The Effect of the Use of Discussion Types on the Listening Skills of Convicts / Prisoners in Training-Improvement Activities Applied in Criminal Execution Institutions

Training and rehabilitation activities for convicts and detainees are of utmost importance for a state’s national policy in terms of their targeted lives after release. The purpose of this study is to put forward whether socio-cultural activities such as debate, panel, forum, open session used in education and training activities for the rehabilitation works performed for prisoners are effective in practice or not and making detections on the issue that activities related with improving the listening skills of prisoners can be improved or not. This research was carried out with 122 volunteer participants selected with random method from convicts/detainees in penal institutions of Bursa province who graduated from primary school, secondary school, high school and university. The study has been analyzed by five different achievements tests which have been designed by the researcher to measure the ability to understand the listening comprehension success and it has been analyzed with a statistical analysis software. The study reveals that with the use of discussion types the success of convicts/prisoners in listening comprehension has showed a significant difference between pre-test and post-test.


Introduction
The perception of the concepts of punishment and execution differed in every period of history (Foucault, 1977), the execution of punishments differed in historical periods (Welch, 2005). The main purpose is to regain and socialize an individual whose security measures are applied to him in criminal execution institutions (CEI) (Smit and Snacken, 2009). While this goal is realized, the society is protected indirectly. Overseeing crime and punishment balance, it is extremely important in terms of both reclaiming and regaining the individual who commits a crime and establishing justice.
Historical experience shows us that the prison management approach, which is based on strict security and enforcement, does not show much positive results for the improvement of convicts / detainees. (Smit and Snacken, 2009: 38-41) The strictly binding solid face of the prisons -the understanding of external authority -did not have dissuasive consequences for the individual (Foucault, 1977: 233), White and Fiona (2004: 215). The modern management concept, which has been in effect for a long time in the world, foresees the mission of completing the period of convicts / detainees in the most productive way, and increasing their personal development in social, cultural, mental, physical and professional means and returning them to the society in a beneficial way (Urgan, 2020: 72-74). The point reached in the subjects of sociology and human-social psychology is now in the form of imprisonment by being treated and educated until the time in which prisoners can regain their citizenship rights (McConville, 2000;Akt: Ağcakale, 2010: 63).
Observing an individual who has been involved in a crime or who has been charged with a crime from the moment he / she comes to the criminal execution execution institution, to be tested in the process, to design and implement appropriate educational plans, in this way it is extremely important to develop the remarkable aspects of the individual in a positive way (Smit and Snacken, Considering these training / improvement activities, one of the most important problems to be solved is activities on listening and speaking. Because convicts / detainees who do not understand what they listen to while explaining their own incentives and listen to them with imperfection in the opposite side find themselves in an environment of verbal conflict. This situation is a frequently encountered one that convicts / detainees in private living spaces with their friends and also through their meetings with the public authorities. Problems such as not understanding what you listen to, designing your own words while communicating, habit of speaking, and not being able to focus on the messages of the other party based on prejudices are frequently observed. Moreover, since a culture of discussion based on the on its principles could not be established, every discussion made results in a fight and unrest. Failure to communicate creates an atmosphere of psychological tension and conflict.
As a result, this situation appears as an obstacle to problem solving in criminal execution institutions.
The aim of this study is to reveal that whether social-cultural activities such as debates, panels, conferences, which are used in training and improvement activities and expressed in the training charts in each activity period, contribute to the improvement of convicts / prisoners in practice, whether or not their activities can be improved to make determinations.

Method
In this quantitative study, a single-subject quasi-experimental research method was used, and repeated measurements were made to examine the effect of different dependent variables. In singlesubject studies, the functional relationship between the dependent variable and the independent variable is revealed by comparing the application data with the baseline level of the subject (Tekin, 2000). In this experimental study, a single group pre-test-post-test model was applied.

