The practice of Tunisian physical education teachers at the end of initial training during the preparation for professional life

Background and Study Aim: The purpose of this article is to describe and analyze students' disruptive behavior and teacher trainee responses before and after a “Body Language and to Speak in Public” training module for school teachers at the end of initial training physical education teachers (PET). Material and Methods: Delayed video scope analysis was conducted using the "Disciplinary Incidents Observation System (DIOS)" of “Brunelle J.” (1996) [7]. Results: The data collected, it can be deduced that the courses directed by trainee students during work readiness internships show a high degree of disruption, since there is a rate of 1.3 and 1.01 DB per minute. Conclusions: The frequency of onset of disruptive behaviors (DB1 and DB2) is slightly lower in sessions facilitated by trainees who have been trained in "Body language and public speaking". Similarly, at the level of disruptive behaviors (DB3), the trainees who underwent the training realized a greater decrease in the frequency of appearance of these behaviors. Faced with these disruptive behaviors, the trainees who attended the training were slightly more interactive in their reactions during the sessions. The impact that this study could have on the initial training of physical education teachers (PET). The results of our studies illustrate the reality of the practice of future teachers during the internship preparation to professional life. Indeed, they constitute a The Journal of Quality in Education (JoQiE) Vol.11, N°17, May 2021 100 repertoire to perceive the different disruptive behaviors of students and the reactions of trainee students to these behaviors. By way of this presentation, our work can certainly be used as part of the initial training of PET and in formalizing the professional skills repository.

in physical education. The specific objective of the study was deals with the description of student's disruptive behavior during the sessions led by the trainee teachers as well as about these reactions to these deviant behaviors of the students before and after the training (BLPS) in the internship preparation to professional life.

Material & methods
This research consists in a quasi-experimental study for the fact that there is manipulation of a variable, namely the training program of « Body Language and Public Speaking (BLPS) » and that there is an observation of its effect on disruptive behavior in physical and sports education.

The Training Program in 'Body Language and Public Speaking'
The training program predicted 12 meetings lasting 2 hours, which makes 24 training hours. The training started in September and it ended in March. Indeed, each meeting is associated to a thematic content that was presented, worked, discussed and experimented. A training meeting implies a theoretical content followed by its implementation. By the 'active experience', the trainees are asked to plan, organize and supervise teaching sequences then, outside meetings, they were invited to implement the elements of content in their training environment. In order to have a more positive effect between the experience and the learnings, successes were systematically underlined while failures were discussed and analyzed, thus allowing to make all the aspects of the training program constructive.
Trainee teachers are invited to: (1) Work on oral expression techniques (breathing, voice, articulation, rhythm and repetition).
(3) Improve the perception of self.
(4) Tame, regulate stress and control the speech.
Note: The former is a University professor at Higher Institute of Dramatic Arts in Tunis (I.S.A.D), communications specialist and expert in 'Body language and speaking in public.

Participants
The sample formed by student volunteers consists of a first reference group (A) n = 10 (6 men, 4 women) and the second experimental group (B) n = 10 (5 men, 5 women).
All volunteers were in the third and final year of university education in PSE. They were launched in a practical training course in a thirty-week long secondary school environment, with four hours of practice for each of them, for a total of 120 hours of annual practice.
Each of the four-hour episodes was a block of four 50-minute lessons each time around the same groups. All participants were previously informed about the aims of the study as well as the arrangements made to preserve their anonymity and the confidentiality of the data collected.
A total of 389 (Mature: 13.22 ± 0.35) high school students, of whom 243 (62.47%) were male and 146 (37.53%) female, participated in this study with an average of 34 students per class.
They were engaged in collective sports activities (either handball or basketball) since the project of their schools only uses collective sports with a view to facilitating their social integration.

