Introduction

The most significant challenge of leadership is to build and sustain an organizational culture that focuses on continual improvement of educational programs, teachers’ capacities and skills, and student learning. The school administrator as the primary instructional leader is insufficient to meet these challenges and, as a result, several experts advocate the dispersal of leadership authority within a school (e.g., Harris, 2003; Fullan, 2001; Lambert, 1998; Leithwood & Riehl, 2003; Sergiovanni, 2001). Such dispersal has the potential to motivate teachers, improve the quality of teaching in the classroom and positively impact student achievement.

Transformational leadership is a desirable style for school leaders involved in improvement efforts because it raises the level of awareness of workers so that they come to value organizational goals and strategies to achieve those objectives. Burns (1978) characterized transformational leaders and distinguished them from managers because of their special ability to bring about organizational innovation and change. Transformational leaders change organizational culture by introducing new beliefs and goals and by changing how group members define their roles.

Although many schools are developing teachers as leaders, researchers have not sufficiently investigated the leadership behaviors of these individuals. Therefore, using data collected from teacher leaders and their principals, I addressed three questions in this study:

1. What are the leadership practices of teacher leaders, as perceived by the teacher leaders and their principals?

2. Are there significant differences between the perceptions of teacher leaders and principals regarding the leadership practices of teacher leaders?

3. Are selected demographic variables of teacher leaders, specifically age, years of experience, educational level, and gender related to the leadership behaviors of teacher leaders?

I collected data from 88 Connecticut teacher leaders and their principals employed in the public schools. Characteristics of the participants are presented in Table 1.

Procedure

After identifying all public schools in Connecticut (excluding technical high schools and charter schools), I randomly selected 500 using a table of random numbers. I sent an email letter to the school principals explaining the purpose of the study and inviting them to participate. If they elected to participate, they selected a teacher leader in their building with whom they were familiar and forwarded a letter of invitation to that individual. Both the principal’s and teacher leader’s email contained a link to an online survey which included the appropriate LPI and the demographic data form.

Materials

Each teacher leader and principal completed the Leadership Practices Inventory (LPI), 3^rd Edition, an instrument developed by Kouzes and Posner (2003) to measure transformational leadership behaviors of practicing leaders. The teacher leaders completed the LPI-Self version and the principals completed the LPI-Observer form.

1. Modeling the way: This scale measures whether leaders lead by example. They also help followers focus on step-by-step accomplishments of large-scale goals, making those goals seem more realistic and attainable.

2. Inspiring a shared vision: This scale measures what leaders actually do to construct future visions and to build follower support for the visions rather than simply providing inspiration.

3. Challenging the process: This scale measures the ways in which the leader searches for opportunities and experiments, even taking sensible risks, to improve the organization.

4. Enabling others to act: This scale measures how the leader makes it possible for followers to take actions by fostering collaboration and supporting followers in their personal development.

5. Encouraging the heart: This scale measures the manner in which leaders recognize followers’ contributions and find ways to celebrate their achievements.

The authors reported internal reliabilities ranging from .77 to .90.

The instrument consists of 30 descriptive statements (six for each of the five leadership factors) regarding various leadership behaviors and activities. The teacher leaders rated how frequently they engaged in the practices using a 10 item Likert-type scale which ranged from almost never to almost always. The principals rated how frequently the teacher leader engaged in the same practices using the same scale. Therefore, raw scores for each leadership practice could range from 6 to 60. The raw scores were converted to Normal Curve Equivalents (NCEs) and percentile ranks based on national norms provided by the authors.

Subjects also completed a demographic survey regarding age, years of experience as an administrator or teacher, educational level, and gender.

Results

Of the 500 principals who were invited to participate, 118 or 23.6% completed the LPI-Observer and demographic form. Of the 118 teacher leaders who received a letter of invitation forwarded by the principal, 88 or 74.58% of them completed the instruments. The final sample consisted of the 88 teacher leaders and their principals who both completed the necessary instruments.

Therefore, I did not include 30 of the principals in the final sample because their teacher leaders did not complete the LPI-Self.

I used NCE means, NCE standard deviations and median percentile ranks to describe the leadership practices of the teacher leaders. I employed a paired-sample t-test to compare the responses of the principals and teacher leaders for each of the five leadership factors. Finally, I performed multiple regression analyses to identify relationships between the demographic data of the teacher leaders and the five leadership practices of the teachers as reported by the teachers and the principals.

Leadership Practices of the Teacher Leaders

Table 2 displays the NCE mean and standard deviation for each of the five leadership factors measured by the LPI. Teachers and principals rated Challenging the Process as a relative strength of these teacher leaders with mean NCE scores of 55.21 and 66.80, respectively. The lowest mean NCE score for the teacher leaders was Encouraging the Heart (45.80) and this was also among the lowest means for the principals (56.72).

Figure 1 graphically presents the median percentile ranks for the teachers and the principals based on the normative data of the LPI-Self and LPI-Observer forms. The teacher leaders were above average for the leadership practice of Challenging the Process. The principals placed the teachers in the top quartile for this behavior (79^th percentile) and the teacher leader ratings fell at the 60^th percentile. Enabling Others to Act was the second highest percentile rank for the teachers (59^th percentile) but this was the lowest teacher leadership behavior rated by the principals, although the administrators’ and the teachers’ median percentile ranks were the same.

