Between Compassionateness and Assertiveness : A Trust Matrix for Leaders

Purpose: Construct the structure of trust-building perceived by principals in Portuguese public schools. Design/methodology/approach: A grounded theory approach applied to a survey of open questions answered by school’s principals. Findings: (1) self-trust is the single trait common to all respondents; (2) compassion is a striking antecedent in the process of other-trust; compassionateness and assertiveness must combine with one another to achieve leadership based on smart trust, what is outlined in a matrix model built from the data. Research limitations/implications: The major limitation of this study was the impossibility to conduct face-to-face interviews given the limitation of free time alleged by the principals. In order to solve that limitation we decide to use the grounded theory approach. Practical implications: The consideration of a symbiosis category of the behaviors of principals can trigger positive expectations among the players who until now did not understand how it is possible to bring together the best of the worlds of management and leadership in a balanced way, with efficiency in authenticity.


Introduction
Trust is a multidimensional construct.The first studies on trust emerged in the areas of Psychology (Deutsch, 1958, in conflict resolution;Rotter, 1967;Zand, 1972, in interpersonal trust) and sociology (Luhmann, 1979).
But the interest of management researchers in the topic began only in the mid-1980s with the investigation of interpersonal relationships between buyers and sellers (Dwyer, Schurr & Oh, 1987).
Research on this research topic increased significantly with the academic work of Moorman, Zaltman & Deshpande (1992) and Moorman, Deshpande & Zaltman (1993) on the trust relation between companies and marketing agencies, the work of Morgan and Hunt (1994) and their commitment-trust model in relationship marketing, and that of McAllister (1995) who categorized trust in two dimensions, cognitive and affective.Giddens (1990), as Coleman (1990), Putnam (1993) and Fukuyama (1995), thought that trust is a social necessity in a rationally oriented world and wrote extensively about trust at the social level and in the world in general.
More recently we find further developments in the work of a large number of authors.Mayer, Davis and Schoorman (1995) proposed a dyadic model rooted in crucial definition of trust as "the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party" (Schoorman, 1995: page 712).Another important and close contribution is the one from Rousseau, Sitkin, Burt, and Camerer (1998), who brings an interdisciplinary definition that is rewarding for social scientists: "Trust is a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another" (Camerer, 1998: page 395).These two positions are developed by Dirks and Ferrin (2002) who assumed on their meta-analytic findings the propensity or predisposition to trust.
Seeing the trust as a way of acting, Khodyakov (2007) proposes that "Trust is a process of constant imaginative anticipation of the reliability of the other party's actions" (Khodyakov, 2007: page 126).For their part, Kramer lists seven rules for "moderate trust", ways of helping people to rely more intelligently in our complex world.These rules are, briefly, know yourself, start small, write an escape clause, send strong signals, recognize the other's person dilemma, look at roles as well as people and remain vigilant and always question (Kramer, 2009: pages 74-77).
According to Six (2007: page 292) "If an organization's management wishes to promote interpersonal trust-building in the organization, then a combination of three types of organizational policies can be effective".On the one hand, creating a culture where relationships are important and show care and concern for the needs of the other person is valued (relations-oriented culture).On the other, leading through normative control instead of bureaucratic control because acting properly is the goal of regulatory control.Finally, using explicit socialization to take newcomers understand the values, the principles of the Organization and how to "get things done around here".
Studying how individuals trust their leaders and others, Mishra and Mishra (2008) define the matrix of behavior, ROCC of trust in the leaders, passing through: Reliability, Openness, Competence and Compassion.In the context of Zanini (2007: page 107), "studies on trust relationships have become even more important due to the growing demand for more complex and specialized work, which is based on knowledge".Thus, following Neves (2011: page 44), trust becomes "a strategic issue".That is, trust is crucial for leaders can create more positive relations, "... promote the behaviors they desire in their employees and diagnose and anticipate problems often derive from a lack of trust".Despite all these recent developments, none of these studies specifically has researched the educational context.This is particularly critical, since educational settings are those where character, values and attitudes are built before any other (perhaps with the exception of family).As such, the present article aims to construct the structure of trust-building perceived by principals in Portuguese public schools.We assume also that those contexts are fundamental to understand the followers' perception of their leaders' authenticity.The rest of the article is organized as usually.First, we describe the method we use to obtain data, and then we analyze the data, we draw conclusions on the data and we present the matrix that emerges from the data analysis.Then, with that matrix, we discuss theoretical implications and practices of principals' trust conceptualization and we conclude by showing the most important limitations.Suggestions for future research will also be explored.

