1. Introduction

While many Canadian educators are deploring the lack teachers, others are wondering why so many are leaving the profession in the first few years. Based on this educational issue, we undertook a Canada-wide survey to explore why new teachers drop out. In this perspective, our three research objectives were: 1) to determine professional aspirations to enter the teaching profession prior to quitting; 2) to identify the main reasons for new teachers to leave the profession; and 3) to identify the individuals to whom new teachers turn when problems arise. To achieve these objectives, we developed two questionnaires that contained both closed and open-ended questions. The first addressed teachers who had dropped out. However, as teachers who have left the field can be difficult to reach (see the Methodology section), we took the original approach of surveying teachers who had witnessed teachers that dropped out. By comparing the responses of the drop-out teachers and the informants, we attempted to obtain a deeper understanding of the issue. In addition to corroborating the literature, our results elucidate a set of systematic and complex relationships that lead to drop-out. We then offer some recommendations to prevent these teachers from dropping out.

2. Teacher Drop-Out: Portrait of a Worrisome Situation

This section aims to shed light on the issue of teacher drop-out. We begin with a definition of teacher drop-out, followed by a presentation of the most frequently cited factors for drop-out in the literature. Finally, we outline the impacts. This section is primarily descriptive, therefore, with the aim of clarifying the whys and wherefores of teacher drop-out. Critiques, both conceptual and methodological, of studies on teacher dropout will be addressed in a further article (in press), based on an extensive literature review on drop-out factors.

2.1. What is Teacher Drop-Out?

Teacher drop-out is an interdisciplinary problem that can be viewed from an economic, organizational, psychological, or educational perspective[29]. We may therefore draw on a variety of reference frameworks, from the idea of human capital[2] to the theory of communities of practice[30]. However, to our knowledge, these frameworks do not capture the complex interdisciplinary nature of drop-out. It would be useful to develop a conceptual framework that is both interdisciplinary yet specific to teacher drop-out, and to combine this with an empirical investigation. Given the lack of such a framework, we limit ourselves here to a description and definition of the main aspects associated with drop-out. The term ‘drop-out’ usually refers to students who leave high school before graduating. However, in Canada’s elementary and secondary schools and in various school commissions and school boards, the term increasingly refers to young teachers who leave the profession. Drop-out among young teachers is generally understood as a voluntary and premature departure from the teaching profession[33]. The term ‘voluntary’ in Macdonald’s definition (e.g.[33]) raises a problem, however. After talks with many teachers across Canada and in focus groups, it is increasingly evident that some teachers are coping with more and more challenges and problems, until the only option is to abandon the profession. In this survey, teacher drop-out is therefore defined as a premature departure from the teaching profession, whether voluntary or not. The literature clearly shows that teacher attrition, far from being confined to retiring veterans, is associated with the induction phase (e.g.[5]). In this perspective, novice teachers (with less than 7 years of experience), not experienced veterans, are the ones who are most often quitting.

Teacher drop-out can be characterized by two notions that usually overlap in the literature, and which we have chosen to distinguish here for a deeper understanding of the concept: drop-out as an inherent feature of any profession, and drop-out as a symptomatic trait of a particular profession. In the first case, drop-out is viewed as a somewhat necessary evil. This view posits that it is better for the teaching profession if teachers who become aware that they lack the skills or the desire to work with students veer towards other professions. In other words, it would be inevitable and even beneficial for some teachers to quit their jobs and leave room for more dedicated or talented teachers[23],[41]. This amounts to a kind of natural culling of the ranks, with clearly positive consequences. In this view, drop-out has always taken place in schools, and everywhere else, according to[32], making it inherent to any profession. To this we may add that today’s climate appears to favor this type of professional redirection. Thus, globalization combined with national and international mobility appears to both condone and even encourage job switching. As a result, single career paths are increasingly rare, and teaching may be considered a short-term assignment[27],[37].

At the same time, teacher drop-out may be viewed as a symptom of professional dysfunction. This second perspective highlights the extent of teacher attrition and the negative fallout for schools (see the section Impacts of drop-out on cost and quality), a problem that merits further research. The issue then becomes how to deepen our understanding of teacher drop-out so that we can redress the problems that ensue, at least partly.

These are not necessarily opposing views. We believe that they are situated on a continuum. That is, teacher drop-out appears to be inevitable and even necessary (notion number one), as long as it remains relatively limited and stable. However, when drop-out rates soar, with negative consequences to education systems, the inherent nature of teacher drop-out takes on an added problematic aspect that places it in the category of a symptom of professional dysfunction (notion number two).

