School and District Leaders Talk about Teacher Attrition

Teacher attrition has become a challenge in many educational systems worldwide. Many studies have focused on teachers' perspectives, while attempting to identify the factors that motivated teachers' decision to leave the profession. The present study aimed to explore teacher attrition from the point of view of school leaders principals and inspectors. Using qualitative and quantitative methods, the results indicate that school and district leaders perceived teacher attrition via two-dimensional structure, including explicit and implicit dimensions. The explicit dimension refers to the act of leaving. The participants indicated that the main reasons that motivate teachers' decision to leave the profession are related to the stressful working environment and poor job conditions. The implicit dimension presents a hidden attrition. Based on cost-benefit theory the study highlights the complex structure of teacher attrition.


Introduction
Teacher attrition has long been identified as a challenge for educational systems worldwide. There is abundant data in the international literature showing that a large proportion of beginning teachers, between 20% and 50%, leave the profession within the first five years (European Union, 2013;Geiger & Pivovarova, 2018;Ingersoll, 2003). Although, many studies have been conducted focusing on early-stage teachers, Graham, Hudson and Willis (2014) demonstrated that the relationship between work experience and attrition evinced a U shape pattern, meaning that also senior teachers tend to leave the teaching profession to a large extent.
A high level of teacher attrition negatively affects school routine and management. It harms the efforts to construct a solid organisational culture and to maintain staff solidarity and satisfaction (Hanselman, Grigg, & Bruch, 2014), and it harms student achievement (Ronfeldt, Loeb, & Wyckoff, 2013). Moreover, teacher attrition means loosing human capital and dealing with teacher shortages (Ingersoll & May, 2012). Replacing teachers is not always an easy task, and sometimes principals are forced to employ unqualified teachers (Changying, 2007;Donitsa-Schmidt & Zuzovsky, 2014). Teacher attrition is also costly. In the U.S.A., for example, state expenditures on teacher attrition cost on average US$1 billion annually (Haynes, 2014).
According to the cost-benefit theory, an optimal level of attrition in organisations is beneficial (Burkhauser, 2017;Siebert & Zubanov, 2009). A particular benefit of attrition is replacement of poorly performing individuals with others who may be better suited for the job (Ableson & Baysinger, 1984). In this way, employee turnover maximises the organisation's productivity, and sets the basis for promoting conceptual innovations. This is also true in the teaching context, provided that the professional qualities of the new coming teachers exceed those of the dropouts (Ronfeldt et al., 2013).
A significant body of literature that aimed to understand the factors and conditions that contribute to teacher attrition has focused on the teacher's perspective (e.g. Borman & Dawling, 2008;Hong, 2012;Struyven & Vanthournout, 2014). These studies mentioned poor job conditions (low salary, part time position), challenging organisational aspects (heavy workloads, large classes, lack of autonomy, lack of opportunities to participate in decision making), non-supportive resources (lack of collegiality and collaboration, difficulties in managing social relations) and difficulties in class management (student behavioural problems, low student engagement) as determinants that motivate the decision to leave the teaching profession. Towers and Maguire (2017) also cited professional identity crisis, which prompts teachers to drop out. Accordingly, a low sense of belonging, dissatisfaction, lack of efficacy, and limited opportunities for professional development were found correlate with teacher attrition. Also personal resources (e.g. human capital, experience) were associated with an increased likelihood of leaving the profession (Arviv Elyashiv & Navon, 2018;Borman & Dowling, 2008). Drawing on data from Belgium, Dupriez, Delvaux and Lothaire (2016) claimed that advanced credentials provide teachers with wider qualifications and recognition in other job markets. As a result, highly-educated teachers tend to leave the profession in the early stages.
Fewer researchers have studied teacher attrition from the perspective of schools and district leaders, despite the fact that principals and inspectors play a key role in determining teachers' satisfaction and career decisions. Nevertheless, their perspectives concerning teacher attrition have rarely been explored. The few existing studies have focused on the principal's influence on teachers' job satisfaction (Burkhauser, 2017;Cobb, 2015;Duyar, Gumus, & Bellibas, 2013;Shen, Leslie, Spybrook, & Ma, 2012) and resilience (Peters & Pearce, 2012). These studies provide evidence that a principal whose pedagogic leadership is based on a democratic school environment, a collegial partnerships, and professional support is associated with teachers' high levels of job satisfaction and commitment.
By contrast, the influence of the district inspector has hardly been mentioned in the professional literature concerning teacher attrition. Arviv Elyashiv & Zimmerman (2013) found that inspectors' support also decreased teachers' decision to drop out, although it should be noted that such support is rarely available to teachers. Findings of that study (which involved interviewing 40 dropout teachers) indicated that the inspectors were perceived as indifferent and unresponsive to the teachers' needs and requests.
Theories of emotional intelligence assume that involving emotions explicitly through leadership practices may increase employee effectiveness and job satisfaction (Beatty, 2000). Crawford (2007) examined leadership practices in schools in Britain. She claimed that the use of the language of emotions is a significant leadership resource at school. It builds trust, collaboration and support, and thus increases teachers' sense of belonging and their tendency to persist in the teaching profession. Indeed, some studies have shown that teachers who have the principal's support are less likely to drop out (Margolis, 2008;Skaalvik & Skaalvik, 2010). This support involves open communication, assistance, and consideration of personal and professional needs. As mentioned, principals and inspectors' perceptions regarding teacher attrition have received limited attention in the research literature. The present study seeks to shed light on this aspect.