Sample and context
Schools are privileged fields for trust analysis because they are in essence spaces of deep relationships at the human and professional levels.The key issue in schools is the type of trust relationship established among professionals, particularly between the principals and the other teachers who help students grow in all their dimensions.This close collaboration relationship and collegiality is built day by day and over time at school.If that complex practice is not positive and constructively experienced, it will be difficult to achieve the mentioned aims.In a context like that, leaders must communicate and pass on values of care, trust and respect.
That is why it is crucial for us to understand which elements make up the trust matrix perceived by school principals.
In order to build a theory on how principals set up their trust matrix, we used a research method developed by Glaser and Strauss (1967) called grounded theory.It is a general methodology for developing theory that is grounded in data systematically gathered and analyzed (Strauss & Corbin, 1993).That theory, consisting of a set of plausible relationships proposed among concepts and sets of concepts, develops and evolves during the research process due to the interplay between data collection and analysis phases.As grounded theory is inductively developed from the phenomenon it represents and theories emerge from data obtained from the phenomenon under study, a research question must be formulated in a form that can be tested by grounded theory itself.The grounded theory does not require an extensive review of the literature, but a generic research of the main theoretical currents on the subject (Locke, 2001).In studies using the grounded theory researchers set out to the field with a generic theme review in study, for the purpose of generating new theory from data observed in real context (Dey, 1999).Otherwise, the investigator is likely to generate obvious theory, or the record of existent literature.The analysis of data collected in research is often referred to as 'coding'.The data analysis process follows a systematic and differentiated hierarchical encoding, the so named grounded theory stages of data analysis -since the lineto-line categorization (open coding), through the establishment of relationships between the categories (axial coding) and ending in the selection of a category -category central -that connects all the other categories (selective encoding) (Strauss & Corbin, 1998, 2007).We combined narrative descriptions with graphics and tabular forms in order to work with the information systematically.Descriptions of the organizations and principals that are included in our sample are presented in Table 1.

School
School type Principals 1

Secondary school including elementary grades
Principal number 1 has been a Computer Science teacher for eighteen years.He took a degree in one of the universities of the region where he was born and lives.He has never worked at another school and has been a principal for 14 years.

Secondary school including elementary grades
Principal number 2 has been a Mathematics teacher for twenty seven years.He took a degree in Civil Engineering.He has been in direction functions for 23 years in two of the five schools where he has worked.

Secondary school
Principal number 3 is a young Economics teacher who has taught for ten years.He has a degree in Management and he has only worked at this school, where he has held the position of principal for 6 years.

4
Secondary school including elementary grades Principal number 4 has been an Accounting teacher for eighteen years and has a Master's degree.He has held this function for 12 years at the school where he has always worked.

5
Elementary school Principal number 5 has been a Portuguese teacher for twenty-five years.He has been the Principal in this school for 10 years but he has worked in another two.

6
Elementary school Principal number 6 has been a History teacher for thirty three years.He has been in direction functions for 20 years in the only school where he has worked.

Secondary school including elementary grades
Principal number 7 has been a Mathematics teacher for thirty four years.He took degree in Mechanical Engineering and has been a principal for 18 years.

Elementary school
Principal number 8 is a Portuguese female teacher, who has been in service for twenty five years.She has run this school for 12 years but has worked in another three schools.

9
Secondary school including elementary grades Principal number 9 has been a Philosophy teacher for thirty-three years.He has been in the function for 10 years in the only school where he has worked.

Secondary school including elementary grades
Principal number 10 has been an Art female teacher for twenty seven years.She has run this school for 10 years.She has a Master's degree in School Administration.

Secondary school including elementary grades
Principal number 11 has been an Economics teacher for thirty three years.She has been a principal for 8 years.She is currently taking a Master's degree in School Administration.

12
Secondary school Principal number 12 has been an English female teacher for twenty seven years.She has run this school for 10 years but has worked in another five schools.
Table 1.Profile of respondents and respective types of school

Data Collection
We  (2008) were written and composed by the principals themselves.Demographically, our principals were 9 men and 3 women, ranging from 33 to 58 years old.The majority were 50 or over.Principals also declared having been 12 years in administration functions on average.