2.2. Why are Teachers Leaving the Profession?

This section presents the main factors underlying teacher attrition from an exhaustive review of the literature. The typology used for the teacher attrition factors is borrowed in part from[5] and from[29]. It presents four main factor types for teachers who leave the profession:

a) Task-related factors: a demanding and time-consuming job[8],[9],[33],[35],[37],[40], management of difficult classrooms[5],[28],[40], unsatisfactory work conditions, particularly low salaries[3],[8],[11],[15],[16],[17],[24],[33], [34], inappropriate teaching subjects[29],[33], restrictive administrative policies[3],[8],[13],[33], and unappealing tasks[13],[33],[34],[35];

b) Individual factors: emotional and psychological characteristics that are incompatible with the teaching profession[17],[9],[13],[24],[35], and sociodemographic and professional factors[3],[4],[17],[19],[21-22],[33],[36],[37];

c) Social environment factors: failed relations with educational and social actors[5],[15],[17],[40], and difficult students and workplace conditions[14],[18],[23],[28];

d) Socioeconomic conditions[10],[15],[37].

We should emphasize that the attrition factors appear to be closely associated, i.e., a given factor may be associated with another or several other factors. For example, age and number of years of experience (individual factors) are often associated with the type of students and the work conditions assigned to the teacher (social environment factors). This interdependence of attrition factors suggests that teacher attrition is more the result of a set of factors than a single factor, which only increases the likelihood that teachers will drop out.

2.3. Impacts of Drop-Out on Cost and Quality

Teacher drop-out is becoming problematic in two respects: the cost incurred and the consequences for the quality of teaching.

In the United States, the Alliance for Excellent Education (AEE)[1] estimated the cost of teacher attrition at almost three billion American dollars in 2004. Losses are felt at the levels of initial training, recruitment, hiring, and professional development. The OECD, the Organization for Economic Co-operation and Development [37] cited the same financial issue.

The consequences of teacher attrition for teaching quality are another major concern reported in the literature. On the one hand, a high attrition rate implies a heavy turnover of teaching staff, which makes it difficult to establish a cohesive school team[1]. In addition, teacher attrition involves a large proportion of novice teachers[37],[42], i.e., teachers who leave the profession prematurely when they have not fully mastered their professional skills. Consequently, we could assume that the teaching quality they provide is lower than that of teachers who have reached the stabilization stage[1],[37]. Add to this the fact that attrition necessarily entails hiring more novice teachers, who are also in the skills-building stage. Teaching quality istherefore doubly affected, both because novice teachers who quit have not achieved optimal teaching skills (remember, they are still in the induction phase), and because this turnover requires the hiring of more novice teachers, who are also building their expertise. At the end of the day, the students are the ones to bear the cost of teacher attrition, in that the quality of the teaching they receive is lower than if the teachers had persisted[1],[37],[41].

2.4. The Drop-Out Teacher: an International Problem

Another worrisome issue is that teacher attrition is an international problem that affects northern and southern countries alike. In the United States, Reference[25] noted that the attrition rate is higher among teachers than in many other professions: 46% of new teachers leave their job in the first five years of service. The comparative study by Reference[42] in eight industrialized countries shows that in the United Kingdom, 40% of beginning teachers abandon the profession in the first three years of service (while Reference[12], found a rate of almost 44%). Meanwhile, Australia has seen an 18% drop-out rate in women aged 25 to 29 years, the data for men being unavailable[42]. However, some countries such as France, Germany, and Portugal have reported attrition rates below 5%. The explanations for this put forward by Reference[42] include centralized education systems and teaching programs, the non-accountability of teachers when students fail, and more formative than summative evaluations of teaching staff. Nevertheless, attrition was a growing trend in these same countries from 1995 to 2000[37], and we could reasonably assume that the rates are higher today.

There is little data available in Canada, and it is equivocal. In 2004, Reference[13], estimated teacher turnover at approximately 30% in the first five years of service. On the other hand, the Reference[38] predicted a 6.7% turnover for its members after the third year of teaching[38]. Elsewhere, the Reference[37] ranked Canada’s public elementary and secondary schools with a 3–6% attrition rate in 1999 (including departures and retirements), and estimated that this rate was more or less the same in 1995. It is therefore difficult to get a clear picture of the extent of teacher attrition across Canada, given that the available statistics vary from 3%[37] to almost 30%[13].