The Research Context
Teaching in Israel is a female-dominated occupation in which women constitute approximately 75% of the teaching force (Israel Central Bureau of Statistics (ICBS), 2013). Most of the teachers are employed by the state and usually are awarded tenure after 3 years, after which they cannot easily be dismissed. Teachers in Israel are represented by organised unions that manage their status and make efforts to protect teachers' rights and improve their working conditions. Teachers' salaries are lower than the average teacher salary in many OECD countries (2017) and also lower than the salaries of other professionals in the Israeli labour market (ICBS, 2018).
Teaching offers moderate levels of autonomy and participation in decision-making (Arviv Elyashiv & Zimmerman, 2015), but offers only a limited path to promotion (Avidav-Ungar & Arviv Elyashiv, 2018). Once every seven years, a teacher is entitled to a year of vacation defined as a sabbatical year. During this year, teachers are expected to devote time to professional development, at the expense of an allocated study fund. In most cases, teachers receive a salary during the sabbatical year.
Every year, approximately 6,000 novice teachers complete teacher-training programmes; 75% of them then enter the educational system (ICBS, 2018). After a period of five years, a third of them leave the profession. Afterward, attrition rates significantly decrease.
Policy makers in the Ministry of Education are very concerned with the high attrition rate among novice teachers. Therefore, in the last decade they have promoted new initiatives in order to prevent many beginning teachers to leave the profession. One of the recent initiatives in this regard includes a grant awarded to school principals that have succeed to design an effective absorption program aimed to integrate novice teachers into school's routine and culture. The award committee's decision is based on the district inspector's recommendation. In the last three years, 40 principals in average were awarded with this prestigious grant. This initiative was based on the assumption that the involvement and support of school leaders, principles and inspectors, in such programs, positively contribute to teacher retention, and eventually reduce attrition (Zimmerman & Zilbershtrom, 2015). Therefore, in the first step it is necessary to understand the perceptions of school principals and district inspectors regarding teacher attrition, the challenges they are facing while dealing with this phenomenon and the steps they believe are needed to reduce its frequency. This study aimed to investigate these aspects.

Research Questions
How do school principals and district inspectors define teacher attrition?
What do school principals and district inspectors see as the challenges related to teacher attrition?
What can be done in order to reduce teacher attrition, according to school principals and district inspectors?

Methods
The study was based on a mix-method approach, using both qualitative and quantitative research designs. The first research question was addressed by conducting a qualitative exploration using semi-structured interviews. The other two research questions were examined by employing a quantitative survey. Following is a description of each of the research designs.