Data analysis
Following the protocol of a grounded theory approach, data were collected as they were analyzed and re-analyzed to make constructs emerge (Eisenhardt, 1989).We often went back and forth in the data analysis, in order to distinguish the constructs from one another other (Locke, 2001).
The process of data analysis can be systematized as follows: first, data were partitioned and examined line by line (Strauss & Corbin, 1998, 2007).Via open coding, we examined the data in order to identify statements which would allow us to understand how principals were building their trust matrix.Following Miles and Huberman (1994), the common statements were coded and generated a pool of first level terms and concepts.Then, the codes were revised to check if each data fragment fitted into each category.Some adjustments were made to obtain more consistent data in line with the codes.Secondly, through a process of axial coding, first level codes were compared to one another in order to identify similarities and differences among them (Strauss & Corbin, 1998, 2007).First level codes, which represented similar ideas, were grouped into a more general and abstract level of conceptual categories, the theoretical dimensions (Locke, 2001).These dimensions represented the different trust behaviors principals adopted.Thirdly, the theoretical dimensions were examined to search for underlying dimensions.Through selective coding, these categories were integrated and refined to outline a framework which took the form of a theory (Strauss & Corbin, 1998, 2007).As we had identified a possible framework, the conformity between the data and our aggregate dimensions were re-examined so that they could be better identified in a model (Glaser & Strauss, 1967;Locke, 2001).A coherent, conforming framework was set up.The aggregate dimensions we arrived at, self and other-trust, emerge from data and help to refine the two basic components of principals' trust.Nevertheless, we used the first order theoretical categories on the construction of our trust matrix; because when we rotated them on our model we find more explainable power of principals' behaviours than with the aggregate theoretical dimensions.