The principals rated the teacher leaders in the top quartile among national leaders (76^th percentile) for Inspiring a Shared Vision; the teacher leaders perceived themselves to be only average (50^th percentile) for this practice.

Differences between Teachers’ and Principals’ Perceptions of Leadership Practices

I used the paired-sample t-test to analyze the differences between the NCE scores for the principals and the teacher leaders for each leadership practice (see Table 3).

The teacher leaders and principals differed significantly (p < .05) regarding their perceptions of how often they practiced four of the five leadership factors, with principals consistently rating teachers higher. They did not differ for the leadership practice, Enabling Others to Act.

Relationships between Demographic Data of Teacher Leaders and Leadership Practices

Using teacher age, educational level, years of experience as a teacher and gender as independent variables, Table 4 displays the stepwise multiple regression results for each leadership practice as self-reported by the teacher leaders. Independent variables were only entered into the regression model if the probability of F was less than or equal to .05.

Selected demographic variables accounted for approximately 18% to 26% of the variance in teachers’ self-reported leadership practices. The model with the highest R value was Challenging the Process (R = .53) with the combination of years of experience, educational level and gender accounting for 26% of the variance in teachers’ self-reported ratings for this leadership practice. The model with the lowest R value was Inspiring a Shared Vision (R = .43) with the combination of educational level, years of experience and gender explaining approximately 16% of the variance in the self-reported ratings of the teacher leaders.

Educational level of the teachers was a significant predictor of all five reported transformational leadership behaviors. As educational level increased, the frequency of their engagement increased for each leadership practice. Surprisingly, the experience level of the teachers was negatively related to four of the five leadership factors. As teaching experience increased, the frequency of self-reported transformational leadership practice decreased for all factors except Modeling the Way. Gender was also predictive of three of the five factors: Inspiring a Shared Vision, Challenging the Process, and Encouraging the Heart. Female teachers were more likely to rate themselves higher for these leadership behaviors. Age of the teacher leaders was a significant predictor for only one leadership factor, Enabling Others to Act. As teacher age increased, frequency of engagement for this practice increased.

Using the teachers’ demographic variables as predictors, Table 5 displays the stepwise multiple regression results for each leadership practice as reported by the principals. Independent variables were only entered into the regression model if the probability of F was less than or equal to .05. For Inspiring a Shared Vision and Enabling Others to Act, none of the predictor variables met the criteria. Therefore, Table 5 has no regression model results for these two leadership practices.

Selected demographic variables accounted for only 7% to 9% of the variance in principals’ perceptions of teacher leadership practices. The model with the highest R value was Challenging the Process (R = .34) with the combination of years of experience and gender accounting for just 9% of the variance in principals’ ratings. The models with the lowest R values were Modeling the Way (R = .28) and Encouraging the Heart (R= .28). For each of these leadership practices, only years of experience of the teacher leaders was a significant predictor of the principals’ ratings, accounting for just 7% of the variance. As teacher years of experience increased, principals’ ratings decreased.

Unlike the teacher leaders, educational level was not a significant predictor of principals’ ratings of the leadership practices. Years of experience were predictors for three of the five factors: Modeling the Way, Challenging the Process and Encouraging the Heart. As teacher experience decreased, principal ratings increased. Gender was a significant predictor for only one leadership practice, Challenging the Process. Principals were more likely to rate females higher for this factor. Age was not a significant predictor for any of the five leadership practices as rated by the principals.

The findings of this study suggested that a relative strength of teacher leaders was challenging the status quo in their schools. These teacher leaders were involved in school improvement efforts, searching for opportunities and experiments and helping colleagues to take sensible risks to improve student achievement. The principals rated the teachers in the top quartile for this leadership practice, probably because they viewed this as the primary role of teacher leaders in order to support organizational change and improvements in student learning. The accountability pressures of No Child Left Behind legislation and the need to achieve adequate yearly progress (AYP) goals may have been driving forces for these teacher leaders to engage in this transformational leadership practice. The significant difference between the principal and teacher ratings for this practice, however, indicated a need for administrators and teacher leaders to engage in frequent dialogue about school improvement efforts.

Teachers tended to rate themselves within the average range compared to other leaders, but the principals perceived their teacher leaders to be above average. These significant differences suggested that principals were overestimating the frequency with which their teacher leaders were engaging in transformational leadership behaviors, practices that were considered critical to the success of school improvement efforts. Principals should become more intimately aware of the work of teacher leaders and the barriers they may be experiencing which interfere with their ability to practice transformational leadership behaviors. Administrators need to have frank discussions with teacher leaders about their work, their leadership skills, and their professional development needs.

Based on this sample of teacher leaders, it appeared that they would benefit from strategies on how to inspire a shared vision in the school, particularly how to build support for the school vision among their colleagues. The gap between the perceptions of principals and teachers for this transformational practice was the greatest. Perhaps dialogue between school administrators and teacher leaders needs to focus on the definition of the school mission and vision, how to involve staff in that process, and how to embed the vision in classroom practice.