Participants
Eight principals (4 men and 4 women) and five inspectors (all women) were interviewed. They were recruited through a "snow ball" sampling. Three were principals of elementary schools and the other five managed secondary schools. As for the inspectors, three participants were responsible for schools located in the country's central inspectorate region, and the other two worked in the peripheral regions. Most of the participants (7 principles and 4 inspectors) worked in the secular Jewish sector. One principal and one inspector represented the Arab sector.

Research Instrument
A semi-structured interview was used to collect data. The six questions that guided the interview concerned participants' attitudes and experiences regarding teacher attrition, documentation procedures, and the teachers' supporting procedures. Each interview lasted between 30 and 40 minutes.

Data Analysis
Data were analysed using an inductive approach. In the first stage, responses were divided into topics. Answers on similar topics were converged into categories that represented the central themes of the interviews (Maykut & Morenouse, 1994).

Participants
School principals: Participants were recruited using a representative sample of 558 Israeli schools (excluding ultra-Orthodox schools). The sampling design followed the rules of the sequential random sampling procedure, stratified by the inspectorate bodies (Jewish-secular, Jewish-religious and Arab), the level of schooling (elementary, secondary), the schools' geographic regions (seven regions), and the school size.
An anonymous questionnaire was sent to these principals following an online request or a telephone call. A total of 126 principals (63 women) responded to the research questionnaire (23% response rate). Although this is a low response rate, the distribution of the schools whose principals completed the research questionnaire approximately corresponds to the sectorial distribution of schools in the general population: 75 of the schools were of the Jewish-secular sector (60%), 16 schools were Jewish-religious schools, (13%, this group is slightly under-represented), and 35 Arab-sector schools (27%). Eighty six participants (68%) were elementary school principals and 40 participants were secondary school principals.
District inspectors: Data were collected among all inspectors in the seven geographic regions via the Ministry of Education. A total of 74 inspectors (42 women) serving in different roles responded to an anonymous questionnaire, representing all seven regions.

Research Instrument
Two questionnaires were developed, one for principals and one for the inspectors. Most of the questions were identical. The questionnaires included: 1. Likert-type questions (12 for the principals and 9 for the inspectors) on a five-point scale (1= not at all agree and 5 = agree very much), 2. Open-ended questions (3 for the principals and 2 for the inspectors), 3. Background information. The questions referred to the respondents' involvement in dropout occurrences among teachers, documentation procedures, and attitudes towards teacher attrition. The questionnaire was tested in a focus group comprised of two principals and two senior researchers.

Data Analysis
The quantitative data were analysed using tools of descriptive statistics, t-tests and  2 tests. The open-ended questions were analysed using an inductive approach.

