for the principal's self-report is 60 whereas that for teacher's ratings was 917. Therefore, comparison of the ratio between chi-square values between principal's self-report and teacher ratings was not appropriate.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Estimates of intra-class correlation (ICC), a ratio of the univariate between-school variance over the total variance, are presented in Table 4 for all items, first-order and second-order constructs for all surveys.

Table 4 shows that around 14% of the variance in teachers' ratings of the principal's authentic leadership, around 15% of the variance in teachers' ratings of trust in their principals, and around 6% of the variance in teachers' ratings of engagement existed between schools. As a result, the nested feature of the data was considered in multi-level structure equation modeling and HLM analyses. Please note that the first-order construct of trust in their principals, instead of the second order trust, was used in the structure model because there was only about 7% of variance existing between schools for trust in their colleagues and we believe that trust in their principals is a more meaningful construct related to their principal's authenticity.

The multi-level structure equation model was found to be satisfactory: [chi square] =24457.40, df = 1707, NFI = .96, NNFI = .97, CFI = 97, GFI = .53, SRMR = .09, and RMSEA = .12. The 90% confidence interval for RMSEA was from .119 to .120. Figure 4 shows the structure equation model of the three major constructs (authenticity, trust, and engagement) between and within schools, respectively.

[FIGURE 4 OMITTED]

Both the between-school and within-school levels of the structure equation model suggest significant positive relationships between principal's authenticity and teacher's trust in their principals and strength of engagement at their schools. That result suggests that the more truthful and open the principals are, the more their teachers trust their principals and the stronger sense of belonging and engagement teachers have.

To compare the moderating effect on the relationships between authenticity, trust, and engagement between principal's self-report of authenticity and teachers' ratings of their principal's authenticity, HLM was used. Two school-level variables were used: principal's self-report of authenticity and teacher's ratings of their principal's authenticity. Since the sample size is relatively large at both levels, maximum likelihood estimation method was employed because this method of estimation was shown to have desirable properties when the sample size is large. These properties include unbiased fixed effects, unbiased estimates of the variance parameters, asymptotically efficient estimates of both fixed effects and variance parameters, and approximately normal distributions of the estimates (Raudenbush & Bryk, 2002). Group-mean centering was used so that the intercept at Level 1 represents the adjusted mean for a teacher's report of their principal's authenticity when his/her levels of trust and engagement are both at the mean of the teacher's school. A two-step strategy for the conditional model was used (Speece, Ritchey, Cooper, Roth, & Schatschneider, 2004). First, simple conditional models were run to examine each variable individually. Then, the variables significant at the first step were examined simultaneously (complete conditional model) at a significance level of p < .05. Estimates of the HLM were presented in Table 5.

The simple conditional models implied in Table 5 suggest that principal's self-report of authenticity had no statistically significant impact on the adjusted mean score of teachers' ratings of principal's authenticity or the relationship between this construct and teacher's level of trust or engagement. The average of teacher's ratings of their principal's authenticity, however, had a negative impact on the relationships between teacher's trust and principal's authenticity and between teacher's engagements with principal's authenticity, respectively. The higher the teachers rated their principal's authenticity, the weaker the strength of these relationships. When teachers rated their principal's authenticity low, they were more than likely to have lower levels of trust and lower levels of engagement. The complete conditional model revealed that teachers' level of engagement was no longer statistically related to their ratings of principal's authenticity when their trust in their principal was controlled. That is, for teachers who had the same level of trust in their principal, their levels of engagement were not related to their ratings of the principal's authenticity anymore. Magnitude of effect, or proportion of the variance explained by the conditional models in comparison to the unconditional model, was 70.75%. This magnitude of effect suggests that the variables that we put in the complete conditional model explained 70.75% of the variance with respect to principal's authenticity rated by their teachers (Raudenbush & Bryk, 2002). The comparison between the HLM model that has both trust and engagement as predictors and the HLM model that has only trust as the predictor suggested that adding engagement as an additional predictor to trust increased the magnitude of effect by only 0.01%.