Procedure: Didactic observation
The ten trainee teachers in cohort B (experimental group) will be compared to the ten other trainee teachers in cohort A (control group) during a practical training course.
The observation was made at two moments: the first collection before training, the second within a week of the end of the training. These observations took place in the exercise sites of the practical pedagogy course. The data collection will cover 40 sessions of 50 minutes each, which were filmed before and after the training. It was done at two points during the 2015-2016 school year: the first collection at the beginning of the 'internship preparation to professional life' (September: before the beginning of the training) and the second at the end of the internship (March: after the end of the training).

The Instrument of Data Collection
The device uses the sound / image coupling in order to be able to relate the behaviors of the different actors and tell them of each one (instructions, private or public remarks, verbal reactions of the trainee and the students). We used two Sony model 4K Handcam cameras with built-in projector and a Boom Tone DJ wireless microphone equipped with a transceiver (VHF 10HL F4 Micro HF) and a range of 100 meters to be able to intercept verbal interventions of the student's teacher.
All the trainee teachers were filmed at least during a session before the recording of the data, in order to accustom the protagonists of the study to the material used. In order to reduce the Hawthorne effect among teachers (behavior modification due to the presence of an observer), the experimenter introduced himself to the teacher as being a student conducting a survey on student motivation in EPS, without making any reference to the Pygmalion effect.
Data collection is done with the help of two camcorders and a wireless microphone. The two cameras are placed in diagonally opposite positions that cover the different angles of the whole area where the session takes place. The data collection will cover 40 sessions of 50 minutes each, which were filmed before and after the training.

The Grid of Observation
In order to analyze students' disruptive behaviors in physical education classes, we have used the works of " Brunelle J." [7], the authors of the DIOS. This system helps describe the disciplinary incidents whose disruptive behaviors (DB) occur during physical education classes based on the moment of occurrence. The observation grid shows 8 categories: (1) Students' DB; (2) Intensity level of DB; (3) DB's moment of occurrence; (4) Number of students involved; (5) Effects of the DB on the proceeding of the session; (6) Student teachers' types of reactions to DB; (7) Effects of the student teachers' reactions on the DB; (8) DB's accessibility or inaccessibility level for student teachers; To analyze the disruptive behavior of students in physical education sessions, we used the version of "DISCIPLINARY INCIDENTS OBSEVATION SYSTEM (DIOS)" [12]. The DIOS is an observation system with predetermined categories. It identifies and describes the content of disciplinary incidents that DB encounter during a physical education session. The DIOS is based on an event observation strategy, that is to say that disciplinary incidents are noted according to their appearance during a session.
The DIOS uses an event observation strategy. For example, disciplinary incidents are coded as they occur during physical education classes. More specifically, the DIOS allows the analysis of a disciplinary incident according to several components (the moment of the lesson where the incident occurs, the number of students involved the disruptive behavior of the students, the reactions of the teacher, the duration of the disciplinary episode, the effect of the teacher's reaction and the source of the incident).
The nineteen behaviors that were chosen to report the most common deviances of students are presented in Table 1. These behaviors are grouped into three levels according to the severity of the disruptive behavior and its influence on the course of the session (Table 1). The possible reactions that the teacher can adopt when there is an emergence of non-observances are twelve in number and are related to the three types of pedagogy (normative: behaviors of imposition, libertarian: permissive behaviors, interactive: behaviors of affirmation and openness). Categories of teacher reactions are presented in Table 2. -Dictate a behaviour -Reprimand -Designate a consequence.
-Make a reminder -Ignore -Apply a consequence -Give a reason.

The coding
Two coders were trained in the use of DIOS for coding video recordings. The coders first worked as a team to become familiar with the observation grid and master all its components. There was a need to practice classifying DB that occur during PET sessions. In the second place, comes the individual coding followed by the confrontation of the grids which showed some divergences. It was therefore necessary at times to return to the definitions of the components of the grid to ensure the compliance of the DB and agree on the same interpretation.
After the training period, the coding of the two coders was subjected to the fidelity test several times before starting the final coding.

Statistical Analysis
The set of dependent variables related to time of learning have been identified by a grid of observation measuring the time of performance of the duties mentioned above.
We used a software of statistical "Statistical Package of Social Science" SPSS 16.0. The threshold of meaning withheld is of 0.05.