Principals and Inspectors Identify Teacher Attrition
The research interviews aimed to explore how the participants, principals and inspectors, identify teacher attrition. Their responses highlighted the multi-layered classification of this phenomenon, clarifying its complexity and its trends.
Inspectors identified teacher attrition as dropout behaviour, meaning, they equate attrition with the phenomenon of teachers who decide to leave the teaching profession permanently. As they addressed the issue, they drew a distinction between two groups of teachers: veteran teachers and beginning teachers. Their responses verified that attrition decisions in both groups arise from different factors experienced at different stages along the professional track. Attrition among veteran teachers is perceived as a result of a deep sense of burnout and stress, which increases over the years. These feelings make the teaching tasks and the classroom encounters a heavy burden, and in many cases, they can even lead to health problems. Consequently, their teaching becomes ineffective and unproductive; eventually many of these teachers decide to leave the profession.
For example, Chen, an experienced inspector said: I have had many meetings with veteran teachers that wanted to stop teaching. They said they could not stay at school any longer. Most of them complained about health problems. They talked of physical and emotional burnout.
The inspectors specified that attrition among veteran teachers is an ongoing process, in which each year teachers are constantly searching for other alternatives in order to reduce teaching hours; sometimes they prefer to take time off, or even an unpaid long vacation.
Veteran teachers usually do not leave teaching unexpectedly. First, they take a sabbatical year, then they decide to take an unpaid vacation for another year, sometimes they even extend it for a second year. Afterwards they are not able to come back… Alternatively, attrition among novice teachers is perceived as a product of poor integration into the school organisation and culture. Novice teachers often face many challenges in the first years of teaching. Failure to provide them with a supportive environment while they are facing a heavy workload and difficult classes may result in dropout behaviour. Gili said: 'Dropping out in the first three years of teaching is a result of an unbefitting absorption process in school. This happens when teachers fail to manage the difficulties of the beginning stage.' Also principals referred to teacher attrition as a dropout behaviour. They explained that teachers decide to leave the profession because they are dissatisfied with the work environment and job conditions. They agreed that these occurrences are more frequent among beginning teachers. Yosi, an elementary-school principal said: Dropout teachers, especially novice teachers, had a great deal of trouble dealing with classroom discipline problems, and the heavy workload. Novice teachers always feel very stressed, but the poor job conditions is the final stroke In the interviews, the principals added another aspect of teacher attrition. This aspect, which is less familiar, refers to hidden attrition. This concept was explained by the principals as professional incompatibility of teachers to perform effectively; nevertheless they choose to stay in the educational system. These teachers become hidden dropouts, According to the principals, these teachers are incompetent. They are characterised with low level of teaching qualifications or inappropriate social skills. Principals said that the absence of authority to dismiss teachers, obligates them to carefully manoeuvre between providing these teachers with a suitable position and minimising the damage caused by their poor professional performance. Therefore, as principals, they constantly need to find creative solutions for these teachers. The participants mentioned some ideas, such as reducing the number of teaching hours, providing alternative positions at school, recommending a vacation (sabbatical year or unpaid vacation) or a move to another school. They claim that these situations usually have a more negative impact on the school climate than when teachers decide to leave the profession.
Tiki, a secondary-school principal said: Around 20 to 30 percent of the teachers are incompetent. The problem with these teachers is that we cannot dismiss them. We are obligated to find them a proper position at school at all costs.
Avi, a secondary-school principal added: Some teachers do not formally drop out. It is impossible to dismiss teachers, even if the teacher is not good. I had some experiences with incompetent teachers [and] I had to make all kinds of arrangements to find them a more suitable role so as to prevent unfortunate incidents in the classroom. I was pleased if one of them moved to another school; it is the only way to replace them.
Both inspectors and principals agreed that teachers' requests to leave the profession should be respected, even if they are good teachers. During the interviews, the participants told about cases of teachers who expressed a profound desire to leave school and turn to a new occupational path. Some of the participants said they convinced these teachers to continue teaching at the school; however, they admitted it was a mistake. They explained that in these cases, leaving the teaching profession is usually inevitable. The delay only creates difficulty for all concerned, but especially for the teachers. It only intensifies their distress and burnout, and finally they become hidden dropouts. Amar, a secondary school principal in the Arab sector, for example, spoke about one of these teachers: One of my teachers once told me he wanted to leave. He was a good teacher and I admired his work. I pressed him and he stayed. At the end of the year, I regretted it. This year he did not perform properly. He came to work, but his heart wasn't in it. Now I know that when someone really wants to leave, there is no point blocking the door; the teacher will find it regardless.