DISCUSSION

The measurement models for the key constructs examined in this study (authenticity, trust, and engagement) provided further validity information of the instruments developed by Walumbwa et al., (2008); Ferris and Travaglione (2003), and Buckingham and Coffman (1999), respectively. The relatively low fit indices of the principal's self-report of authenticity measurement model, however, suggested that the data do not support the construct structure of this instrument in self-report form and cautioned further users of the self-report form of authenticity instrument. This is not surprising because the literature suggests that other ratings are more trust-worthy than self-ratings (Atkins & Wood, 2002; Atwater & Yammarino, 1992; Furnham & Stringfield, 1994). To our knowledge, this is the first study that examined the construct validity of the self-report form of the authenticity instrument (Walumbwa, Wang, Wang, Schaubroeck, & Avolio, 2010).

The finding that principal authenticity is positively and significantly related to teacher levels of trust and engagement is consistent with previous studies in business (Ostrem &Wheeler 2006) and in education (Bird et al., 2009). It seems reasonable that teachers would respond positively to the four components of authentic leadership: strong self-awareness, relational transparency, balanced processing, and moral integrity. A school principal is faced with many varied and complex situations on a daily basis. Coping with this triage-like working condition requires a certain degree of courage, confidence, and consistency of performance. This self-efficacy engenders trust amongst faculty and sets examples for teachers to follow in the pursuit of their responsibilities.

Principals with strong relational transparency patterns of behavior develop deep, open, and meaningful relationships with their staff members. These principals are concerned about the further professional development of their subordinates and are supportive of their career advancement and future success. Teachers appreciate this notion of "we're in this together" sense of teamwork and it portends to strengthen their levels of trust and engagement (Dipaola & Tschennan-Moran, 2001; Tschannen-Moran, & Hoy, 2000).

Decision-making and problem-solving responsibilities are pervasive in the principalship. Faculties appreciate a leader who approaches these tasks with equanimity, free from bias and preconceived notions of outcomes. Principals, who seek input and advice from diverse sources of information and then rationally sifts the data according to rubrics reflecting established organizational goals rather than personal leanings, garner trust from their staff. "Letting the chips fall where they may" means everyone has an equal chance for consideration rather than having cliques and favoritism rule the day. Such patterned behavior generates confidence and consistency with operational matters thereby creating clarity of purpose throughout the school.

Finally, principals who demonstrate moral integrity display a strong sense of right from wrong discernment. This is appreciated by teachers who deal daily with conflicting demands from parents and students. Principals, who are able to effectively and efficiently frame issues in light of "what's best for kids" and to judge accordingly towards outcomes that reflect a client-based enterprise, win the trust of their faculty members (Bogler & Somech, 2004; Hoy et al., 2008). Teachers who know with clarity the parameters and the perimeters of what is acceptable and that which is not, are able to complete their responsibilities with confidence and dispatch.

Implications for Future Research and Practice

The focus of this study was on the relationship between school building principals and their respective faculties. Two different approaches were used to examine these relationships: SEM and HLM. While SEM suggested that both trust and engagement were positively related to authenticity, HLM revealed that only trust was statistically significantly related to authenticity. One possible reason is the multicolliearity between the independent variables in HLM (Iacobucci, Saldanha, & Deng, 2007) since these two constructs are highly correlated (r = .79). Another possible reason is that engagement is not related to teacher's ratings of their principal's authenticity when their levels of trust to their principal are the same. The structure equation model for the between-schools and within-schools show that the relationship between trust and authenticity was stronger than that between engagement and authenticity. The HLM approach showed that adding engagement as an additional predictor of authenticity had a minimal impact on the improvement of the model (the magnitude of effect increased by 0.01%). The same approaches could be applied to other leader-follower relationships within educational settings. Such pairings as superintendents and principals; boards of education and superintendents; deans and department chairs; or, chancellors and deans have the potential of extending our understandings of the value of authentic leader behavior and its relationship to follower behavior.

Another arena of research exploration could be the extension of the leader-follower pairing to the inclusion of outcomes. Do students benefit from authentic leaders and trusting, engaged teachers? Is there a relationship with student performance?