The teacher leaders reported low levels of engagement in practices related to Encouraging the Heart. Since their ratings for this leadership practice were below national averages, the teacher leaders would benefit from training in the importance of recognizing and celebrating the contributions and achievements of other teachers. This may be uncomfortable for the teacher leaders due to cultural norms of equality or perhaps they believe it is not their role to do so. The impact of No Child Left Behind, particularly the difficulty of achieving AYP, may be having a negative influence on teachers’ attitudes; teacher leaders and principals may feel there is very little to celebrate as more and more schools are identified as not meeting AYP.

Demographic variables of the teachers accounted for more variance in the ratings of their own leadership practices compared to the principals’ ratings. Teacher leaders who reported the highest levels of transformational leadership behaviors were more likely to have higher levels of education but lower levels of teaching experience. Principals should consider the educational level of candidates before appointing them to positions of leadership. Follow up studies could explore this relationship further to determine if specific types of advanced education were related to leadership practices. For example, many of the teachers in this sample may have had advanced training in school administration or within content areas for which they hold leadership responsibilities. This advanced training may have given them both the confidence and the skills to lead others.

Future studies could also explore the phenomena that less experienced teachers reported higher levels of transformational leadership behaviors and principals also perceived that more experienced teachers were less engaged in these leadership practices. Principals may wish to consider special support and encouragement for experienced teachers who assume leadership roles. These individuals may have been immersed for many years in a school culture that values norms of equality among teachers. These teacher leaders may find it more difficult to break from such traditions and practice transformational leadership behaviors (Barth, 2001; Smylie, 1992).

References

Barth, R. 2001. Teacher leader. Phi Delta Kappan 82 (6): 443-450.

Burns, J. 1978. Leadership. New York: Harper & Row.

Fullan, M. 2001. Leading in a culture of change. San Francisco: Jossey-Bass.

Harris, A. 2003. Improving schools through teacher leadership. Education Journal 59: 22-23.

Kouzes, J. and B. Posner. 2003. Leadership Practices Inventory, 3^rd Edition.. San Francisco, CA: Wiley

Lambert, L. 1998. Building leadership capacity in schools. Alexandria, VA: Association for Supervision and Curriculum Development.

Leithwood, K. and C. Riehl. 2003. What we know about successful school leadership. Philadelphia, PA: Laboratory for Student Success, Temple University.

Sergiovanni, T. 2001. Leadership: what’s in it for schools? London: Routledge Falmer Gunter.

Smylie, M. 1992. Teachers’ reports of their interactions with teacher leaders concerning classroom instruction. The Elementary School Journal 93 (1): 85-96.

Total 88 100.00 88 100.00

High School 14 15.91 14 15.91

35-39 7 7.95 3 3.41

40-44 5 5.68 7 7.95

45-49 4 4.55 4 4.55

50-54 21 23.86 16 18.18

55 + 23 26.14 56 63.64

Years of Experience:

Less than 5 3 3.41 3 3.41

5-9 29 32.95 5 5.68

10-14 15 17.05 23 26.14

15-19 17 19.32 17 19.32

20-24 1 1.14 19 21.59

25-29 3 3.41 13 14.77

30 or more 20 22.73 8 9.09

Table 3. Differences between teacher leader and principal ratings for leadership practices

Leadership Practice t

Enabling Others to Act .51

Modeling the Way 2.79*

Challenging the Process 4.99**

Encouraging the Heart 3.64**

Inspiring a Shared Vision 5.44**

* p < .05

** p < .001

Table 4. Stepwise regression of selected demographic teacher variables to self-reported leadership practices of teacher leaders

____________________________________________________________________

Modeling the Way .44 .19 .18 6.39 20.05 .000

Educational Level 4.48**

Inspiring a Shared Vision .43 .18 .16 8.25 6.46 .001

Educational Level 3.62**

Years of Experience -2.65**

Gender 2.48**

Challenging the Process .53 .29 .26 7.41 11.19 .000

Years of Experience -4.38**

Educational Level 4.14**

Gender 3.00**

Enabling Others to Act .50 .25 .22 5.13 9.26 .000

Educational Level 4.39**

Years of Experience -3.68**

Age 2.06*

Encouraging the Heart .46 .21 .19 9.26 7.61 .000

Years of Experience -3.83**

Educational Level 3.31**

Gender 2.06*

____________________________________________________________________

Note. Criteria for variables entered is probability of F < or = .05

* p < .05 ** p < .01

Table 5. Stepwise regression of selected demographic teacher variables to leadership practices of teacher leaders as reported by principals

____________________________________________________________________

Modeling the Way .28 .08 .07 5.31 7.40 .008

Years of Experience -2.72**

Challenging the Process .34 .11 .09 4.58 5.37 .006

Gender 2.90**

Years of Experience -2.12*

Encouraging the Heart .28 .08 .07 6.72 7.49 .008

Years of Experience -2.74**

Note. Criteria for variables entered is probability of F < or = .05

* p < .05 ** p < .01

Figure 1. Median percentile ranks for leadership practices as rated by teacher leaders and principals