Findings
In this section we explain how the behaviors perceived by the principals were coded and grouped in higher level categories.We start by presenting the theoretical categories and aggregate theoretical dimensions which emerged from the data, and form the principals' trust matrix.This research and the theoretical findings stemming from it may be scarcely representative for the literature and knowledge of the theme, but they correspond to what we found and concluded in this work.In the following pages, we present the theoretical categories, as well as the aggregate dimensions which emerged from our analysis and coding of the data.A synopsis of first level codes and the corresponding theoretical categories is presented in Figure 1.We began by looking for information on what principals perceived as an approach to the confidence construct.One of the principals expressed it as follows: "Trust for me is being confident that others will fulfill their mission.When I don't care about the work done by others, I trust them.But I have to trust myself when I choose the team." What this principal means to say is that the fact that he personally chooses the members of the team he works with, has been a guarantee of success and that leaves him carefree and, above all, confident.In literature, Giddens (1990: page 34) considers that trust is a credit on the reliability of a person, considering the results.Giddens also suggested that trust involves risk: "trust in others is a psychological need of persistent and recurring nature" (Giddens, 1990: page 97).This means trust is a belief that people will do the right things in certain ways at certain moments.For this principal, risk is minimized by his/her self-confidence when it comes to choosing the team.So, trust represents someone's intention to take a risk when relating to others.Nothing is guaranteed.This readiness to trust is defined as the "general willingness to trust others" (Mayer et al., 1995: page 715).
Two other principals raise the question of the personal choice of team members taking into account one of the cross-cutting elements in literature on trust: competence in performance: "Trust is built on the competent performance of the people who are usually part of the teams which support the direction.As I choose them, the process has to work." "Trust is being quite sure that we will not fail.That is, we have to develop skills to avoid mistakes.Throughout these years I have made some decisions and have been successful."Mayer et al. (1995) proposed a model where the trustor determines the reliability of others on the basis of his/her beliefs in the competence of the trustee (knowledge, skills and competences), benevolence (to the extent that the trustor believes that the trustee will act in the best interest of the trustor) and integrity (to the extent that the trustor realizes that the trustee will act according to a set of principles which the trustor considers acceptable).A trust relationship is more likely to develop in the case of the trustee being perceived as having high ability, benevolence and integrity.
Not quite what another principal thinks: "Trust is built by my team.But this function is often lonely and so we have to believe in ourselves and make the choices we believe in." In other words, we cannot always run the risk of staking everything, especially when we are alone, with little support or psychologically vulnerable.Then we have to move on and believe our decisions were the best.We recall what Neves postulates: "ultimately, the basis of trust is the belief that others do not take advantage of us.So, "... not being able to trust others, as a life principle, can bring costs in the medium and long term" (Neves, 2011: page 11).
Principals' statements reviewed so far were aggregated under a category called "Personal Decision" and referred to behaviors in which principals' trust is the power to choose the people who become part of their teams.Principals may be isolated or running risks but they assume being confident in their ability to choose the people who make up their teams because it has produced positive results.
But trust is also placing ourselves in others' boots.So, some principals say: "Trust is putting others in our place.I am not afraid of being removed from the function I have.The others are indispensable." "Trust is accepting others as they are.I have no reason to distrust.Everybody supports me.I am desired." These principals put the basis of their own confidence in others.If these principals for some reason cannot continue, someone else among the others will have to take on their function.
They are aware that their trust comes from believing in the team that they themselves have chosen.They do not fear the future, they even say they are desired and it is the others that are indispensable.Here we find an approach to Mayer, Davis and Schoorman's dyadic model of confidence, rooted in the crucial definition of trust as "the will one party is vulnerable to the actions of another party, based on the expectation that the other party will perform a particular This kind of statement was categorized under "Personal Consideration" and it refers to behaviors by which principals' trust requires commitment from both parties and the principal considers the others who are members of his team, as very important and, in some cases, perfectly indispensable.Trust relationships are so decisive that directors grow and progress considering the importance of others in their own personal performance.
Our data point to a series of statements directly related to principals' competence, knowledge and personal experience.In fact, this is a fundamental feature in the trust-building of some of the principals surveyed.One of them states: "Those who help me to manage the school acknowledge my capacities and competence, which guarantees them a better future." What this principal means is that the future of others depends positively on his ability and competence.Fostering clear self-praise, this principal makes a parallel between the trust issue and interdisciplinary definition, just like Rousseau, Sitkin, Burt and Camerer: "Trust is a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another" (Rousseau et al., 1998: page 395).That level of trust, according to Neves (2011: page 44), is beyond institutional trust because it is at the level of relations between people, namely leaders and collaborators, that trust becomes "a strategic issue".That is, trust is crucial for leaders to be able to create positive relations, "... promote the behaviors they wish in their employees and diagnose and anticipate problems which often stem from a lack of trust".
But not everything was that easy to achieve.Maybe that is why another principal confesses that: "At the beginning it was difficult because there were many doubts, which are no longer there.I am confident because I believe in my experience and knowledge regarding the function I perform." This principal, who had doubts because he was inexperienced when he started, sees trust as a way of acting, just like Khodyakov (2007) who proposes that "trust is a process of constant imaginative anticipation of the reliability of the other party's actions on the basis of (1) the reputation of the partner and actor, (2) the assessment of the current circumstances of the action, (3) assumptions regarding the partner's actions of and (4) the belief in the other party's honesty and morality" (Khodyakov, 2007: page 126).
No wonder there is another principal who assertively states: "I trust my skills and have always relied on my training to perform this function." Ultimately, despite having confidence in his abilities and competencies, this principal requires more security and has sought to improve his performance.The complexity of the demands which a school principal's function involves makes the principal seek to keep up to date.In Zanini's context (2007: page 107), "studies on trust relationships have become even more important due to the growing demand for more complex and specialized work, which is based on knowledge [...]".
This set of positions on the fact that the principal's competence, knowledge and experience are the basis of trust, led us to the category called "Personal Competence".The statements that relate to the existence of positive, imaginative and (in) formed behaviors as a trace of principals' personal trust are in this category.
In addition to these categories which stemmed from the concept of trust perceived and declared by principals, categories related to Compassion also emerged, which are a crucial element of trust, to the extent that it means putting the interests of others at a level identical or superior to ours, which implies using time and acting empathetically (Mishra & Mishra, 2008: page 10).
In our study, Compassion was one of the elements of trust principals gave most importance to in their statements.Principals perceive compassion as a very important constituent of trust behavior and is close to the idea inherent to the structure of advanced leadership for Avolio et al. (2004), in which trust intervenes and is a key variable in the relationship between authentic leadership and the followers' attitudes and behaviors.
So, there is a group of principals who have a clear stand on the role of compassion in their trust relationship with their followers.One of them considers that: "Compassion does not sit well with being a professional." This Director, who had assumed trust as a personal process of choice of competent people for his workforce, does not tolerate faults, errors or lack of rigor in the performances of his followers.As Kramer (2009: page 70), this principal brings out the fact that trust can be a strength or a weakness, depending on whether we are dealing with trustworthy or untrustworthy people.In a world of growing complexity, where we increasingly count on the capacity of experts, it is important to know that it is not enough to rely only on ourselves.And that has to do with demand, rigor and assertiveness.The term and concept assertiveness was popularized to the general public by authors like Alberti andEmmons (1970, 2001) and Smith (1975).They think that assertiveness it's a process of knowing you have a right to be in your place in the world, a right to occupy the space you are in, and a right to get what you want, or at least to negotiate it strongly, in the face of others' demands and expectations that you accommodate their wishes and desires.That sounds a lot like a definition of self-confidence.
Assertiveness and self-confidence are inextricably linked together.Interest towards assertiveness was also maintained in the beginning of the twenty-first century, but it shifted towards the formation of assertive skills in various spheres of public life.Great attention was drawn to the application of assertiveness in pedagogical practice in secondary schools (Christopher, Edwards, & Eppler, 2012).
There's another important aspect to assertiveness, and that's about control.Confidence allows you to exert control over a situation -or at least to exert more control over it.Assertiveness does the same.According to Ames (2009: page 112) "assertiveness is a characterization of how a person responds in a situation in which her positions and/or interests are, or could be, in conflict with others' positions or interests".The author further considered that "Assertiveness may be an essential component of effective leadership and a common -if not leading -culprit when leadership goes awry" (Ames, 2009: page 116).
Therefore, it is not difficult to accept that two other principals are even more assertive: empathic concern are typically considered highly interrelated (Batson, 2009).According to Neff (2003), self-compassion involves "being open to and moved by one's own suffering, experiencing feelings of caring and kindness toward oneself, taking an understanding, nonjudgmental attitude toward one's inadequacies and failures, and recognizing that one's experience is part of the common human experience" (Neff, 2003: page 224).Compassion is defined as feeling sorrow or concern for others along with a desire to alleviate their suffering (Goetz, Keltner & Simon-Thomas, 2010).Thus, it should be noted that some of the principals stated: "Being a principal requires a lot of understanding of the human, fallible side of others." "After hard times, I believe that everybody makes mistakes." These concerned principals assume that, as making mistakes is a human characteristic, assuming responsibility as a principal is above all accepting and adapting to people's weaknesses.Other directors also appealed for understanding: "Accepting people is to trust the good side that defines us." "People are very important for me." According to Six (2007: page 292), if the administration of an organization wishes to promote the construction of interpersonal trust in the organization, then the combination of three types of organizational policies can be effective.On the one hand, by creating a culture where relationships are important and where showing care and concern for the needs of others is valued (relation-oriented culture); on the other hand, leading through normative control instead of bureaucratic control, because acting adequately is the goal of regulatory control.
Finally, it is necessary to use explicit socialization to make newcomers understand the values and the principles of the organization and how to "get things done around here".
This set of statements by the principals was categorized as "Compassionateness" and considers compassion behaviors, which is a crucial element of trust, a synonym of camaraderie, benevolence and naivety.
Finally, almost residual, there were principals who assumed compassion but ignored its complexity.This ultimately determines some kind of indecision, absence or ignorance regarding the importance of trust as an organizational dimension.
In connection with compassion two of the directors claim that: "Sometimes I feel sorry for some teachers namely fixed-term teachers." "When someone goes wrong, I feel sorry because it could have been avoided." These principals clearly mix up compassion and pity, which cannot be in keeping with the trust which they had considered as a personal decision in the choice of the members of their teams.
We point out that this position is not even close to benevolence (to the extent that the trustor believes that the trustee will act in the best interest of the trustor) stated by Mayer et al. (1995).They will hardly agree with Avolio and Gardner (2005: page 326) who argue that "if leaders demonstrate impartial treatment of information, personal integrity and authentic relational motivation, leader-follower relations will produce high levels of respect, friendship and confidence".
The set of positions in which compassion, a crucial element of confidence, is synonymous with pity behaviors is categorized as "Absence".
After grouping the first-order codes in theoretical categories, we conducted an axial coding for these categories and grouped them into two main aggregate theoretical dimensions (Figure 1).
These were: Self-Trust and Other-Trust.The first aggregate theoretical dimension includes the Personal Decision, Personal Consideration and Personal Competence categories.This dimension was based on the similar reference of all categories to behaviors centered on the personal dimension of trust.
The second dimension, Other-Trust, includes the categories of Assertiveness, Compassionateness and Absence.All these categories have the consideration of compassion as a crucial element of trust between the leader and his followers in common, although in one of them, which is residual, the ignorance of the complex dimension of compassion in the construction of trust relationships was clear.
Finally, as far as the other elements of trust mentioned by Mishra and Mishra (2008) above are concerned, no evidence emerged of any theoretical category.We can clearly observe that reliability is a quasi-static antecedent of trust, to the extent that, as it is not taken for granted, it is either reduced to (in) existence or it is built over time.As to Openness, it is commonly mistaken for the circulation of communication, an extremely acute problem in schools because the abundance, accumulation and erosion of information and legislation mark daily life.Finally, for most respondents, Competence is the result of "position", i.e., competence is reduced to the belief that people acquire the skills required for the job by performing that job.Whatever training people may amass, it is on in situ updating that principals have been focusing.