Since having trusting and engaged employees is generally seen as being a desirable set of working conditions, leader authenticity should also be sought, developed, and rewarded in practice. Hiring boards and departments of human resources should seek candidates who demonstrate the components of authenticity: self-awareness, relational transparency, balanced processing, and moral integrity. These patterns of behavior can be observed in their conduct during the interview process; can be queried through actual questioning; and, can be documented through reference checking with past work settings.

University preparation programs and professional development enterprises can include the concept of leader authenticity in their course content. Adding the study of authenticity to their knowledge domains would improve the chances that their clients will become aware of the positive relationships between leader behavior and follower response in this regard. Creating simulation exercises that would include opportunities for instructive role-playing would allow participants to practice authentic behavior in guided situations. Teaching methods of gathering and analyzing data from subordinates concerning their perceptions of leader authenticity would be informative for those aspiring to become future leaders.

Leader supervisors should include authentic behavior as a desirable strategy and reward its implementation. Explicit feedback during goal-setting, formative, and summative evaluation sessions would reveal the importance of authenticity to the leader being assessed. Tying the demonstration of authentic behavior to compensation programs would directly bind desired behavior to expected practice. Appendix A. Authentic Leadership Questionnaire (Self) Instructions: The following survey items refer to your leadership style, as you perceive it. Please judge how frequently each statement fits your leadership style using the following scale: 0 1 2 3 4 Not at all Once in a while Sometimes Fairly often Frequently, if not always As a leader I... 1. say exactly what I mean 0 1 2 3 4 2. admit mistakes when they are made 0 1 2 3 4 3. encourage everyone to speak their mind 0 1 2 3 4 4. tell you the hard truth 0 1 2 3 4 5. display emotions exactly in line with feelings 0 1 2 3 4 6. demonstrate beliefs that are consistent with actions 0 1 2 3 4 7. make decisions based on my core values 0 1 2 3 4 8. ask you to take positions that support your core values 0 1 2 3 4 9. make difficult decisions based on high standards of ethical conduct 0 1 2 3 4 10. solicit views that challenge my deeply held positions 0 1 2 3 4 11. analyze relevant data before coming to a decision 0 1 2 3 4 12. listen carefully to different points of view before coming to conclusions 0 1 2 3 4 13. seek feedback to improve interactions with others 0 1 2 3 4 14. accurately describe how others view my capabilities 0 1 2 3 4 15. know when it is time to reevaluate my position on important issues 0 1 2 3 4 16. show I understand how specific actions impact others 0 1 2 3 4 Copyright [c] 2007 Authentic Leadership Questionnaire (ALQ) by Bruce J. Avolio, William L. Gardner, & Fred O. Walumbwa. All rights reserved in all medium. Distributed by Mind Garden, Inc. www.mindgarden.com. Appendix A: Authentic Leadership Questionnaire (Teacher Form) Instructions: The following survey items refer to your leader's style, as you perceive it. Judge how frequently each statement fits his or her leadership style using the following scale: 0 1 2 3 4 Not at all Once in a while Sometimes Fairly often Frequently, if not always My leader... Q8. says exactly what he or she means 0 1 2 3 4 Q9. admits mistakes when they are made 0 1 2 3 4 Q10. encourages everyone to speak their mind 0 1 2 3 4 Q11. tells you the hard truth 0 1 2 3 4 Q12. displays emotions exactly in line with feelings 0 1 2 3 4 Q13. demonstrates beliefs that are consistent with actions 0 1 2 3 4 Q14. makes decisions based on his or her core values 0 1 2 3 4 Q15. asks you to take positions that support your core values 0 1 2 3 4 Q16. makes difficult decisions based on high standards of ethical conduct 0 1 2 3 4 Q17. solicits views that challenge his or her deeply held positions 0 1 2 3 4 Q18. analyzes relevant data before coming to a decision 0 1 2 3 4 Q19. listens carefully to different points of view before coming to conclusions 0 1 2 3 4 Q20. seeks feedback to improve interactions with others 0 1 2 3 4 Q21. accurately describes how others view his or her capabilities 0 1 2 3 4 Q22. knows when it is time to reevaluate his or her positions on important issues 0 1 2 3 4 Q23. shows he or she understands how specific actions impact others 0 1 2 3 4 Copyright [c] 2007 Authentic Leadership Questionnaire (ALQ) by Bruce J. Avolio, William L. Gardner, & Fred O. Walumbwa. All rights reserved in all medium. Distributed by Mind Garden, Inc. www.mindgarden.com Appendix B. Trust Survey Please judge how frequently each statement fits your level of trust in your leadership using the following scale: 0 1 2 3 4 Strongly disagree Disagree Neutral Agree Strongly agree Q24. I feel that information can be shared openly within my school. 