3. Survey Objectives

In light of the few studies and the statistical variations concerning drop-out rates among new teachers in Canada, the overall objective of this Canada-wide survey was to better understand why they are leaving the profession in the first few years of service. Under the scope of this study, it is impossible to cover all subjects taught. A number of specific objectives contribute to this overall objective, as follows:

1. To determine professional aspirations to enter the teaching profession prior to quitting;

2. To identify the main reasons for new teachers to leave the profession;

3. To identify the individuals to whom new teachers turn when problems arise.

These objectives cover the primary concerns of the educators we have talked with in recent years, who motivated us to carry out this Canada-wide survey. They are also consistent with the literature on teacher drop-out, for which we provide an overview below.

4. Methodology

We now present the methodology we used, beginning with the data collection and study participants followed by the analysis procedure.

4.1. Data Collection and Participants

To address the geographic and professional dispersion of drop-out teachers, we used online questionnaires, which have the advantage of being administrable via the Internet, thereby transcending the usual limitations of time and space. They contained both closed questions, mostly rated on a Likert scale, as well as open-ended questions. One questionnaire was developed for teachers who had dropped out, and a second for teachers who had witnessed the events surrounding a specific dropping out event – simply put, they had to be clearly aware of the context which lead to the dropping out of a specific new teacher. The two questionnaires addressed the same themes, such as reasons for quitting, requirements for preventing attrition, human support available to drop-out teachers experiencing problems, teaching as a career choice, and professional aspirations to become a teacher. The informants were included in the study to compensate for a major methodological shortcoming in the empirical research on drop-out: lack of participants. We should also highlight that it was not possible to include other important information such as students or parents in this study. This could be part of future research.

It is important to emphasize that teacher attrition is particularly difficult to study because the participants of interest (drop-out teachers) are by definition ‘out of touch’ with the teaching profession. In other words, we know where they aren’t, but we don’t know where they are. It is therefore difficult to obtain a large sample of participants and consequently to better understand the real reasons for dropping out[29]. We are therefore fully aware that our sample of dropped-out teachers was not necessarily representative of the whole group of drop-outs.

On the other hand, the inclusion of teachers who witnessed the events surrounding the dropping out would be liable to bias the results, for several reasons. First, perceptions of teacher attrition could differ significantly between drop-out teachers and informants, who would have an outsider’s viewpoint. Moreover, perceptions of a same fact could differ between drop-out teachers and informants, depending on individual interests (e.g., school principals would surely be reluctant to admit that they had failed to support teachers who were having problems). Finally, although the informants were in a position to discuss the behavioral variables (what the drop-out teachers did) and the status variables (who the drop-out teachers were), they appeared less able to provide information on the thinking variables (what the drop-out teachers thought).

To mitigate this methodological bias, our analyses systematically included a comparison of the two groups to highlight any differences. By comparing the drop-out teachers and informants, we were also able to put forward reasonable hypotheses about how the drop-outs were perceived by the various school actors. Furthermore, the analyses of variance between the two subsamples show no significant differences. This suggests, among others, that the two respondent groups hold essentially the same views on teacher attrition, or at least on certain underlying problems.

The questionnaires were pretested on 26 teachers and 11 education actors (school principals and pedagogical counselors). The questionnaires were posted online for three weeks from March to April 2008. To guarantee that the questionnaires would be widely distributed, we mobilized three of the largest teachers associations in Canada and we published ads in five newspapers in two languages (French and/or English) in the cities of Halifax, Toronto, Calgary, Vancouver, and Montreal. This collection procedure enabled us to include 34 drop-out teachers and 167 in our study, for a total of 201 participants.

4.2. Data Analysis

The data collected from the questionnaires were mainly quantitative (responses to closed questions), but also qualitative (responses to open-ended questions). The Likert scores on the questionnaire were quantitatively analyzed with descriptive and inferential statistics using SPSS 19. This allowed us to draw a sociodemographic portrait of the participants, uncovering some interesting points bearing on teacher drop-out. The initial analysis results were further complemented by a qualitative analysis of the open responses to the online questionnaire. This consisted of a content analysis (see[21],[31]) with semi-open coding, initially constructed from the various factors influencing attrition. The aim was to highlight the relationships between the different moderators of attrition identified in the quantitative analysis.

5. Presentation and Analysis of Results

The results of the study are presented in two main parts: a quantitative analysis of the responses to the online survey questionnaire and a qualitative analysis of the responses to the online survey questionnaire.