Challenges of Teachers' Attrition
The survey questionnaire examined principals and inspectors' views regarding the challenges of teacher attrition. Two main challenges were recognised: managing teacher personnel and supporting teachers who have expressed a desire to drop out.
As for the first challenge -managing teacher personnel, Table 1 shows that less than half of both principals and inspectors considered dropouts to be good and effective teachers (14.7% of the inspectors compared to 43.1% of the principals). Moreover, in another question that was addressed only to the principals, they claimed that this dropout behaviour affects the learning performance at the school to a moderate extent (M = 3.48, SD=1.18). Principals' responses to the open-ended questions clarify and elaborate these results. Their responses (n = 41, 65% of the responses to open-ended questions) indicated that in many cases teacher attrition is desirable and necessary. These cases refer to unmotivated teachers, who do not fulfil their duties properly. Hence, attrition is perceived as the right solution. The principals admitted that these cases make it possible to employ more motivated and productive teachers. Examples of principals' responses: 'A frustrated teacher who does not do his work effectively makes the class uneasy and restless' and 'A teacher who does not do her job properly creates a negative environment and harms the professional relationships and collaborations among teachers'.
Obviously, principals' responses (n = 45, 71% of the responses to open-ended questions) also indicated that attrition among good and effective teachers is undesirable. They mentioned that losing good teachers might cause negative consequences for teaching quality and students' achievements. For example: 'The loss of good teachers produces a loss of knowledge, a negative atmosphere in the teachers' room, and a negative impact on students' wellbeing and achievement' and 'Dropout of high-quality teachers creates a need to compromise by hiring mediocre teachers whose quality of instruction is lower'.
Furthermore, as shown in Table 1, a minority of both the principals and the inspectors thought that teacher attrition produces teacher shortages. The principals reported that they were able to fill the position of teachers who had dropped out within a reasonable time period and more than a third (39.4%) stated that they had been able to recruit replacements with similar expertise within a month. Nearly half (43.4%) of the principals claimed that this recruitment process was completed within 1-6 months. Only 17.1% of the principals stated that finding replacements for the dropout teachers was a difficult task.
Another challenge of teacher attrition is supporting teachers who have expressed the desire to leave the teaching profession. The findings are presented in table 2.  Table 2 indicates that only a quarter of the principals and inspectors agreed that clear procedures to deal with teacher attrition are available. However, according to their responses, these procedures are needed only to a moderate extent (principals: M = 3.47, inspectors: M = 3.31). Half of the participants mentioned that it is necessary to develop clear procedures in order to address these teachers' needs.
The results also indicate that principals more than inspectors tend to support teachers that have decided to drop out. As shown in Table 2, 71% of the principals, compared to only 37% of the inspectors, mentioned that they support these teachers and pay attention to their needs. Moreover, a larger proportion of principals than of inspectors engage in a personal conversation with teachers who have expressed the wish to dropout, (97% vs. 82% of the inspectors,), document the relevant events and actions (57% vs. 22% of the inspectors), and inform their supervisors (63% vs. 16% of the inspectors).

Reducing Teacher Attrition
Principals and inspectors responded to the following open-ended question: what should be done to reduce teacher attrition. Four categories were emerged from the responses of 90 principals and 41 inspectors. Table 3 presents these categories. Provide mentoring and support 24% 54% Offer to assist dropout teachers 12% 12% As can be seen in Table 3, in the responses to this open-ended question, the strategy most frequently mentioned by the principals for coping with the phenomenon of teacher attrition was to improve occupational conditions. Sixty percent of the principals, compared to only one fifth of the inspectors, suggested this strategy. In this manner, the respondents emphasised the need to increase teachers' salary, reduce working hours, and improve teachers' status.
of the inspectors' responses suggested providing guidance and personal coaching, as well as empathic environment, to assist teachers who had experienced difficulties and burnout. This strategy was mentioned less frequently by the principals (24% of the responses).
Another strategy offered by the principals (29% of the responses) was to improve the workplace environment in the schools, for instance, by reducing the number of students in a class, granting autonomy to teachers, including teachers in school-related decision-making, and creating opportunities for professional development. The inspectors rarely referred to this issue. Both principals and inspectors advised the development of organisational procedures to assist teachers who have expressed a desire to drop out. Yet, they stressed that these procedures should be implemented only for effective teachers.