1 2 3 4 5 Q25. I think that processes within my school are fair. 1 2 3 4 5 Q26. I honestly express my opinion at my school with the knowledge that employee views are valued. 1 2 3 4 5 Q27. There is a widely held belief that my school is moving forward for the better. 1 2 3 4 5 Q28. I feel that my principal keeps personal discussions confidential. 1 2 3 4 5 Q29. I think that my co-workers act reliably from one moment to the next. 1 2 3 4 5 Q30. I perform knowing that my organization will recognize my work. 1 2 3 4 5 Q31. I feel that my principal trusts his/her employees to work without excessive supervision. 1 2 3 4 5 Q32. I think that my school offers a supportive environment. 1 2 3 4 5 Q33. I have positive feelings about the future direction of my school. 1 2 3 4 5 Q34. Employees at my school generally feel that coworkers appreciate their quality performance. 1 2 3 4 5 Q35. I believe that my school recognizes and rewards employees' skills and abilities. 1 2 3 4 5 Q36. I feel that my principal listens to what I have to say. 1 2 3 4 5 Q37. Most employees at my school believe that co-workers will be supportive if problems arise. 1 2 3 4 5 Q38. I feel that my co-workers are truthful in their dealings with me. 1 2 3 4 5 Q39. I proceed on the basis that my principal will act in good faith. 1 2 3 4 5 Q40. I behave on the basis that my coworkers will not disclose personal information. 1 2 3 4 5 Q41. I think that my principal appreciates additional efforts I make. 1 2 3 4 5 Q42. I act on the basis that my school follows plans with action. 1 2 3 4 5 Q43. I feel confident that my coworkers appreciate my good work. 1 2 3 4 5 Q44. I act knowing that my principal will keep his/her word. 1 2 3 4 5 Q45. Most employees at my school believe that co-workers are reliable. 1 2 3 4 5 Q46. I believe that my principal follows words through with action. 1 2 3 4 5 Q47. I proceed with the knowledge that my co-workers are considerate of my interests. 1 2 3 4 5 Q48. I will act on the foundation that my co-workers display ethical behavior. 1 2 3 4 5 Q49. Employees commonly believe that they are treated fairly at my school. 1 2 3 4 5 Q50. I believe that my co-workers support me if I have problems. 1 2 3 4 5 Q51. I feel that my principal is available when needed. 1 2 3 4 5 Q52. I feel that I can trust my co-workers to do their jobs well. 1 2 3 4 5 Q53. I believe that my co-workers give me all the information to assist me at work. 1 2 3 4 5 Q54. It is generally accepted that my school takes care of employee interests. 1 2 3 4 5 Q55. I act on the basis that my principal displays integrity in his/her actions. 1 2 3 4 5 Appendix C. Engagement Survey Instructions: Please judge how frequently each statement fits your level of engagement at work using the following scale: 0 1 2 3 4 Strongly disagree Disagree Neutral Agree Strongly agree 57. Do you know what is expected of you at work? 1 2 3 4 5 58. Do you have the materials and equipment you need to do your work right? 1 2 3 4 5 59. At work, do you have the opportunity to do what you do best every day? 1 2 3 4 5 60. In the last seven days, have you received recognition or praise for doing good work? 1 2 3 4 5 61. Does your supervisor, or someone at work, seem to care about you as a person? 1 2 3 4 5 62. Is there someone at work who encourages your development? 1 2 3 4 5 63. At work, do your opinions seem to count? 1 2 3 4 5 64. Does the mission/purpose of your company make you feel your job is important? 1 2 3 4 5 65. Are your associates (fellow employees) committed to doing quality work? 1 2 3 4 5 66. Do you have a best friend at work? 1 2 3 4 5 67. In the last six months, has someone at work talked to you about your progress? 1 2 3 4 5 68. In the last year, have you had opportunities at work to learn and grow? 1 2 3 4 5

REFERENCES

Atkins, P. W. V., & Wood, R. E. (2002). Self- versus others' ratings as predictors of assessment center ratings: Validation evidence for 30-degree feedback programs. Personal Psychology 55, 871-904.