Inference statistics
Given the small number of observations and the non-normality of the distribution of the whole of the values of the variables, we chose the Mann-Whitney U-test of independent samples and Wilcoxon signed rank test of associated samples to compare the values of The Journal of Quality in Education (JoQiE) Vol.11, N°17, May 2021 110 the variables related to the learning time of two groups of trainee teachers.

Disruptive behaviour of students before and after training:
The results shown in Table 3 show the frequency of onset of disruptive behaviour before and after training in both groups of trainee students. Table 3. Frequency of disruptive behavior adopted by students before and after training in ten sessions led by student physical education trainees. For the control group; during the first two months of work experience preparation (T0); the classification of disruptive behaviours by level shows that about 50.12% of the behaviours are of second level, that is to say that they are likely to disturb the class in the short or medium term. More specifically, the behaviours " Fooling around " (24), "Make noise" (84), " Squabbling " (126), and " Bulling " (52) are the second most commonly reported second-level deviances.

Disruptive
First-level disruptive behaviours, which have a small influence on the life of the class but may still disturb the student trainee, account for approximately 42.27% of disruptive behaviours adopted by students. The main deviances of this category are "Talkative " (158) and "Distracted" (87). Third-level disruptive behaviours, which actually disturb the good progress of the class from the moment they occur, are much less frequent (7.52%) and are expressed mainly by deviances such as " Resisting instructions" (13)," Dangerous behaviour "(12).
The Journal of Quality in Education (JoQiE) Vol.11, N°17, May 2021 114 Finally, it is interesting to note that the disruptive behaviours "Distracted" and "Talkative" (first level) as well as " Squabbling " and " Making noise" (second level) alone account for 64.54%.
At the end of the work experience preparation course (T1); 52.17% of the behaviours are second level. More specifically, the behaviours " Squabbling " (94), " Deforming the rules " (71) and " Making noise" (68), constitute the most often identified deviances.
First-level disruptive behaviours account for approximately 40.17% of the disruptive behaviours adopted by students. The main deviances of this category are "Talkative" (124) and " Distracted" (57).
At the level of third-level disruptive behaviour, which accounts for 7.65% of all disruptive behaviours and is expressed mainly by deviances such as " Resisting instructions " (09), " Lashing out at matériel" (09). It is interesting to note that disruptive behaviours "Talkative" (first level) as well as " Squabbling ", " Deforming the rules " and "makes noise" (second level) alone account for 79.55% of all behaviour's disruptors.
Finally, it should be noted that all the disruptive behaviours (1st, 2nd and 3rd level) were reduced by 18.44% at the end of the work experience training period.
For the experimental group; before the beginning of the training (T0); the classification of disruptive behaviours by level shows that about 51.89% of the behaviours are second level. More specifically, the " Squabbling" (109) and " Bulling " (72) are the second most frequently identified second-level deviances.
At the level of first-level disruptive behaviours that account for about 40.72% of disruptive behaviours adopted by students. The main deviations of this category are "Talkative" (163) and " Distracted" (70). Finally, third-level disruptive behaviours are less frequent (7.39%) and are expressed mainly by deviances such as " Resisting instructions " (16), " Dangerous behaviour" (08). Finally, it is interesting to note that the disruptive behaviours " Distracted" and "Talkative" (first level) as well as " Squabbling ", " Bulling" (second level) count alone for 73.11%.
After three months of training (T1), 49.77% of the behaviours are second level.
First-level disruptive behaviours account for approximately 41.82% of the disruptive behaviours adopted by students. The main deviances of this category are "Talkative" (116) and "distracted" (57).
At the level of third-level deviant behaviour which represents 8.41% of all disruptive behaviours and is expressed mainly by deviances such as " Resisting instructions " (11), " Mugging" (08).
We also find that the disruptive behaviours "distracted" and "Talkative" (first level) as well as " Squabbling"," Making noise " and " Bulling " (second level) alone account for 79.55%.
Finally, it should be noted that all the disruptive behaviours (1st, 2nd and 3rd level) suffered a decrease of 30.82% in the control group. Hence, this decrease in the frequencies of appearance of deviant behaviours adopted by the students during the sessions led by the trainee teachers is more important at the experimental group than the control group. Before training (T0), the frequency of onset of Type 1 disruptive behaviours (DB1) was insignificant between the two groups. This means that there is no difference between the two groups in the frequency of occurrence of DB1.