The trust matrix perceived by principals
We will now present the principals' trust matrix built on the theory that emerged from the data collected and analyzed by using the grounded theory approach to data analysis.
-635-Journal of Industrial Engineering and Management -http://dx.doi.org/10.3926/jiem.1046 As far as the aggregate theoretical dimension Self-Trust is concerned, it seemed consensual to consider that it corresponded to an individual attitude characteristic of public school principals.
Basically it is about how much each one of us believes in himself/herself.For Covey and Merril (2006) it is the first wave to be generated in the process of building confidence: "The first wave, Self-Trust, is about the confidence we have in ourselves-in our ability to set and achieve goals, to uphold commitments, to walk our talk-and also our ability to inspire trust in others.The whole idea is to become a person who is worthy of trust both for ourselves and others.The key principle underlying this wave is credibility, which comes from the Latin root, credere, meaning "to believe" (Covey & Merril, 2006: page 44).
In reference to the aggregate dimension Other-Trust, we decided to go back and analyze the initial categories and revisit some of the concepts that emerged, stronger explaining principals' behaviors and that could be read in the light of other theoretical positions on trust contained in literature and, more broadly, on leadership.
Thus, at the intersection of the theoretical structure of cultural values (Schwartz, 1999  those who perform it.Taking into account the construct created by Avolio and colleagues since the beginning of the 21st century, these can only be called Authentic principals.

