INVESTIGATING LEVELS OF OPENNESS AND TRUST BETWEEN PRINCIPALS
 AND THEIR TEACHERS (A version of this manuscript was published in the following: Southern Regional Council on Educational Administration: 2001 Yearbook)

A B S T R A C T

The purpose of this research was to investigate the level of openness and trust that exists between teachers and their respective principals. Openness and trust are two factors that are essential for human relations and effective leadership.  A secondary purpose was to see if the gender of the teacher or principal had any effect on levels of openness and trust.  One hundred and sixteen graduate students in an educational leadership preparation program evaluated their principals’ level of openness and trust.  The data from this research clearly indicate that improvement is needed on these two constructs that affect an organization’s culture and productivity.
 
 

 Introduction

Garmston (1998) suggested that sometimes groups have characteristics that make them so dysfunctional that attempting to implement change without improving some fundamental qualities is futile.  Qualities that can make a group dysfunctional according to Garmston are the following: ineffective communication patterns, limited trust and respect for the leader, and feelings of not being listened to by the leader. Garmston identified a leadership problem that has been identified by a number of other writers and researchers.   According to Sass (1989), interpersonal communication skills, human relations, and leadership are the most important skills for educational leaders.  Harrill (1990), Harrison (1993), Hutchison (1988), Jolly (1995), and Rouss (1992) all support the premise that human relations and interpersonal skills are competencies needed for effective leadership. Human relations is defined as those formal and informal interactions that occur between people. Bulach (1998), based on his assessment of the leadership skills of 51 aspiring school leaders, concluded that more than 50% have weaknesses in the human relations area. In other words, their interactions with others tended to produce a negative situation. Similar findings were reported by Kramer (1993) in his meta-analysis (35 studies) of school site leader behaviors.  Sergiovanni (1998) in discussing school climate and culture also stressed the importance of human relations.
Bulach, Boothe, and Pickett (1998) surveyed 375 teachers to identify those behaviors their principals practiced that were mistakes.  An analysis of the data produced 14 categories of perceived mistakes or harmful behaviors that principals practiced.  Perceived mistakes in human relations and interpersonal communications were the most frequently reported.  Specific behaviors in the human relations area were a lack of trust and an uncaring attitude.  The most frequently perceived behavior in the area of interpersonal communications was failure to listen or a lack of openness. Clearly, openness and trust are essential for human relations.

A number of organizational theorists talk about the importance of the openness and trust constructs for effective leadership.  For example, Kaiser (1992) stated that leaders have to operate in an atmosphere of openness, honesty, and trust. Saxl, Miles, and Lieberman (1989) stated that building trust is the most important skill leaders need to improve organizations. McGarry (1991) stated that communication and trust are essential if educational services are to be improved. Finally, there is overwhelming support for the importance of  human relations and more specifically the constructs of openness and trust for effective leadership. However, most of this support is based on opinion and logic and is not supported by data or research. Given that openness and trust are thought to be so important for effective leadership, to what extent do these constructs exist between principals and their teachers?

Purpose of this Research

The purpose of this research was to investigate the level of openness and trust that exists between teachers and their principal.  A secondary purpose was to see if the gender of the teacher or principal had any effect on levels of openness and trust.
 

Hypotheses:
 

#1. Overall level of trust between principals and teachers is present.
#2. Overall level of openness between principals and teachers is present.
#3. The telling and listening dimensions of openness are present.
#4. The character dimension of trust is present.
#5. The ability and truthfulness dimensions of trust are present.
#6. The predictability and confidentiality dimensions of trust are present.
#7. There is a difference in expected and reported behaviors for levels of openness and trust.
#8. There is a difference between male and female principals in expected and reported behaviors for levels of openness and trust.
#9. There is a difference between male and female teachers in expected and reported behaviors for levels of openness and trust .

Definitions:

Trust : an interpersonal condition that exists when interpersonal relationships are characterized by an assured reliance or confident dependence on the character, ability, truthfulness, confidentiality and predictability of others in the group (Bulach,1993).
Openness: an interpersonal condition that exists between people when: (1) facts, ideas, values, beliefs, and feelings are readily transmitted; and (2) the recipient of a transmission is willing to listen to that transmission (Bulach,1993).
Gender : male or female.
Expected level of openness and trust: the level of openness and trust that teachers perceive is expected by their principal.
 Reported level of openness and trust: the level of openness and trust that teachers report they practice with their principal.