The first level disruptive behaviors (DB1)
After three months of training (at T1), the frequency of deviant episodes was not significant between the two groups except at the 'Game' (p = 1.229.10-4) and 'conclusion' moments (p = 0.007). In situations of 'play' and 'conclusion', the frequency of occurrence of these behaviours was greater in Control group than Experimental group.
At the end of the vocational training course, the variation in the frequencies of DB1 appearances decreased significantly in both groups (Control group and The Journal of Quality in Education (JoQiE) Vol.11, N°17, May 2021 117 Experimental group). However, at the time before the course the frequencies of appearance of the DB1 remain very high for the Control group (p = 1) and the Experimental group (p = 0.305). Before training (T0), the frequency of onset of Type 2 disruptive behaviors (DB2) was insignificant between the two groups, except at the 'Educative' phase (p = 4.8.10-8).

The second level disruptive behaviors (DB2)
During this phase, sessions led by the GCONT scored a higher number of DB2. At the end of the work experience preparation course, the variation in DB2 occurrence frequencies decreased significantly in both groups (Control group and Experimental group).   The trainees 'reactions to the disruptive behavior of their students are shown in Table 7 and reveals that trainees' Experimental group and Control group reacted 491 and 552 times respectively to disruptive behaviors displayed by their students in ten sessions before the start of the formation (T0).

Trainee teachers' reactions to the disruptive behaviour of their students before and after the training
The nature of the trainees' reactions reveals that the normative approach is dominant among them (Experimental group and Control group). Indeed, more than 50% of his reactions constitute tax behaviors. The most revealing reactions of this trend for both groups are "dictates behavior" and "reprimand". In addition, the Control group is regularly libertarian (40.58%) than the Experimental group (32.79%). Finally, it should be noted that the interactive pedagogy is far from being used by the trainees of the two groups: Control group (12.68%) and Experimental group (8.76%) at the beginning of the internship preparation to professional life. After 3 months of training, the reactions of training teachers to the DBs of pupils reach 495 for GCONT and 407 for GEXP during 10 sessions after the end of the training.
After the training, the normative approach dominates the nature of the reactions of the trainees of the two groups (Experimental group and Control group). In fact, the tax behaviors cover more than 45% of Control group and 51.6% of Experimental group reactions. The most revealing reactions for both groups remain "dictates behavior" and "reprimand". In addition, the Control group is regularly libertarian (38.18%) than the Experimental group (30.22%). Finally, it should be noted that the Experimental group after the training was more interactive in its reactions to their students than the Control group since they reach 18.18% of the reaction set, while the Control group reaches 16.77%.