Participants
The participants of this study are randomly selected amongst the prisoners located in prisons in Bursa, namely Bursa E-Type Closed and Open Penitentiary Institution, Bursa H-Type Closed Penitentiary Institution and Yenişehir Female Closed Criminal Execution Institution, excluding terror crime detainees / convicts. 122 convicts / detainees in those institutions voluntarily agreed to participate into this study. However, 26 persons, out of 122 convicts / detainees who had succesfully participated participating in the pre-test, did not participate into the post-test application due to the force majeure such as referral / transfer / evacuation. Thus, 96 persons in total participated to the posttest application.

Collection of Data
Data set of the research was procured in the form of the first listening comprehension achievement test (1st LCAT) titled "Crime and Punishment", developed with the opinion of the experts in the field, the 2nd listening comprehension achievement test titled "Family Society and World Relationship" (2nd LCAT), 3th Listening Comprehension Achievement Test (3rd LCAT) titled "Pain", 4th Listening Comprehension Achievement Test (4th LCAT) titled "Dangerous Enemies of the Path to Achievement" and 5th Listening Comprehension Achievement Test (5). LCAT) titled "Study the Mortar, Study, Remembering the Fruit of the Work". Listening texts were chosen in line with adult education rules (taking into account child prisoners / convicts) and within the framework of the education-improvement vision envisaged to be implemented in criminal execution institutions.
These achievement tests are arranged in accordance with the content of the texts played and the acquisition gains. Before applying achievement tests, information such as the subject and purpose of the research, what the tests do and how they should be answered were given, and the questions of the participants were answered.

Data Collection Tools
In order to determine the effect of the use of discussion types in the training and improvement activities of convicts/detainees accommodated in criminal execution institutions, five different listening comprehension tests, which appeal to different values, have been developed. The texts used in the achievement tests with the values to be processed during the application process were presented to the expert opinion and were selected in harmony. 1th Listening Achievement Achievement Test (1st LCAT) titled "Crime and Punishment": It was created considering the levels of convicts / detainees for the value of justice. 2th Listening Achievement Test (2nd LCAT) titled "Family Society and World Relationship": It was created considering the levels of convicts / detainees for family value. 3th Listening Achievement Test (3rd LCAT) titled "Pain": It was created considering the levels of convicts / detainees for empathy value. The 4th listening comprehension achievement test (4th LCAT) titled "Dangerous Enemies of the Path to Achievement": It was created considering the levels of convicts / detainees for achievement value. The 5th listening comprehension achievement test (5th LCAT) titled "Study the Mortar, Study, Remembering the Fruit of the Work": It was created by considering the levels of convicts / detainees for the diligence value.

Findings for Validity and Reliability Studies of Achievement Tests Used in the Research
In order for the achievement tests used in the research to be highly valid and reliable, a reliable achievement test was created and applied to convicts / detainees by providing the achievement test development steps and conditions. For this reason, item analysis was conducted to ensure the structural validity of the test items, and item discrimination indices and difficulty levels calculated in the item analysis are given in Table 1.
When the table is examined, it can be seen that the difficulty indexes of the tests ranged between 0.25 and 0.94, and the discrimination indexes ranged between -0.50 and 1.00. High item discrimination increases the validity of the test. If the item's discrimination index is 0.40 and above, the item is very good, if it is between 0.30-0.39, the item is quite good, if it is between 0.20-0.29, the item can be used in mandatory cases, but it needs to be corrected and improved. If the item is 0.19 and smaller, it is very weak, if it cannot be improved by corrections, it should be removed from the test (Turgut, 1995;Akt: Tekin, 2000).
Then, with the pilot application, achievement tests consisting of 14, 16 and 11 questions were applied to 96 convicts / detainees, and the item analysis was repeated. The discrimination index scores that emerged according to the item analysis are presented in Table 3.
The reliability of the tests was calculated using the Kuder-Richardson 20 (KR-20) formula on 14, 16 and 11 items, respectively. KR-20 reliability coefficient of 1 st test is 74. KR-20 reliability coefficient of the 4 th achievement test is 80. Accordingly, the KR-20 reliability coefficient of the 5 th achievement test is 79. It can be said that the achievement tests numbered 1, 4 and 5 are valid and reliable tests. transfer, evacuation, etc.) and after the application process, the study was finalized by applying posttests to 96 participants. Achievement tests used in the application process are within the scope of values education and chosen in relation to the discussion topics.