Methodology

One hundred and sixteen graduate students (98 females and 18 males) in the leadership preparation program at the University of West Georgia were involved in this study. They reported their perceptions of the levels of openness and trust that existed between themselves and their principals.  The principal data pool consisted of 45 males and 71 females. Some students had the same principal. Consequently, data were sometimes collected on the same principal.
The Group Openness and Trust Scale (GOTS) used in this study was developed by Bulach (1974) and later revised in 1993 and in 1998. The GOTS can be used to measure levels of openness and trust within groups or levels of openness and trust between teachers and their principal. It consists of 60 behaviors about openness and trust. Teachers respond to each behavior on a 5-point Likert scale ranging from strongly disagree to strongly agree.  Four of the items are negative and must be reverse scored.  Scores can range from 1.0 (strongly disagree) to 5.0 (strongly agree). The instrument has construct validity and reliability using the Cronbach alpha is a +.93.

Thirty items measure the extent to which teachers believe that the principal expects them to be open and trusting (expected level of openness and trust). The next thirty items are the same except they ask the teacher to report what they do on these behaviors (reported level of openness and trust). For example, an expected behavior would be as follows: “ My principal expects me to tell the truth when it needs to be told.” The reported behavior would be as follows: “I tell the truth when it needs to be told.”  Since neither expected behavior nor reported behavior may represent actual behavior, the two behaviors (expected and reported) are combined for an average score. It is believed that the most valid measure of openness and trust lies somewhere between expected and reported behaviors.
The openness construct has two dimensions measured by 12 expected and 12 reported behaviors. The two openness dimensions are telling and listening. An example of a behavior in the telling dimension is “. . tell him what I think of his/her ideas.” An example of a listening behavior is “. . ask him/her what I think of his/her ideas.”

The trust construct has five dimensions measured by 18 expected and 18 reported behaviors. The five dimensions are as follows: character, ability, predictability, confidentiality, and truthfulness. An example of a behavior in the character dimension is  “. .believe that s/he cares about me.” An example of a behavior in the ability dimension is “. . have faith in his/her ability.” An example of a behavior in the predictability dimension is “. . believe that s/he will respond favorably in a given situation when my welfare is at stake.” An example of a behavior in the confidentiality dimension is “. . rely on him/her to keep a confidence.” An example of a behavior in the truthfulness dimension is “. . believe what I hear him/her say.”

ANALYSIS

Descriptive statistics were used to analyze and report the data. In order for hypotheses #1-6 to be accepted, scores close to 4.0 (an agree response) must be present. In order for hypotheses #7-9 to be accepted, the level of significance must be .05 or lower.  A t-test for dependent groups was used to determine if there was a significant difference between "expected" behaviors and "reported" behaviors (hypothesis #7).  A t-test analysis for independent groups was used to determine if there were differences in openness and trust levels for male and female principals and male and female teachers (hypothesis #8-9).

Results

The overall level of trust between principals and teachers, as measured by the five dimensions, was 3.74.  Agreement that trust was present would have resulted in scores that were 4.0 or higher (Table 1 has the individual dimension scores). Consequently, hypothesis #1 that the overall level of trust between principals and teachers would be present was rejected.

Table1
 
 
            Trust Dimensions                                               Mean
Character
3.46
Truthfulness
3.99
Ability
3.96
Confidentiality
3.73
Predictability
3.57
   
Overall Trust
3.74
            Openness Dimensions                                    Mean
Listening
3.09
Telling
2.96
   
   
   
   
Overall Openness
3.03

 
 

The overall level of openness between principals and teachers, as measured by the two dimensions was 3.03 ( Table 1 has the individual dimension scores). A score of 3.0 indicates that a number of teachers disagreed that openness was present. Consequently, hypothesis #2 that the overall level of openness between principals and teachers would be present was rejected.

The score on the telling dimension of the openness construct was 2.96. The score on the listening dimension of the openness construct was 3.09.  These scores indicate disagreement that openness was present on both dimensions.  Consequently, hypothesis #3 that the telling and listening dimension of openness would be present was also rejected.