Discussion
According to the study inductive nature and utilized methodology we assume not to formulate any starting hypotheses.Thus, the development of theory results from an interactive process for the collection, coding, analysis, and planning the steps to follow in the study, in the sense that the emerging results will direct the next data collection, a process that became known as the theoretical sampling (Glaser & Strauss, 1967).As emphasized by the authors, this process requires, on the part of the investigator, a theoretical sensitivity along the collection and codification of the data, so that this go realizing where the data are taking and what he should do next (Bowen, 2006).This theoretical sensitivity is a central aspect in grounded theory and reflects the researchers' ability to use their personal and professional experiences, as well as its methodological knowledge in the service of data interpretation, developing the insight with which a researcher comes to the research situation.The aim of this study was to construct the structure of trust-building perceived by principals in Portuguese public schools.We were able to draw this matrix by identifying the two dimensions that explain the various principals' behaviors in the trust relationship they established with their followers.The identification of these two dimensions and their quadrants are the main contributions of this article.There have been, until now, no researchers who had committed themselves to the study of the dimensions of trust relations between school principals and teachers.According to our findings, the two dimensions, although contrasting, are not fully independent from each other.This has been highlighted with the creation of the name for the missing quadrant and where the symbiosis of empathy and the demand and rigor of the task is established.
Another major contribution of this study refers to the current debate of the effectiveness of "pure" positive behaviors and psychological states for leadership in organizations.As some have concluded, reality in work settings is far more complex than simply seeing things as positive or negative (Lopes, Cunha & Rego, 2011).In the same line of reasoning, the present study found that "pure" positive behaviors, as being compassionate, might need to be tempered or complemented with other behaviors such as assertiveness, to become really effective.
The major limitation of this study was the impossibility to conduct face-to-face interviews given the limitation of free time alleged by the principals.However, the conclusions, in spite of the reserve of representativeness due to the fact that the data were collected among a small, specific population, can and should be regarded as characterizing elements of a trust matrix possible to apply to other sets of public school principals.One of the clues for future research is to enlarge the basis of data collection and, through the method used here, test the quality of the findings.