Atwater, L. A., & Yammarino, F. J. (1992). Does self-other agreement on leadership perceptions moderate the validity of leadership and performance predictions? Personnel Psychology, 45, 141-164.

Begley, P. (2001). In pursuit of authentic leadership practices. International Journal of Leadership in Education, 4, 353-365.

Bird, J. J., Wang, C., Watson, J., & Murray, L. (2009). Relationships among principal authentic leadership and teacher trust and engagement levels. Journal of School Leadership, 19, 153-171.

Bogler, R. (2001). The influence of leadership style on teacher job satisfaction. Educational Administration Quarterly, 37, 662-683.

Bogler, R., & Somech, A. (2004). Influence of teacher empowerment on teachers' organizational commitment, professional commitment and organizational citizenship behavior in schools. Teaching and Teacher Education, 20(2), 277-289.

Branson, C. (2007). Effects of structured self-reflection on the development of authentic leadership practices among Queensland primary school principals. Educational Management Administration & Leadership, 35(2), 225246.

Buckingham, M., & Coffman, C. (1999). First break all the rules: What the world's greatest managers do differently. New York: Simon & Schuster.

Connell, J., Ferres, N., & Travaglione, T. (2003). Trust in the workplace: The importance of interpersonal and organizational support. Journal of Management Research, 3(3), 113-118.

Cook, J., & Wall, T. D. (1980). New work attitude measures of trust, organizational commitment and personal need non-fulfillment. Journal of Occupational Psychology, 53, 39-52.

Dipaola, M. & Tschannen-Moran, M. (2001). Organizational citizenship behavior in schools and its relationship to school climate. Journal of School Leadership, 11, 424-447.

Easton-Brooks, D. & Davis, A. (2009). Teacher qualification and the achievement gap in early primary grades. Education Policy Analysis Archives, 17(10), 1-19.

Fan, X., & Sivo, S. A. (2005). Sensitivity of fit indices to misspecified structural or measurement model components: Rationale of two-index strategy revisited. Structural Equation Modeling, 12, 343-367.

Ferres, N., & Travaglione, T. (2003, December). The development and validation of the workplace trust survey: Combining qualitative and quantitative methodologies. Paper presented at the 9th Asia Pacific Researchers in Organization Studies Conference, Oaxaca, Mexico.

Fredrickson, B. (1998). What good are positive emotions? Review of General Psychology, 3, 300-319.

Furnham, A. & Stringfield, P. (1994). Congruence of self- and subordinate ratings of managerial practices as correlate of supervisor evaluation. Journal of Occupational and Organizational Psychology, 67, 57-67.

Gardner, W., Avolio, B., Luthans, F., May, D., & Walumbwa, F. (2005). "Can you see the real me?" A self-based model of authentic leader and follower development. The Leadership Quarterly, 16, 343-372.

Gardner, W., & Schermerhorn, J., Jr. (2004). Unleashing individual potential: Performance gains through positive organizational behavior and authentic leadership. Organizational Dynamics, 33, 270-281.

George, B., Sims, P., McLean, A., & Mayer, D. (2007). Discovering your authentic leadership. Harvard Business Review, 85, 129-138.

Gere, A. R., Buehler, J., Dallavis, C., Haviland, V. S. (2009). A visibility project: Learning to see how preservice teachers take up culturally responsive pedagogy. American Educational Research Journal, 46(3), 816-852.

Goddard, R. D., Tschannen-Moran, M., & Hoy, W. K. (2001). A multilevel examination of the distribution and effects of teacher trust in students and parents in urban elementary schools. The Elementary School Journal, 102(1), 3-17.

Harter, J. K., Schmidt, F. L., & Keyes, C. L. (2002). Well-being in the workplace and its relationship to business outcomes: A review of the Gallup studies. In C. L. Keyes & J. Haidt (Eds.), Flourishing: The positive person and the good life (pp. 205-224). Washington D. C.: American psychological Association.