5.1. Quantitative Analysis

The descriptive statistics presented in this section are divided into two parts: the first deals with the drop-out teachers’ responses and the second with the informants’ responses. It is noteworthy that, parallel to the descriptive statistics, we performed an inferential statistical analysis (analysis of variance) of certain responses, particularly concerning the difficulties encountered, professional aspirations, and teaching as the first career choice. These variables were initially applied to all participants. We then confronted the results for the drop-out teacher group with those for the informant group to identify significant between-group variations in perceptions of drop-out. No significant differences were found, indicating that perceptions of teacher drop-out do not differ significantly between those who experienced it and those who witnessed it, with some minor exceptions.

5.1.1. Drop-Out Teachers

A total of 34 drop-out teachers participated in the survey (22 of whom were women). A large proportion of the respondents reported having a teaching diploma, i.e., a bachelor’s degree (55.2%) or master’s degree (41.4%). Half the drop-outs had quit teaching within two years (10.7%) or less (39.3%) of the time of the study, which suggests that the experiences related in this study were relatively recent, at least for most respondents.

At the time of dropping out, 70.3% of respondents had five or fewer years of teaching experience. These results support the argument that teachers are particularly inclined to abandon the profession during the induction period, which lasts for the first seven years of teaching[43].

The results on professional aspirations to become teacher are mixed. Although the majority of respondents (53.8%) agreed somewhat (19.2%) or moderately (34.6%) that they had professional aspirations to go into teaching, at least 46.1% of the drop-out teachers surveyed agreed strongly (42.3%) or completely (3.8%). Similarly, 50% of respondents agreed strongly (35.7%) or completely (14.3%) that teaching was their first career choice.

Figure 1. Main difficulties of drop-out teachers in teaching: amount of work to be done at home (too much)

Figure 2. Main difficulties of drop-out teachers in teaching: workload (too heavy)

The main difficulties that drop-out teachers had to cope with in their jobs were structural and organizational requirements: the amount of work to be done at home, which 80% agreed was too much (Figure 1), and the workload, which 77.8% of respondents agreed was too heavy (Figure 2). Overall, it seems that new teachers found the job too time-consuming.

The other difficulties mentioned feature less prominently in the results analysis. With regard to professional relations, 40.7% of respondents agreed that they had a difficult relationship with the school’s administration. This result is explained by the qualitative analysis (see section 4.2.1.Reasons for leaving the profession). Relations with colleagues were also reported as problematic by 51.8% of respondents, but to a lesser degree than relations with the administration. Moreover, the relative importance of these results must be considered, insofar as relations with the administration and colleagues were not a major problem for 29.6% and 37.1% of respondents, respectively. Relations with parents, albeit less problematic, were generally reported as similar to relations with the administration and colleagues.

Classroom management was a frequently cited explanation for teacher attrition (see section Why are teachers leaving the profession?). Directly linked to classroom management was a lack of respect by students, which was noted as very problematic by 26.9% of respondents, but surprisingly, as only somewhat problematic by 34.6% of respondents. Moreover, poorly motivated students appear to have been a challenge for 33.3% of respondents, whereas 29.6% were apparently not bothered by this problem.

In sum, aside from the amount of work to be done at home and the heavy workload—challenges on which the respondents agree—the ratings of other difficulties diverge widely. It would seem that the factors varied greatly across the individual drop-out experience, which leads us to believe that specific characteristics of teaching conditions and/or individual teachers play an important role in explaining it.

5.1.2. Key Informants on Teacher Drop-Out

The informants who witnessed the events surrounding the teacher drop-outs are more numerous in this study, comprising 167 respondents, the great majority of whom are women (82.7%). Half the respondents referred to a drop-out that occurred two years previous (17.3%) or less (34.6%) to the time of the survey.

Among the difficulties that the drop-out teachers had to cope with, the informants agreed that too much work to be done at home and a too heavy workload were the main causes (respectively 72.5% and 65.8%) of their eventual departure from the teaching profession (Figures 3 and 4). These findings corroborate the statements of the drop-out teachers.

Note also that other aspects related to classroom management, such as student misbehavior (34.6%), poor motivation (48.1%), and lack of respect for the teacher (30.7%), were frequently cited as being very problematic for the drop-out teachers.