Practical implications
As pointed out by several authors (e.g.Miles & Huberman, 1994), we took care to return the model to a sample of respondents or similar professionals, and get them the model validation makes sense.
But we are aware that for Glaser and Strauss (1967), studies using the grounded theory that feature higher quality are those that produce "good theory", which the authors define as adjusted, emerging categories are set and explain the data collected, relevant, dealing with the real concern of participants, workable, can be applied to many situations in the study area that seek to explain, and modifiable, changing in the light of the changes that reality is being targeted on a daily basis.
We can consider similarly that grounded theory, it is claimed, is a theory which is inductively derived from the phenomenon it represents and meets four central criteria: fit, understanding, generality and control (Strauss & Corbin 1990).Fit entails that the theory fits the substantive data.Understanding entails that the theory be comprehensible to all involved in the area of study.Generality entails that the theory is applicable in a variety of contexts.Control implies that the theory should provide control with regard to action toward the phenomenon.
Finally, other authors consider that the quality of the theory generated by grounded theory can be evaluated by the evaluation criteria of qualitative studies: trust (trustworthiness), credibility, transferability, dependability and reliability (Lincoln & Guba, 1985).
We believe that this study means the beginning of the use of a trust building matrix on the part of the principals to improve relationships with their followers.Everyone can gain by knowing the importance of compassionateness and assertiveness in daily relationships.Above all, it is important that principals and teachers have access to this research, to the extent that we can only change our practices and improve the way we lead and are led if we reflect on the theories that are founded in the professionals' statements.
Principals' behaviors will surely be perceived differently by the principals themselves.What happens in the difficult practices of the direction of schools needs to be put into perspective, not in the sense of a penalty but on the path to improvement.The consideration of a symbiosis category of the behaviors of principals can trigger positive expectations among the players who until now did not understand how it is possible to bring together the best of the worlds of management and leadership in a balanced way, with efficiency in authenticity.
based our research on testimonies of elementary and secondary public school principals on the North coast of Portugal (n = 12), in forty four schools in the Viana do Castelo, Braga and Porto regions.The selection was casual or out of convenience, to the extent that the respondents were those who were available for the purpose and the data were those we had access to.The region where the data were collected, from Vila do Conde to Valença, was defined according to geographical proximity and also the coherence with two other studies by the same author.Surveys were carried out between June and July 2013 and the questionnaires were collected by personal contact.The answers to an open-ended question survey based on ROCC typology (Reliability, Openness, Competence and Compassion) by Mishra and Mishra

Figure 1 .
Figure 1.First order codes, theoretical categories and aggregate theoretical dimensions : page 29) and the Organizational Values Inventory by Tamayo, Mendes & Torres (2000) and by using Covey and Merrill's smart trust matrix (2006), we decided to build the trust matrix perceived by the principals who accepted our invitation to take part in this study on trust.Such matrix is drawn in Figure 2.

Figure 2 .
Figure 2. Matrix of principals' perceived trust action, which is important for the trusting part, regardless of the ability to monitor or control the other party"(Mayers et al., 1995: page 712).This definition is confirmed by another Walumbwa, Luthans and May (2004)when I deliver my fate in the hands of others.I do not fear because I will receive in the same measure as I give.I believe my colleagues are good people."Thisdependencerelationshipleadsus to the position taken by Avolio Gardner,Walumbwa, Luthans and May (2004)who contend that trust intervenes in the leadership structure and is a key variable in the relationship between authentic leadership and the attitudes and behaviours of the followers.Another principal state, referring to his own capacity to reflect this relationship in the document which defines his application for the principal's function: "Knowing myself, I invest in relationships.My intervention project shows I am a person who trusts others.Generally, I admit that you can only try to cheat on me if you do not know me."