Hipp, K. A., & Bredeson, P. V. (1995). Exploring connections between teacher efficacy and principal's behavior. Journal of School Leadership, 5(2), 136-150.

Hoy, A. W., Hoy, W. K., & Kurz, N. M. (2008). Teachers' academic optimism: The development and test of a new construct. Teaching and Teacher Education, 24(4), 821-835.

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1-55.

Iacobucci, D., Saldanha, N., & Deng, X. (2007). A meditation on mediation: Evidence that structural equation models perform better than regressions.

Ilies, R., Morgeson, F., & Nahrgang, J. (2005). Authentic leadership and eudaemonic well-being: Understanding leader-follower outcomes. The Leadership Quarterly, 16, 373-394.

Jensen, S., & Luthans, F. (2006). Relationship between entrepreneurs' psychological capital and their authentic leadership. Journal of Managerial Issues, 18, 254-273.

Johnson, S. M., & Birkeland, S. E. (2003). The schools that teachers choose. Educational Leadership, 60(8), 20-24.

Kahn, W. (1990). Psychological conditions of personal engagement and disengagement at work. Academy of Management Review, 33, 692-724.

Kaplowitz, M. D., Hadlock, T. D., & Levine, R. (2004). A comparison of web and mail survey response rates, Public Opinion Quarterly, 68, 94-101.

Kenny, D. A., & Judd, C. M. (1986). Consequences of violating the independence assumption in analysis of variance. Psychological Bulletin, 99, 422-431.

Koh, W. L., Steers, R. M., & Terbog, J. R. (1995). The effects of transformational leadership on teacher attitudes and student performance in Singapore. Journal of Organizational Behavior, 16, 319-333.

Kramer, R. M., & Tyler, T. R. (1996). Trust in organization: Frontiers of theory and research. London: Sage.

Kreft, I., & de Leeuw, J. (1998). Introducing multilevel modeling. Newbury Park, CA: Sage.

Lambert, R. (2002). Evaluating management climate in Head Start programs: The measurement properties of the policy and program management inventory. NHSA Dialog, 6, 37-52.

Luthans, F., & Avolio, B. (2003). Authentic leadership development. In K. Cameron, J. Dutton, & R. Quinn (Eds.), Positive organizational scholarship: Foundations of a new discipline (pp. 241-258). San Francisco: Berrett-Koehler.

MacCallum, R. C., Roznowski, M., & Necowitz, L. B. (1992). Model modifications in covariance structure analysis: The problem of capitalization on chance. Psychological Bulletin, 111, 490-504.

Marsh, H. W., Hau, K. T., & Wen, Z. (2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler's (1999) findings. Structural Equation Modeling, 11, 320-341.

Marzano, R., Waters, T., & McNulty, B. (2005). School leadership that works: From research to results. Alexandria, VA: Association for Curriculum and Supervision.

May, D., Chan, A., Hodges, T., & Avolio, B. (2003). Developing the moral component of authentic leadership. Organizational Dynamics, 32, 247-260.

Mayer, R. C., & Davis, J. H. (1999). The effect of the performance appraisal system on trust for management: A field quasi-experiment. Journal of Applied Psychology, 84, 123-136.

Mayer, R., Davis, J., & Schoorman, F. (1995). An integrative model of organizational trust. Academy of Management Review, 20, 709-734.

McKown, C., & Weinstein, R.S. (2007). Teacher expectations, classroom context, and the achievement gap. Journal of School Psychology, 46, 235-261.

Michie, S., & Gooty, J. (2005). Values, emotions, and authenticity: Will the real leader please stand up? The Leadership Quarterly, 16, 441-457.

Ostrem, L., & Wheeler, D. (2006, October) Engagement and trust as outcomes of servant leadership. Paper presented at the Gallup Leadership Summit, Washington, DC.

Palmer, N., & Fleig-Palmer, M. (2006, October). Emergent adult's authentic leadership development accelerated through self-awareness. Paper presented at the Gallup Leadership Summit, Washington, DC.

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods. Thousand Oaks, CA: Sage.