As for relational aspects, relations with the administration (50.3%) and colleagues (44.1%) were reported at slightly lower levels than by the drop-out teachers themselves. On the other hand, the informants reported more difficulties in relations with parents of students (56.6%). In other words, we observe an inverse trend: whereas relations with the administration and colleagues were perceived as the main relational difficulty by the drop-out teachers, the informants attributed more difficulties to relations with the parents of students. Knowing that almost all the informants were teachers or school principals, we wonder whether this might be a case of offloading responsibility. By this we mean that teachers and principals were probably loath to assume responsibility for their colleagues’ departures, and might therefore have unconsciously ‘scapegoated’ the parents. We must emphasize that this could be an inherent bias in the methodology of this study. Conversely, it could be that teachers and the administration, overwhelmed by day-to-day challenges and not enough time to deal with them, were not in a position to come to the aid of beginning teachers, even if they wanted to. In this case, these inverse trends could be explained by the fact that drop-out teachers placed too much blame on their colleagues and the administration.

Figure 3. Main difficulties of drop-out teachers in teaching: amount of work to be done at home (too much)

Figure 4. Main difficulties of drop-out teachers in teaching: workload (too heavy)

5.2. Qualitative Analysis

A content analysis was applied to the open responses to the online questionnaire, which produced a semi-open coding[21] using QDAMiner software. The qualitative results thus obtained enabled a more in-depth understanding of several of the points advanced in the quantitative analysis (see section 4.1.Quantitative analysis), particularly for identifying certain drop-out factors and their relationships (e.g., lack of support by the school administration and colleagues; teaching as a career choice) and for relating them to associated aspects (e.g., solutions for preventing drop-out).

5.2.1. Human Resources Available to Drop-out Teachers Having Problems

Lack of support for drop-out teachers who are having problems has largely been reported as a drop-out factor in quantitative results. In this section, this support is related to the relationships between the drop-out teachers and other school actors.

First of all, the drop-out teachers said that they turned to the administration to resolve their difficulties. Although some respondents reported receiving unconditional support from the administration, others begrudged the lack of involvement in their problems, which is directly connected to the above-mentioned lack of support. Some teachers in immersion programs also noted a lack of understanding on the part of the school administration, and the absence of specific immersion teaching conditions. In other words, the drop-out teachers seemed to primarily seek the help of an interceder—the school administration—which did not seem to be prepared to help them.

Another point was frequently brought up by the informants: drop-out teachers hesitated to go to the school administration when they were having problems because they were afraid to look incompetent. This fear probably also involved job insecurity. Going to see the school principal to talk about problems in the classroom somehow means admitting that one is not yet ready for a permanent job. Therefore, it was probably not so much a question of lack of support as a lack of trust between the administration and the drop-out teacher.

Fellow teachers also provided support in order to deal with problems, particularly because they were closer to the drop-out teachers and were less in a position to judge. In this case, experienced colleagues were often cited as the preferred contact persons. However, albeit to a lesser extent, some respondents felt that a lack of team spirit and time were obstacles to collegial support. This resulted in a kind of isolation for the drop-out teachers, as reported by both drop-out teachers and informants.

Note how that support by the administration and support by colleagues were often connected in tandem. In other words, the drop-out teachers turned first to their colleagues and afterwards to the administration when problems persisted. Or inversely, they appealed first to the administration when they were having a problem and then to their colleagues when they did not receive the help they wanted.

5.2.2. Choice of Teaching as a Career and Professional Aspirations of Drop-Out Teachers Prior to Quitting

Teaching as a career choice and professional aspirations showed contrasting results in quantitative terms (5.1.1.Drop-out teachers). The qualitative results allowed us to deepen our understanding of why. First, some of the drop-out teachers showed a lack of interest in the teaching profession from their very entry into the profession. For some of them, teaching was a possibility among many professional interests, while for others it was a temporary ‘starter’ job. In another case, teaching was a default profession because there was little choice at the university. This last point appears to be connected to the lack of recognition given to the teaching profession—a secondary aspect—which means that drop-out teachers might not embrace a job that they do not much value.

In opposition to this initial lack of interest, and with almost equal frequency, the initial motivation to teach reported by some of the drop-out teachers appears to indicate that the drop-out teachers chose this profession voluntarily. How then to explain the subsequent departure of these respondents? Two explanations are considered, and were mainly raised by the drop-out teachers themselves. The most often cited is a disconnect between the initial perception of teaching and the harsher realities of the job. That is, teachers tended to idealize teaching. To a somewhat lesser extent, a lack of teaching skills was also mentioned to explain initial motivation and subsequent departure.