Data Analysis
In the research, firstly, demographic characteristics of the participants were examined. In this context, parametric and nonparametric tests were analyzed to test the hypotheses of the research.
Whether the data are suitable for normal distribution was examined with the Shapiro Wilk test.
With the Kolmogorov-Smirnov test, it is possible to examine whether the data collected from a sample show normal distribution (Altunışık et al., 2001: 165). The significance level of the result value of the Kolmogorov-Smirnov test is greater than 0.05 indicates the suitability of the data for normal distribution (Karaatlı, 2010: 10). Descriptive statistics for continuous variables are given in those who are suitable for normal distribution (mean ± standard deviation) and those who are not suitable for normal distribution (median (minimum-maximum)). Descriptive statistics for categorical variables are given as frequency and percentage. In order to determine the significant difference according to the variables for the hypotheses, the t-test which is parametric in those who show normal distribution in cases where two groups are present in independent group comparisons of continuous variables, the non-parametric Mann Whitney U test in those who do not comply with normal distribution and in the case of more than two groups. One-Way ANOVA test, which is parametric, was used, and nonparametric Kruskall Wallis test was used for those who did not conform to normal distribution. When there was a significant difference in the parametric One-Way ANOVA test, the post hoc Tukey test Educational Policy Analysis andStrategic Research, V 15, N 4, 2020 © 2020 INASED 93 was used to determine which groups the difference was between. When a significant difference was found in the nonparametric Kruskall Wallis test, the non-parametric Mann Whitney U test was used to determine which groups the difference was between.
Reliability of achievement tests in this study was found by making item analysis and calculating KR-20 values. In addition, in the test of the hypotheses included in the study, α value was taken as 0.05. Therefore, the analysis results in the study were interpreted at the 95% confidence level.
The analyzes in the study were obtained by using statistics program.

1-Demographic Findings of the Research
In the research, the following findings related to the socio-demographic characteristics of the participants were reached.  Table 4 shows the distribution of convicts / detainees living in criminal execution institutions according to their demographic characteristics.

2-Significance of Achievement Tests
The results of the 5 achievement tests developed are analyzed according to pre-test and posttest. The pre-and post-tr6ial scores of the convicts/detainees who participated in the study from the listening comprehension achievement tests are given in the tables below. According to the test results given in Table 5, there is a significant difference between the preand post-trial scores of the convicts/detainees taking the achievement Test titled Crime and Punishment (p<0.05). When sequence totals of difference scores are taken into account, this observed difference appears to favor positive rows, the final test score. According to these results, it can be said that the education was productive. According to the test results given in Table 6, it is seen that there is no significant difference between the scores of the convicts / detainees participating in the research before and after the experiment they received from the 2nd listening comprehension achievement test (p> 0.05). According to the results, it can be said that the training did not change the achievement of convicts / detainees. According to the test results given in Table 7, it is seen that there is no significant difference between the scores of the convicts / detainees participating in the research before and after the experiment they got from the 3rd listening comprehension test (p> 0.05). According to the results, it can be said that the training did not change the achievement of convicts / detainees. According to the test results given in Table 8, it is seen that there is a significant difference between the scores of the convicts / detainees participating in the research before and after the experimental scores (p <0.05). When the rank totals of the difference scores are taken into consideration, it is seen that this observed difference is in favor of the Positive Rank, the post-test score. According to these results, it can be said that the education provided was productive. According to the test results given in Table 9, it is seen that there is a significant difference between the pre-test and post-test scores of convicts / detainees participating in the study from the 5th listening comprehension achievement test (p <0.05). When the rank totals of the difference scores are taken into consideration, it is seen that this observed difference is in favor of the positive ranks and post-test points. According to these results, it can be said that the education provided was productive.