The first of the five trust dimensions tested was the character dimension.  The overall score was 3.46 indicating some disagreement that this dimension was present (Table 1).  Consequently, hypothesis #4 that the character dimension of trust would be present was rejected.The overall score for the truthfulness dimension was 3.99 and 3.96 for the ability dimension  indicating some agreement that these dimensions were present (Table 1). Both scores are close to 4.0 which is an agree response that the behaviors associated with these dimensions were present. Consequently, hypothesis #5 that the ability and truthfulness dimensions of trust would be present was accepted.

The overall score for the predictability dimension was 3.67 and 3.73 for the confidentiality dimension  indicating some disagreement that these dimensions were present (Table 1). Consequently, hypothesis #6 that the predictability and confidentiality dimensions of trust would be present was rejected.

The t-test (see Table 2) to determine if there was a significant difference in expected and reported behaviors yielded significant t’s on the following dimensions: listening (p < .05), character (p < .002), truthfulness  (p < .000), ability (p < .000), predictability (p < .000).
 
 


 

The Bonferoni technique was used to correct for errors associated with the use of multiple t’s. This caused the alpha level for listening to be greater than .05 (p > .05) (no longer significant). The remaining factors were less than .05 (p < .05). Consequently, hypothesis #7 that there would be a difference in expected and reported behaviors was only accepted for the dimensions of character, truthfulness, ability, and predictability and rejected for the following dimensions, listening, telling, and confidentiality.

The t-test (see Table 3) to determine if there was a difference in expected and reported behaviors for levels of openness and trust for male and female principals yielded only one significant t-score. The expected behavior for ability had a t-score of 2.72.  However, when it was adjusted for errors associated with multiple t’s (14 x .007 = .09) it was no longer significant.  Consequently, hypotheses #8 that there would be a difference in expected and reported behaviors for levels of openness and trust for male and female principals was rejected.

The t-test (see Table 3) to determine if there was a difference in expected and reported behaviors for levels of openness and trust for male and female teachers did not produce any significant t-scores. Consequently, hypothesis #9 that there would be a difference in expected and reported behaviors for levels of openness and trust for male and female teachers was rejected.

Discussion

While the average score for all five dimensions of trust did not reach 4.0 for an agree response, there was some agreement on  the truthfulness and the ability dimensions.  However, on closer analysis, it appears that the average score on these two dimensions is somewhat distorted by the higher expectations principals have on these two dimensions.  For example, teachers reported that principals expect teachers “to believe what they hear principals say" for a score of 4.13.  However, when teachers reported what they do on this behavior they expressed some disagreement with this behavior with a score of 3.56.  A similar pattern was found for the other three behaviors that comprise this dimension of trust.  Consequently, on closer examination it appears that teachers are unwilling to trust as much on these dimensions as their principal would like.

The level of openness is much worse than the level of trust.  While the average trust level is 3.74, it is only 3.03 for openness.  An example of an openness behavior “The principal expects me to tell him what I think of the way s/he does things” had a score of 2.67 which means that they don’t expect teachers to be open on this behavior.  Teachers also reported with a score of 2.57 that they are not open and meet the principal’s expectation on this behavior.  Another example of a behavior with a low score (2.60) was “The principal expects me to disagree if I don’t agree with what is being said or done.” Similar scores, indicating a lack of openness, were found for other behaviors on both the telling and listening dimensions of openness.

The low scores on the openness construct could contribute to the low trust scores.  The correlation between these two constructs in this study was +.56 (p < .001).  In other studies, using the same instrumentation, correlations have approached +.70.  This finding verifies the opinion of Kouzes and Posner (1993) that openness is required for trust to develop.  They go on to say that “Building trust begins by building a personal relationship through listening” (p.100). If principals are going to improve levels of trust, they are going to have to work on levels of openness.  Since openness and trust are closely intertwined, the question of which is more important cannot be answered. Teachers will not be open with someone they do not trust, nor will they trust someone whom they perceive is not open.
While teachers tended to disagree on the presence of the behaviors associated with the telling and listening dimensions of the openness construct, there was agreement on two of the behaviors.  The behavior in the telling dimension “Teachers share positive thoughts with the principal instead of keeping it to themselves” had a score of 3.85, indicating modest agreement that this behavior occurred.  The behavior in the listening dimension “Teachers accept the principal’s comments and reactions” had a score of 3.9.  All the other scores in the listening dimension indicate that teachers don’t often ask for feedback from their principal.  In the telling dimension the scores indicate they don’t often give their principal feedback unless it is positive.  This could lead to a false sense of performance, where people think they are doing all right when in reality, both the principal and the teachers could be dissatisfied with each other’s performance.
 