Discussion
Teacher attrition has become a challenge in many educational systems worldwide. The analysis of this phenomenon has been the subject of recent studies, most of which focused on the teachers' perspective, while attempting to identify the factors that motivated their decision to leave the profession. The present study aimed to explore teacher attrition from the point of view of the direct supervisors of the schools, i.e. the principals and inspectors. Although, school and district leaders have a significant role in determining teachers' satisfaction, professional development, and career decisions, their perspectives regarding teacher attrition have not been sufficiently investigated. Three questions were posed: 1. How do principals and inspectors identify teacher attrition? 2. What are the challenges of teachers' attrition? 3. What can be done to reduce teacher attrition?
Using qualitative and quantitative methods, the results indicate that school and district leaders perceived teacher attrition via a two-dimensional structure, including explicit and implicit dimensions.
The explicit dimension presents the formal aspect of teacher attrition. This dimension refers to the act of leaving the profession. This aspect has been examined extensively in the professional literature. In this manner, principals and inspectors indicated that the main factors that motivate teachers' decision to leave the profession relate to a stressful work environment and poor job conditions. Consequently, these teachers experience professional deficiency. It gradually crushes feelings of meaningfulness and commitment (Hackman & Oldham, 1980), and it generates the mechanisms that explicitly drive attrition. Novice teachers are more vulnerable to these circumstances than veteran teachers, mostly because they consider the conditions of a teaching career in comparison to other occupational alternatives (Arviv Elyashiv & Zimmerman, 2015, Dupriez, et al., 2016.
How do school and district leaders deal with teacher attrition challenges? Both principals and inspectors reported that there are no established and clear procedures for dealing with teacher attrition. Indeed, a review of the Ministry of Education's instructions from 1998 to 2018 indicates the lack of a clear and formal policy or procedures for managing teacher attrition. Yet, many of the participants agreed that these kind of procedures are not really necessary. Hence, given that managing teacher attrition is not guided by a uniform policy or clear procedures, it is managed at the local level and procedures are determined by personal attitudes, preferences, and interpersonal relationships at the school or district level. In practice, the most frequent tool that most principals and inspectors use is a having a one-on-one personal conversation with the potential dropout. This instrument helps clarify the teacher's particular needs and difficulties, and constitutes an opportunity to rethink the decision to leave the teaching profession. Principals also tend to document these occurrences and inform their immediate superiors. Inspectors rarely implement the latter procedures.
It is worth mentioning that the principals and inspectors agreed that the teacher attrition rarely leads to a shortage of teachers. In many cases, they even agreed with an individual teacher's decision to drop out and viewed it as an opportunity to employ more motivated and productive teachers. They reported that they were able to fill the position of teachers who dropped out within a reasonable time period.
Two major steps were suggested in order to reduce the explicit act of dropout among teachers: improving occupational conditions and increase mentoring and supportive initiatives. Improving the occupational conditions of teachers is not in the purview of either principals or inspectors. The Ministry of Education, in collaboration with the Ministry of Finance and the teachers' unions, determines these terms. Recent reforms in the Israeli education system, titled 'Ofek Hadash' (New Horizon) and 'Oz Latmura' (The Courage to Change), aimed to promote a fundamental change in teachers' job rewards and to forge new paths for professional developments. The consequences of these reforms in terms of occupational conditions and teachers' satisfaction have been partially examined (National Authority for Measurement and Evaluation, 2010;Oplatka, 2010). These studies demonstrated that the reforms led to a slight improvement in teachers' salaries. Veteran teachers enjoyed these improvements more than did novice teachers. Novice teachers' salaries remained low compared to the national average salary. Further research is needed, particularly to examine the correlation between theses improvements and teacher attrition.
The second suggestion, to increase mentoring and supportive initiatives available to teachers, is imperative for decreasing dropout behaviour among novice teachers, but is important for veteran teachers as well. Based on the theories of emotional intelligence, using the 'language of emotions' may provide a method to support both novice and veteran teachers. These methods have positive benefits for strengthening teachers' efficacy and satisfaction (Beatty, 2000). They also increase teachers' sense of belonging and the desire to persist in the educational system (Crawford, 2007). As for novice teachers, internship programmes usually involve a personal mentoring process that supports their socialisation, professionally and emotionally, as they take their first steps as teachers. These programmes have been expanded in the recent years, aiming to introduce novice teachers to the professional culture and the specifics of the professional framework (Hobson, Ashby, Malderez, & Tomlinson, 2009), and to reduce the likelihood that they will decide to leave the teaching profession (Arviv Elyshiv & Zimmerman, 2015).
The implicit dimension presents the informal aspect of teacher attrition. This dimension refers to hidden attrition, and is a less familiar topic in the professional literature. The participants describe what I have labelled implicit attrition -as the phenomenon whereby teachers who exhibit professionally incompetent behaviour and ineffective performance opt to remain in-rather than drop out of-the profession. They choose to hold their position at school and stay in the educational system despite the possibility that they are aware of their performance level and even feel frustrated by it. In Israel, once tenure is granted, usually the decision whether to stay or drop out is in the hands of the teachers. School and district leaders explained the difficulty of managing these situations, as they do not have the authority to dismiss teachers with tenure, even incompetent teachers. Formally, these teachers are members of the teaching staff, but practically, they have a hard time getting involved in the school's daily routines in a positive manner, collaborating with their colleagues, and engaging in interpersonal interactions with students. In these situations, the solution involves creating specific arrangements at the school level, to keep these teachers out of the classroom. Eventually, they become hidden dropouts.