3-Evaluation of Findings Related to the Pre-test and Post-test Achievement Scores According to the Demographic Features of the Participants
Below are the results of pre-test and final-test achievement scores before listening activities based on discussion types based on demographics of prisoners/detainees who participated in the study.
In the case of the two groups, showing conformity to a normal distribution the parametric t-test to show conformity to a normal distribution non-parametric Mann-Whitney U test for showing conformity to a normal distribution in the case of more than two groups parametric one-way ANOVA test, normal distribution show compliance non-parametric Kruskal-Wallis test was applied. According to the characteristics of convicts / detainees in Table 10, there is no significant difference between 1st LCAT pre-test achievement score variable according to the participants variable, because p = 0.566 and 0.05 is more than the variable of participants. There is a significant difference between the 1st LCAT post-test achievement score variable according to the participants variable since p = less than 0,027 and 0,05 for the 1st LCAT post-test achievement score variable. The difference was caused by convicts / detainees living in Yenişehir Woman Closed CEI. According to the characteristics of convicts / detainees in Table 11, there is no significant difference for the 4th LCAT pre-test achievement score variable since p = 0,691 and more than 0.05.

Evaluation of pre-test and post-test achievement scores by Participants
There is no significant difference for the 4th LCAT post-test achievement score variable since p = 0,753 and more than 0.05.  Table 12 shows the averages of 5th LCAT pre-test and post-test achievement scores for the participants characteristics of convicts / detainees. Since the p-test achievement score variable is less than p = 0.002 and 0.05, there is a significant difference according to the participants variable. The difference was caused by convicts / detainees living in Bursa H Type CEI. For the LCAT post-test achievement score variable, there is a significant difference compared to the participants variable since p = 0.044 and less than 0.05. The difference was caused by convicts / detainees living in Bursa Açık CEI.

4-Evaluation of Pre-test and Post-test Achievement Points According to the Conference Activity
As a result of the normality test, since the 1st LCAT pre-test and the 4th LCAT pre-test data were observed to come from the normal distribution, it was analyzed with the t-test, which is a parametric test type and used to test the variables in which the two groups are involved. As a result of the normality test, it was observed that the data of the 1st LCAT post-test, the 4th LCAT post-test, and the 5th LCAT pre-test and post-test did not come from the normal distribution. Therefore, for these tests, the Mann Whitney U test, which is a non-parametric test type and used to test the variables in which the two groups are involved, is used.  There is no significant difference according to the conference activity variable since p = 0,907 and 0,05 for the 4th LCAT pre-test in Table 14. Since p = less than 0,044 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the conference activity variable. Since there is more than p = 0.871 and 0.05 for the 5th LCAT pre-test in Table 15, there is no significant difference according to the conference activity variable. Since p = less than 0,027 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the conference activity variable.

5-Evaluation of Pre-test and Post-test Achievement Points According to Debate Effectiveness
The average scores of the convicts / detainees before and after the application were examined.
As a result of the normality test, it was observed that the data of the 1st LCAT post-test, the 4th LCAT pre-test and the 5th LCAT post-test came from the normal distribution. Therefore, whether the scores of achievement differ according to the effectiveness of the debate was analyzed with the t-test, which is a parametric test type and used to test the variables in which the two groups are involved. Since the 1st LCAT pre-test, the 4th LCAT post-test and the 5th LCAT pre-test did not come from the normal distribution, the Mann Whitney U test, which is a non-parametric type of test and used to test the variables of two groups, was used. There is no significant difference according to the debate effectiveness variable in Table 16, since p = 0,400 and 0,05 for the 1st LCAT pre-test. Since p = less than 0,017 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the debate efficiency variable. There is no significant difference according to the debating effectiveness variable since Table   17 has more than p = 0.826 and 0.05 for 4th LCAT pre-test. Since p = less than 0.032 and 0.05 for the post-test, there is a significant difference between the post-test achievement score variable according to the debate efficiency variable. Since p = 0,432 and 0,05 for the 5th LCAT pre-test in Table 18, there is no significant difference according to the debate effectiveness variable. Since p = less than 0,048 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the debate efficiency variable.