The character dimension with a score of 3.46 was the lowest of the five dimensions in the trust construct. Scores of 3.1 and 3.2 respectively on the behaviors “Teachers control their reactions and feeling to what the principal does and says,” and “Teachers keep their distance from the principal” would indicate that principals frequently do not know what their teachers are thinking or feeling.  This coupled with the low scores on the listening and telling dimensions could lead to a situation where both teachers and principals are in the dark regarding what each really thinks about what is happening in the school setting.
The ability and truthfulness dimensions with scores of 3.96 and 3.99 offer some hope that trust can be developed between teachers and their principal. Scores this high indicate that a number of teachers had an agree or completely agree response on the eight behaviors that make up these two trust dimensions. However, the score for expected behavior was higher in every instance than it was for reported behavior. The widest spread in scores was for the behavior “The principal expects me to have faith in his ability” with a score of 4.2 versus the reported behavior of what teachers “say they do on this behavior” with a score of 3.68.  This indicates that a number of teachers have faith in their principal’s ability, but not as  many as the principal expects.

There were some positive behaviors in the confidentiality and predictability dimensions of the trust construct.  Teachers with scores of 4.0 and 3.95 respectively admit their mistakes and problems when necessary and they believe that their principal will respond favorably when their welfare is at stake.  Behaviors that show a lack of trust involve the gossip or rumor mill.  Teachers reported that they don’t always work directly with the principal when there is a problem---the predictability dimension.  This occurs when there is some doubt regarding how the principal will respond to the problem.  Consequently, the problems are discussed in the gossip or rumor mill--the confidentiality dimension.

The difference in expected and reported behaviors occurred in every openness and trust dimension.  Teachers consistently reported that they were less open and trusting on all behaviors except for two of them: (1) on the behavior “Teachers ask the principal about his/her feelings” the reported behavior score was 2.71 and the expected score was 2.62; and (2) on the  behavior “Teachers openly disagree with the principal when they do not agree with what is being said or done” the reported score was also 2.71 and the expected score was 2.60. Keep in mind that even though reported scores are higher, they are still disagree responses.
Teachers agreed with scores above 4.0 that their principal expected openness or trust on eight of the 30 behaviors. Teachers, on the other hand agreed with scores of above 4.0 on only one of the 30 behaviors.  The behavior with the highest expected behavior score (4.33) was “The principal expects teachers to believe that s/he is honest.”  This also happened to be the only one, with a score of 4.02, where the teachers reported agreement that they practiced that behavior.

This finding that teachers are less willing to be open and trusting (expected scores were higher than reported scores) with their principal differs from openness and trust data gathered on 40 schools in Georgia and 37 schools in Kentucky.  In that data, teachers invariably reported that they were more willing to be open than was expected. While the reported scores were higher than expected scores, they still were not open. The data from these schools is a measure of openness and trust levels between the teachers, whereas the data in this study is a measure of openness and trust levels between the principal and the teachers.  Apparently, teachers are more willing to be open and trusting with their colleagues than they are with their principal.  Of the 77 schools on which levels of group openness and group trust were collected, only five schools had teachers who agreed that trust existed with their colleagues. None reported that openness existed.  The finding that teachers do not trust their principal is disturbing, but it is even more disturbing to know that teachers, in most schools, are not open and do not trust each other.
The data regarding the sex of the principal and the teachers was somewhat of a surprise.  It was anticipated that females would be more trusting than males, but that was not the case. Perhaps it is a good sign that there is no difference.  Only one of the dimensions  “expected ability” came close to being statistically significant.  Female principals with a score of 17.0 expected teachers to trust their ability more than male principals with a score of 15.8.  Since there are four behaviors (4 x 4 = 16) in this category, each behavior received an average agree response of 4.0 or higher.  One possible explanation for this is that more female principals than male principals have a background in elementary education.  Since Georgia only has one administrative certificate for all levels of administration, it is easy for someone with a high school background to become an elementary principal.  This happens with regularity because of the shortage of males who are going into administration. For example, in this study, only 18 of the 116 teachers in the educational leadership program were male.  A male principal, with a high school background, probably would not expect teachers to trust their ability since they have little or no experience at the elementary level.  This could account for the difference in expectations.