Whether a poor perception of the realities of teaching or a lack of skills was involved, initial training was sometimes blamed, primarily by the informants. The argument was that university teaching programs do not prepare, or poorly prepare, new teachers for the realities of the job, recalling the debate between theory and practice. An initial training program that provides the requisite skills might therefore help prevent teacher attrition.

5.2.3. What Can be Done to Prevent Teacher Attrition?

The requirements for preventing teacher attrition are partly related to the reasons for the attrition, as mentioned above. We should first specify that, unlike the previous section, which presents some differences between drop-out teachers and informants, the responses on the needs of drop-out teachers who were having problems were largely shared by all participants. Support in general and administrative support in particular were the most often cited needs. There were also requests for specific teaching conditions, such as a lighter workload and more time for daily preparation, or the chance to teach the same grade two years in a row, especially for beginning teachers. This is all the more understandable when we know that new teachers are usually asked to change grades several times during the induction period. Reducing isolation, for instance through more communication and collaboration with colleagues and other school actors, was a frequently mentioned need. It appears to be directly connected to lack of support as a reason for quitting. However, this is far from the concept of mentoring, for instance by an experienced colleague rather than a member of the administration, and this seems to be the most important requirement to prevent beginning teachers from quitting, as corroborated by the literature[26].

6. Discussion

To conclude, it is noteworthy that the majority of the drop-outs reported in this study occurred within the first five years of teaching (including 50% within the first two years), which confirms the argument that the professional induction phase is particularly conducive to teacher attrition[17],[18].

In terms of difficulties encountered by the drop-out teachers, excessive workload outside the workplace (at home, etc.), too often heavy workloads, and the resultant lack of time were the main points related to work conditions. Although it came up, low salary was not a prominent factor in teacher attrition, a finding that has been reported previously[6],[37],[39].

Classroom management and difficult students were also major challenges for new teachers, whatever the subject being taught. In fact, classroom management is a destabilizing feature that confronts beginning teachers from the start[40]. Moreover, young teachers usually prioritize this aspect over teaching the subject or student learning[28].

The relational aspects of teaching were further problems that new teachers must cope with. Having a difficult relationship with some of the students’ parents was repeatedly mentioned as a reason that drove future teachers to quit, especially by the key informants. On the other hand, the drop-out teachers placed more emphasis on problematic relations with the administration and colleagues. In this respect, lack of support by the administration and colleagues, particularly when problems with parents occurred, was the most recurrent theme among the respondents, which corroborates the findings of recent studies[5],[9],[16]. Besides the perceived lack of support, the respondents also reported a fear of looking incompetent. This may be explained in part by the job insecurity that most beginning teachers experience: asking one’s employer for help could be viewed as an obstacle to securing a permanent job.

Teaching as a career choice and professional aspirations to become a teacher show two inverse trends. In the first case, some of the drop-out teachers expressed a lack of initial interest in the profession, which partly explains their eventual departure. However, and in equal proportions, the drop-out teachers were motivated to go into teaching. Two explanations may be posited to clarify the relationship between the initial motivation and the subsequent departure: either the teachers idealized the teaching profession, a point that has been raised in other studies (e.g.,[18],[20]), or else they had poor teaching skills. In both cases, initial training was repeatedly blamed, in that it inadequately or partially prepared the students for teaching, indirectly raising the odds that they would leave the profession.

So what can be done to prevent teacher attrition? Based on our results, better support in general and better administrative support in particular as well as university training that better provides the requisite skills are suggested, along with specific conditions for beginning teachers (including lighter workloads), and strategies to provide them with help when they experience problems. Above all, the most often cited need by the respondents concerns an aspect of professional induction: mentoring, which appears to have a positive impact on professional insertion and the prevention of attrition[26].

In view of these results, it is important to suggest some directions for future research. First, it would be instructive to conduct similar surveys on a regular basis in order to deepen our understanding of why teachers drop out. This would be consistent with the conclusions of the meta-analysis by Reference[5], which emphasizes the need for reliable empirical data. In addition, it would be important to examine the relationships between the diverse moderators of teacher attrition to better understand how they interact. More studies should be conducted to examine and compare the perceptions of a variety of actors, as we have done. Finally, in line with the project Current Trends in the Evolution of School Personnel in Canadian Elementary and Secondary Schools (http://www.teachcan.ca), it would be useful to conduct a follow-up study on teachers, from university training to professional induction into teaching.