Discussion
At the level of disruptive behaviors (DB), the data collected with the help of DIOS first showed that the courses run by trainee students during a work experience training course show a high degree of disruption since there is a rate of 1.3 and 1.01 DB per minute. We also find that the highest number of deviant student behaviors was secondlevel (DB2) with a percentage of over 49% of all behaviors. In addition, all type 1 and 2 disruptive behaviors constitute 90% of inappropriate behaviors and are behaviors with little influence on the life of the class when they appear. Whereas, third-level deviances, which actually interfere with the smooth running of the class from the moment they occur, are much less frequent and constitute between 7% and 8% of students' deviant behavior during sessions led by trainee students. These results are clearly in line with other research using "DIOS" [16].
For first-level disruptive behavior (DB1), the frequency of onset was insignificant between the two groups (Control and Experimental group) during the different moments of the session, at the beginning of the internship preparation to professional life. Indeed, DB1 represents approximately 42.27% for the Control group and 40.17% for the Experimental group. This is consistent with the study by 'Stephan Dostie' [15] who states that students commit particularly high-level non-observances when they are close to the teacher during periods of explanation.
After three months of training (T1), the frequency of deviant episodes was insignificant between the two groups except at the 'Game' (p = 1.229.10-4) and 'conclusion' moments (p = 0.007). In both situations, the frequency of occurrence of these deviant behaviors was greater in the Control group than the Experimental group. This is explained by trainees' ignorance behaviors to disruptive behaviors in certain phases of the session, which encourages their repetition and even their amplification [7].
However, the high frequency of disruptive behaviors in the classes observed suggests a more specific analysis of the moments when these non-observances occur. Similarly, it should be noted that the main deviations of this category before and after the training for both groups were s "chamaille".
At T0, the frequency of onset of third-level disruptive behaviors (DB3) was insignificant between the two groups (Control group and Experimental group) at different times of the session. Indeed, the DB3s represent approximately 7.52% for the Control group and 7.39% for the Experimental group of the set of behaviors.
At T1, the frequency of the DB3s was no significant between the two groups (Control group than the Experimental group). However, it should be noted that at the level of the variation of the frequencies of appearances of the DB3, a significant decrease is detected and more important in the Experimental group than the Control group. and' Conclusion '(p = 0.346). These results converge towards the study of "Cicurel F." [10] who asserts that the high frequencies of DB3 is one of the clues for the teacher that these episodes are too long for the attention span of his students and that these organizational routines are no longer effective. However, it should be noted that trainees trained in "Body language and public speaking" were slightly more interactive in their reactions to the different deviant behaviors of their students than other trainees.
In terms of finalization, the data collected with the help of the DIOS at the end of the work experience preparation period first allowed us to note the high frequency of disruptive behaviors in the sessions observed. More specifically, more than 90% of these non-observances may potentially disturb the class in the short or medium term (DB1 and DB2). For third-level deviances (DB3), which actually disturb the smooth running of the class from the moment they occur, are much less frequent and constitute between 7% and 8% of non-observances shown by students.
In addition, disruptive behaviors appear more frequently at certain times of the session. In fact, students regularly adopt inappropriate behaviors during transitions, explanations, educational and play situations [25]. However, in the course of the sessions led by trainee students who have undergone the training of "Body language and speaking in public "; the frequency of occurrence of DB1 and DB1 are slightly lower.
For DB3s, a larger decrease was detected in sessions led by trainees who attended the training than their counterpart.
On the other hand, normative pedagogy dominates the nature of student trainees' reactions to the different disruptive behaviors along the internship preparation to professional life. While, the trainees who attended the training were slightly more interactive in their reactions to the different deviant behaviors of their students than the other trainees.

Conclusions
From the data collected with the help of the DIOS, it can be deduced that the courses directed by trainee students during work readiness internships show a high degree of disruption, since there is a rate of 1.3 and 1.01 DB per minute. In addition, all Type 1 and Type 2 disruptive behaviors constitute 90% of inappropriate behaviors and they are a little influence on the life of the class when they occur. Whereas, third-level deviances, which actually interfere with the smooth running of the class from the moment they occur, are much less frequent and constitute between 7% and 8% of students' deviant behavior during sessions led by trainee students.
However, the frequency of onset of disruptive behaviors (DB1 and DB2) is slightly lower in sessions facilitated by trainees who have been trained in "Body language and public speaking". Similarly, at the level of disruptive behaviors (DB3), the trainees who underwent the training realized a greater decrease in the frequency of appearance of these behaviors.
Faced with these disruptive behaviors, the trainees who attended the training were slightly more interactive in their reactions during the sessions.
The results of our studies illustrate the reality of the practice of future teachers during the internship preparation to professional life. Indeed, they constitute a repertoire to perceive the different disruptive behaviors of students and the reactions of trainee students to these behaviors.
By this presentation, our work can certainly be used as part of the initial training of PET and in formalizing the professional skills repository.