Suggestions for Improving Levels of Openness and Trust

There are a number of things principals can do to improve levels of openness and trust.  Principals need to spend more time listening to their teachers and encouraging them to give him/her feedback on a wide variety of items, e.g., allow teachers to evaluate the principal, have group meetings where faculty and administration can share how things are going and how they can be improved.  The vision and mission of the school could be discussed, but apparently this does not occur.  For example, the lowest score (2.4) was on the behavior “Teachers tell the principal what they think of his/her values and beliefs.”  While educational experts agree that the vision a principal has for a school is very important, apparently it is seldom discussed.
Other suggestions deal with being predictable, keeping confidences, and showing teachers that you care about them. One of the easiest ways to show that you care is by listening to teachers. Kouzes and Posner (1993) stated that leaders have to stay open to others and that “Trust is maintained when people see that we are not ‘know-it-alls’ and are interested in learning from others” (p.262). Listening is an openness dimension that is a building block for trust.  When a person listens to someone, a message is conveyed that you value that person, and that you have time for them because they are important.  In other words, you care about them.  When people believe that you care, the process of trusting is under way. According to Fullan (1998), the old take charge leadership style must be replaced with a sharing style and high-level communication skills.

One other suggestion that impacts the predictability dimension of trust is authenticity. Many leaders read publications that describe methods and strategies for improving management and leadership skills.  As their style is adapted to the new methods, it is easy to lose authenticity.  These "self-help" books provide an illusion of security for principals and teachers. Living authentic lives is very important for openness and trust. In other words, teachers need to know who their principal really is and what they stand for. A principal who changes roles and approaches frequently will be perceived as unpredictable and unauthentic.

Authenticity exists when principals are open and honest with themselves and secondly, when they are able to remove those barriers that keep them from revealing themselves to others.  This means that principals must first examine their "fears" from which emerge obstacles to their living authentic lives.  These obstacles may include their heavy reliance on methodology, strategy, and acting as if they are giving a performance (Goffman, 1959). Bolman and Deal (1995) in their book Leading with Soul, suggested that leadership first begins with a connection to the heart and not with management skills.  The closer leaders are seen as being authentic and able to live in harmony with their moral and spiritual beliefs, the more they will be perceived as being open and trustworthy.

Conclusions

The current levels of openness and trust  in our school systems could be a result of  teachers and  who received no training related to these constructs. Perhaps curriculum needs to be developed for graduate and undergraduate students.  This was a conclusion reached by Bulach (1998) who described the effect of human relations training on selected leadership skills. It is possible that training of this type for an entire school faculty would create some bonding or feeling of community that would increase levels of openness and trust.  Experiential training at the undergraduate and graduate level could provide students with the necessary skills enabling them to be more open and trusting as they enter the workforce.  Training of this type was described by Bulach and Potter (1998) in their work with leadership assessment centers.

This manuscript started with a review of literature on human relations and narrowed to levels of openness and trust as important for human relations. The findings of Bulach et al. (1998) reinforce the previous research that human relations are a crucial area for effective leadership. While there are no definitions of human relations, the words trust, communications, and listening skills are recurring themes.  It seems logical that trust would be at the heart of human relations.  In a marriage, which should epitomize successful human relations, trust is essential.  Closely related to trust is a construct called openness.  When people trust, they leave themselves open, and this can subject them to risk being hurt by the person they trust. As a result, there is a tendency for teachers and principals to be closed rather than open.
While there is little disagreement that levels of openness and trust are important for an effective organization, there is little agreement that they are present in the schools of Georgia.  The data from this research clearly indicate that improvement is needed on these two constructs that affect an organization’s culture and productivity. According to Cherniss (1998), emotionally intelligent educational leaders possess people skills. Openness and trust are two basic people skills. It is time to stop talking about how important openness and trust are for an organization and do something about it.  It is “Time to walk the talk!” Principals need to listen to their teachers!  If principals become open and trusting role models, perhaps a culture can be created where teachers will start being open and trusting with each other. A process for creating such a culture is described by Bulach (2001).