Shamir, B., & Eilam, G. (2005). "What's your story?" A life-stories approach to authentic leadership development. The Leadership Quarterly, 16, 395-417.

Shaw, R. B. (1997). Trust in the balance: Building successful organizations on results, integrity and concern. San Francisco, CA: Jossey-Bass Publishers.

Sparrowe, R. (2005). Authentic leadership and the narrative self. The Leadership Quarterly 16, 419-439.

Speece, D. L., Ritchey, K. D, Cooper, D. H., Roth, F. P., & Schatschneider, C. (2004). Growth in early reading skills from kindergarten to third grade. Contemporary Educational Psychology, 29, 312-332.

Spritzer, G. (2006). Leading to grow and growing to lead: Leadership development lessons from positive organizational studies. Organizational Dynamics, 35(4), 305-315.

Stapleton, L. M. (2006). Using multilevel structural equation modeling techniques with complex sample data (pp. 345-383). In G. R. Hancock & R. O. Mueller (Eds.), Structural equation modeling: A second course. Charlotte, NC: Information Age Publishing.

Stronge, J., Ward, T., Tucker, P., & Hindman, J. (2007). What is the relationship between teacher quality and student achievement? An exploratory study. Journal of Personnel Evaluation in Education, 20(3/4), 165184.

Tschannen-Moran, M., & Hoy, W. K. (2000). A multidisciplinary analysis of the nature, meaning, and measurement of trust. Review of Educational Research, 70(4), 547-593.

Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., & Peterson, S. J. (2008). Authentic leadership: Development and validation of a theory-based measure. Journal of Management, 34(1), 89-126.

Walumbwa, F. O., Wang, P., Wang, H., Schaubroeck, J., & Avolio, B. J. (2010). Psychological processes linking authentic leadership to follower behaviors. The Leadership Quarterly, doi: 10.1016/j.leaqua.2010.07.015.

Wynn, S., Carboni Wilson, L., & Patall, E. (2005). Beginning teachers' perceptions of mentoring, climate and leadership: Promoting retention through a learning communities perspective. Paper presented at the American Educational Research Association, San Francisco, CA.

Yammarino, F. J., & Atwater, L. E. (1997). Do managers see themselves as others see them? Implications of self-other rating agreement for human resources management. Organizational Dynamics, 24(4), 35-44.

Yammarino, F. J., Dionne, S. D., Schriesheim, C. A., & Dansereau, F. (2008). Authentic leadership and positive organizational behavior: A meso, multi-level perspective. The Leadership Quarterly, 19(6), 696-707.