Conclusion
The dual structure of teacher attrition highlights its complicity and challenges. This structure aligns with the cost-benefit theory (Ableson & Baysinger, 1984). Our results indicate that principals and inspectors adopt economic approach while addressing teacher attrition. This approach involves a constant calculation of the balance between performance and attrition (Hancock, Allen, & Soelberg, 2017). It distinguishes between attrition that is detrimental and attrition that is beneficial to the system. When explicit attrition characterises teachers who up to a certain point in time had demonstrated strong motivation, well-honed pedagogical skills, and positive interpersonal relationships, this case of attrition is viewed as detrimental; it is costly and undesirable, as it involves losing valuable human capital and expertise, as well as formal and tacit knowledge (Kacmar, Andrews, Van-Rooy, Steilberg, & Cerrone, 2006). This type of attrition is a symptom of the system's dysfunction, and principals and inspectors alike seek to reduce its occurrence. Nevertheless, both leaders conveyed the understanding that encouraging these teachers to remain in the educational system despite their expressed wish might lead to an even less desirable outcome. It may intensify distress and burnout, and finally it could produce hidden dropouts. Hence, this type of attrition is detrimental to the system.
By contrast, when the phenomenon of explicit attrition occurs among ineffective teachers with poor pedagogical skills, it is considered beneficial to the system. This situation provides an opportunity to improve job matching (Ronfeldt et al., 2013). Replacing these teachers is a desirable outcome; otherwise, it too might produce implicit attrition, which has a negative impact on school environment and harms the learning climate and students' outcomes. It is better if incapable teacher leave the profession voluntarily. Nevertheless, in quite often this is not the case. Many of these teachers decide to stay. This decision is detrimental to school management, yet principals and inspectors have no recourse to prevent it. Table 4 summary of two-dimensional structure of attrition. The unrecognized status of hidden attrition among teachers generates a vicious circle that reinforces the frequency and intensity of these events. School leaders usually try to take responsibility, but their efforts focus only on the local level. Consequently, this problem still prevails. Policy makers should be made aware of the implicit dimension of teacher attrition and its negative impact, and design formal policies to enable school leaders to take the necessary steps towards minimizing or even eliminating it.
The current study contributes to the professional literature by exemplifying school perspective regarding teacher attrition. This perspective reflects a dual structure that demonstrates a big loss of human capital in the teaching profession. Whereas most studies concentrate on the loss resulting from the explicit aspect of teacher attrition, this study shows that leaving the teaching profession is only one dimension of this phenomenon. The other dimension involves hidden aspects that concern those who stay in the educational system, but preform ineffectively. This raises some additional important questions: What are the characteristic of hidden dropout teachers? What are the differences between them and the explicit dropout teachers? And what are the steps needed in order to improve the latter teaching capabilities and performances on school level? These issues may further elaborate the discussion of teacher quality and merit further examination.

Acknowledgements
This study was supported by the Israeli Ministry of Education and the Mofet Institute.
National Authority for Measurement and Evaluation (2010). Evaluation of the "New Horizon" reform at the end of two years of its implementation in primary education. Jerusalem: Ministry of Education (Hebrew).