References

    Bolman, L. G., & Deal, T. E. (1995). Leading with soul. San Francisco, CA: Josey-Bass Inc.
    Bulach, C. R. (2001). A four step process for identifying and reshaping school culture. Principal Leadership, 1(8), 48-51
    Bulach, C. R. , Boothe, D., & Pickett, W. (1998). "Should nots" for school principals: Teachers share their views. ERS SPECTRUM: Journal of School Research and Information, 16(1), 16-20. (80 %)
    Bulach, C. R. (1998). The impact of human relations training on selected leadership variables. Paper presentation at the Southern Regional Council of Educational Administrators  Conference at Savannah, GA on 11-3-1998.
    Bulach, C. R., & Potter, L. (1998). Assessment centers: Meeting the needs of educational leaders. NASSP Practitioner, 24(3), 1-4.
    Bulach, C. R., & Malone, B. (1994). The relationship of school climate to the implementation of school reform. ERS SPECTRUM: Journal of School Research and Information 12(4), 3-9.
    Bulach, C. R. (1993). A measure of openness and trust. People in Education, 1(4), 382-392.
    Bulach, C. R. (1974). An investigation of the relationship of group openness and trust for group decisions involving risk. An unpublished doctoral dissertation , University of Cincinnati, OH.
    Cherniss, C. (1998). Social and emotional learning for leaders. Educational Leadership, 55(7), 26-28.
    Fullan, M., (1998). Leadership for the 21st century: Breaking the bonds of dependency. Educational Leadership, 55(7), 6. 6-10.
    Garmston, R. J. (1998). Too dysfunctional to be trained. Journal of Staff Development, 19(4), 56-57.
    Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY: Doubleday and Company, Inc.
    Harrill, J. L. (1990). A descriptive study of South Carolina district-level curriculum/instructional leaders' perceptions of needed competencies and skills. Dissertation Abstracts International, 51(08), 2620. (University Microfilms No. AAI9101463)
    Harrison, P. T. (1993). The critical elements of effective principal preparation: A
Delphi study (effective principals). Dissertation Abstracts International, 54(03), 0753. (University Microfilms No. AAI9322195)
    Hutchison, C. I. (1988). Leadership skills. Performance and Instruction, 27(8), 2-5.
    Jolly, R. A. F. (1995). The effectiveness of secondary educational administration preparation at Kansas regents universities. Dissertation Abstracts International, 56(11), 4224. (University Microfilms No. AAI96099508)
    Kaiser, J. S. (1992). Educational Administration. Mequon, WI: Stylex Publishing Co.
    Kouzes, J. M., & Posner, B. Z., (1993). Credibility. San Francisco, CA: Jossey-Bass Publishers
    Kramer, B. H. (1993). Empirically indicated effective school-site leader behaviors:
    A meta-analysis and theoretical comparison (effective leadership). Dissertation Abstracts International, 54(09), 3282. (University Microfilms No. AAI9803729)
    McGarry, T.P. (1991). The superintendent. Educational Leadership, 48(6), 80-81.
    Ruoss, E. G. (1992). Essential competencies for independent school leaders (leadership training). Dissertation Abstracts International, 54(05), 1623. (University Microfilms No. AAI9324891)
    Sass, M. W. (1989). The AASA performance goal and skill areas' importance to effective superintendency performance as viewed by professors of educational administration and practicing superintendents. Dissertation Abstracts International, 50(09), 2742. (University Microfilms No. AAI9006076)
    Saxl, E. R., Miles, M. B., & Lieberman, A. (1989). Assisting change in education. Alexandria VA: Association for Supervision and Curriculum Development, 1-5.
    Sergiovanni, T. J., & Starratt, R. J. (1998). Supervision: A redefinition. Boston, MA: Mcgraw-Hill.