6-Evaluation of Pre-test and Post-test Achievement Points by Forum Activity
The average scores of the convicts / detainees before and after the application were examined.
As a result of the normality test, it was observed that the data of the 1st LCAT post-test, the 4th LCAT pre-test and the 5th LCAT post-test came from the normal distribution. For this reason, whether the scores of achievement differ according to the effectiveness of the forum was analyzed with the t test, which is a parametric test type and used to test the variables in which the two groups are involved.
Since the 1st LCAT pre-test, the 4th LCAT post-test and the 5th LCAT pre-test did not come from the normal distribution, the Mann Whitney U test, which is a non-parametric type of test and used to test the variables of two groups, was used. Since there is more than p = 0.107 and 0.05 for 1st LCAT pre-test in Table 19, there is no significant difference according to forum activity variable. Since p = less than 0,016 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the forum activity variable. There is no significant difference according to the forum activity variable since p = 0,245 and 0,05 for 4th LCAT pre-test in Table 20. Since p = less than 0,013 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the forum activity variable. Since there is more than p = 0.309 and 0.05 for the 5th LCAT pre-test in Table 21, there is no significant difference according to the forum activity variable. Since p = less than 0,045 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the forum activity variable.

7-Evaluation of Pre-test and Post-test Achievement Points by Panel Activity
The average scores of the convicts / detainees before and after the application were examined.
As a result of the normality test conducted, it was observed that the data of 1st LCAT, 4th LCAT and 5th LCAT pre-test and post-test did not come from the normal distribution. For this reason, whether the scores of achievement differ according to the panel effectiveness was analyzed with the Mann Whitney U test, which is a non-parametric test type and used to test the variables with the two groups.  Since there is more than p = 0,729 and 0,05 for the 4th LCAT pre-test in Table 23, there is no significant difference according to the panel activity variable. Since p = less than 0,041 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the panel activity variable. Since p = 0.095 and 0.05 for the 5th LCAT pre-test in Table 24, there is no significant difference according to the panel activity variable. Since p = less than 0,048 and 0,05 for the post-test, there is a significant difference between the post-test achievement score variable according to the panel activity variable.
In addition, after the analysis, the variables of gender, age, education level, income level, participation in courses, participation in cinema activity, number of books read, and prison type are among the pre-test and post-test scores before the application of listening activities based on discussion types. There is no significant difference between.
Considering the demographic findings of the research, the significance of the achievement tests, the pre-test and post-test achievement scores before the listening activities based on the types of discussion based on the demographic characteristics of the participants, the following summary table was created based on these findings:

Discussion and Conclusion
This research was shaped on the question "What is the effect of the use of discussion types on the listening skills of convicts / prisoners" and at the end of the research, the positive effect of the use of discussion types on education-improvement activities was determined. Doğan (2010), in his study entitled "Making Use of Activities in Improving Listening Skill", discussed the education of listening skill and at the end of the research, it was shown that gamification of activities was important for more effective participation of students. In this study, it was revealed that the use of discussion types had positive effects on the listening skill of convicts / prisoners. Epçaçan (2013) concluded that listening skill is a developable skill in the study of listening skill education and listening techniques, and that listening skill is a developable skill and especially in the measurement dimension of listening education and the need for more studies in this field. In addition, the application of strategies and techniques to improve listening skill has been put forward as an element affecting listening skill. At the end of this research, the positive effect of the use of discussion types on listening skill was found in training-improvement activities. Considering the use of discussion types within a technique to improve listening skill constitutes a similar aspect of this research in terms of the results presented.
Aytan (2011) examined the effects of active learning techniques on listening skill in her study "Effects of Active Learning Techniques on Listening Skill". It has been determined that the education given with active learning techniques improves students' listening skills. It can be thought that the above mentioned study shows similarity with this research in which the positive effects of the use of discussion types on listening skill were determined.
If a general evaluation is made in the light of this information, when the literature is examined, it is remarkable that some academic studies have been carried out for criminal execution institutions.
Crabbe claims that offender learning should be about values rather than about costs (Crabbe, 2016).
He adds that prison education needs to be inspiring and motivational and it necessiates links to employment and 'real life' on release. Cormac assesses that while prison education can work with, it needs to distinguish itself from, state-sponsored rehabilitation programmes and stand on the integrity of its profession, based on principles of pedagogy rather than be lured into the evaluative and correctional milieu of modern penality (Behan, 2014). It will be possible to gather these studies under the main headings such as historical development of institutions, periodical approaches, the concept of imprisonment, the legal framework of punishment, regional prisons, security principles of the institutions, personnel attitudes and problems (Focault, 1977). Research in the literature does not reveal an approach to language skills of convicts / detainees (both children and adults). It seems that there has been no study on the listening skills of convicts / detainees so far.
In this study, the effects of the use of discussion types on the listening skills of convicts / prisoners in the training and improvement activities implemented in criminal execution institutions were examined with some sub-problems. Five different listening comprehension tests were applied to convicts / detainees. By looking at the findings of these tests and other findings related to the research, it is possible to reach the following results: It was determined that there was a significant difference between the scores of the convicts / detainees participating in the study before and after the 1st LCAT. When the rank totals of the difference scores are taken into consideration, it is seen that this observed difference is in favor of the positive ranks and post-test points. According to this finding, it can be said that the education provided was efficient. When 1st LCAT findings are considered as a whole, it can be said that the significance before and after the experiment was mainly caused by the group of participants in Bursa E Type Closed Penitentiary Institution.
It was found that there was no significant difference between the scores of the convicts / detainees participating in the research before and after the experiment they received the 2nd LCAT and the 3rd LCAT. According to this finding, it can be said that the training did not change the achievement of convicts / detainees.
It was determined that there was a significant difference between the scores of the convicts / detainees participating in the study before and after the 4th LCAT. When the rank totals of the difference scores are taken into consideration, it is seen that this observed difference is in favor of the positive ranks and post-test points. According to this finding, it can be said that the education provided was efficient. Considering the 4th LCAT findings as a whole, it is possible to talk about a achievementful educational process in general, not on the basis of the institution of execution, since there is a statistically significant borderline significance throughout the test.
It was determined that there was a significant difference between the scores of convicts / detainees participating in the study before and after the 5th LCAT. When the rank totals of the difference scores are taken into consideration, it is seen that this observed difference is in favor of the positive ranks and post-test points. According to this finding, it can be said that the education provided was efficient. 5th LCAT, when findings are considered as a whole, it can be said that the significance level before and after the experiment was mainly caused by the participant group in Yenişehir Women's Closed Criminal Execution Institution, Bursa E-Type Closed Criminal Execution Institution and Bursa H-Type Closed Criminal Execution Institution.
When examining the demographic data, the results of which were determined were based on the three scales named as the 1st LCAT, the 4th LCAT and the 5th LCAT, and the sub-problems were examined considering these scales. The 2nd LCAT and the 3th LCAT were not evaluated because they did not give meaningful results.
When all the scales are considered as a whole according to the participants characteristics of the convicts / detainees participating in the research, it can be mentioned that the participants has an impact on the education process performed throughout the tests. In addition, it can be said that the variables of gender, age, education level, income level, participation in the course activity, participation in the cinema activity, number of books read, type of criminal execution institution found in the study have no effect on the education process.
When the convicts / detainees participating in the research are evaluated according to the variables of conferences, debates, forum and panel activities, it is seen that the theoretical and practical training process on the types of conferences, debates, forums and panels has been influenced by the statistical significance between the results of the pre-tests and the results of the post-tests.