Chuang Wang, University of North Carolina at Charlotte

James J. Bird, University of North Carolina at Charlotte Table 1 Means, Standard Deviations, Inter-Correlations, and Internal Consistencies for Authentic Leadership Questionnaire (n = 60 for principals and n = 917 for teachers) 1 2 3 4 5 6 1 -- .88 ** 91 ** 92 ** .82 ** .12 2 -- 73 ** 74 ** .63 ** 3 -- 79 ** .64 ** 4 -- .69 ** 5 -- 6 -- 7 8 9 M 4.64 4.50 4.59 4.88 4.59 4.02 SD 0.55 0.58 0.59 0.67 0.64 0.82 Alpha .93 .81 .76 .87 .76 .96 7 8 9 10 1 2 .11 3 .08 4 .21 5 .07 6 94 ** 93 ** 91 ** 91 ** 7 -- .81 ** .80 ** .85 ** 8 -- .81 ** 78 ** 9 -- 78 ** M 3.86 4.06 4.26 3.88 SD 1.02 0.81 0.82 0.95 Alpha .91 .84 .88 .81 Note. (a). 1/5 were principal's self/report, and 6/10 were means of teachers' ratings; (b). 1 & 6 were total authentic scores, 2 & 7 were self/awareness, 3 & 8 were transparency, 4 & 9 were ethical/moral, and 5 & 10 were balanced processing; (c). ** p < .01. Table 2 Means, Standard Deviations, Inter-Correlations, and Internal Consistencies for Teacher Engagement and Trust (n = 917) 1 2 3 4 5 Engagement (1) -- .79 ** .77 ** .64 ** .72 ** Trust Total (2) -- .96 * .81 ** .92 ** Trust School (3) -- .67 ** .90 ** Trust Colleagues (4) -- .58 ** Trust Principal (5) -- M 4.08 3.97 3.85 3.95 4.12 SD 0.58 0.67 0.83 0.65 0.76 Alpha .86 .97 .95 .93 .94 Note. ** p < .01. Table 3 Goodness-of-Fit Statistics of the Measurement Models for Each Construct [chi square] df NFI NNFI CFI GFI Authenticity 147.36 100 .73 .83 .86 .68 (Principal) Authenticity 705.20 100 .98 .98 .98 .92 (Teacher) Engagement 387.13 52 .96 .96 .96 .94 Trust 3482.21 461 .98 .98 .98 .81 SRMR RMSEA LCI HCI Authenticity .11 .110 .069 .140 (Principal) Authenticity .04 .080 .074 .085 (Teacher) Engagement .05 .082 .075 .090 Trust .07 .083 .080 .086 Note. These models were second-order for authenticity and trust but first order for engagement. NFI = normed fit index; CFI = comparative fit index; GFI = goodness of fit index; SRMR = Standardized Root Mean Square Residual; RMSEA = Root Mean Square Error of Approximation; ICL = lower bound 90% confidence interval of RMSEA; HCI = higher bound 90% confidence interval of RMSEA. Table 4 Estimates of ICC for Items, First-Order, and Second-Order Constructs Authenticity Variables ICC Q8 10.07 Q9 16.83 Q10 11.33 Q11 13.98 Q12 13.04 Q13 15.47 Q14 9.88 Q15 9.99 Q16 16.55 Q17 7.26 Q18 18.47 Q19 17.05 Q20 15.68 Q21 13.86 Q22 15.94 Q23 17.66 Self-awareness 15.79 Relational Transparency 13.05 Internalized Moral Reasoning 12.97 Balanced Processing 14.26 Authenticity 13.94 Trust Variables ICC Q24 14.35 Q25 18.02 Q26 10.73 Q27 24.93 Q30 7.38 Q32 17.65 Q33 22.38 Q35 12.19 Q42 14.11 Q49 18.81 Q54 13.52 Q37 8.74 Q38 5.83 Q40 4.11 Q43 4.58 Q45 10.31 Q47 7.80 Q48 9.68 Q50 3.76 Q53 6.01 Q28 17.44 Q31 18.62 Q36 14.54 Q39 16.87 Q41 11.61 Q44 18.10 Q46 21.16 Q51 22.98 Q52 9.18 Q55 2.33 Trust School 15.82 Trust Colleagues 6.76 Trust Principal 15.28 Trust Total 13.54 Engagement Variables ICC Q57 5.97 Q58 7.37 Q59 5.28 Q60 7.74 Q61 4.99 Q62 4.19 Q63 6.94 Q64 8.29 Q65 9.51 Q66 2.85 Q67 5.22 Q68 6.00 Engagement 6.19 Table 5 Estimates for Principal's Authentic Leadership with Hierarchical Linear Models Simple Conditional Model Coefficient SE t P Teacher Level Trust-Principal 0.89 0.02 37.36 <.001 Engagement 0.76 0.05 16.64 <.001 School-Level (Adjusted Mean Authenticity) Self-Report 0.01 0.08 0.09 .93 Teacher-Rated 0.99 0.01 141.44 <.001 School-Level (Trust-Principal) Self-Report -0.01 0.05 -0.26 .80 Teacher-Rated -0.11 0.05 -2.18 .03 School-Level (Engagement) Self-Report 0.02 0.07 0.27 .79 Teacher-Rated -0.41 0.09 -4.70 <.001 Complete Conditional Model Coefficient SE t P Teacher Level Trust-Principal 0.86 0.03 25.14 <.001 Engagement 0.02 0.03 0.68 .50 School-Level (Adjusted Mean Authenticity) Self-Report Teacher-Rated 0.99 0.01 141.44 <.001 School-Level (Trust-Principal) Self-Report Teacher-Rated -0.10 0.06 -1.63 .10 School-Level (Engagement) Self-Report Teacher-Rated -0.01 0.05 -0.26 .79