Examination of attitudes to learning and educational stress in prospective primary school teachers: İzmir- Buca sample

Many factors interact with each other in learning and internalizing a subject along with performing a new task. Attitudes and stress are the two of these factors. The aim of this study was to examine attitudes to learning and educational stress in third and fourth year students as prospective primary school teachers. The relational model was used and data were collected with Educational Stress Scale for Adolescents and Scale for Attitudes to Learning. The population of the study includes the students in the Education Faculty of a university where the researcher worked and a total of 189 third-year and fourth-year students formed the study sample. While the students’ attitudes to learning differed in terms of subscales of educational stress, they either mostly agreed or were indecisive about their attitudes and stress. They also got low scores for educational stress; and a significant difference in attitudes to learning and expectations from learning in favor of the female students was observed. Similarly, the female students got higher scores for pressure from study, self-expectation and educational stress in general. No significant difference was found between the third-year and the fourth-year students in terms of their attitudes to learning, but the fourth-year students had a higher self-expectation. The sections of the students did not create a difference. There were relations between subscales of the scales. In view of these results, it can be suggested that educational stress can be reduced by using appropriate interventions designed to decrease worries about grades and workload and to support expectations of students.


INTRODUCTION
The factors affecting individuals during the education and teaching processes are teachers, learning environment, teaching methods and techniques, learning strategies, background information, interactions with family and social milieu, attitudes, motivation, anxiety, self-efficacy, self-respect, self-concern and educational stress.
Learning and internalizing a subject and performing a new task can be achieved by interplay of abovementioned and more factors.Although learning and internalizing a subject (Özden, 1997: 24) along with accomplishment of what has not been achieved before (Kara, 2010) can have various definitions such as source of stress by teachers, also complained about the forgoing sources of stress.Beausaert et al. (2016) performed four longitudinal studies between 2011 and 2014 to reveal sources of stress and burnout in 3572 administrators of primary and secondary schools in Australia.They showed that a person's surroundings could be sources of stress and that social support could reduce stress and burnout.
As emphasized in the literature above, stress can be due to personal characteristics, working conditions, insufficient resources and a person's surroundings.Considering conditions in Turkey, problems likely to appear after graduation can also be a cause of stress for the students obtaining the right to study at university after a very difficult exam, KPSS (a national exam administered in Turkey to employ individuals as state officers/teachers when they become fourth year students).Therefore, third-year students were enrolled into the present study.

Attitudes to learning
The competencies individuals achieve at the end of their learning period are related to their experiences during their learning period.One of the factors effective in this process is attitude.The term attitude is defined as intentions of individuals influencing their acceptance or rejection of the opposite (Başaran, 1990).Attitudes are acquired by means of modelling behavior of others, identification and experiences.Attitudes to learning may result from the same factors.
Since attitudes can be effective in learning as mentioned above, there have been studies about effects of attitudes on learning in various fields of study including science, mathematics, Turkish language, learning a foreign language, use of technology and acquisition of communication skills (Akamca and Hamurcu, 2005;Ünal and Ergin, 2006;Özgen and Pesen, 2008;Bosede, (2014) ;Yaman, 2014;Harb et al., 2014;Kovac and Zdilar, 2017;Wan and Lee, 2017).Also, there have been many studies showing that attitudes towards teaching as an occupation varied with gender, field of study, age and personality (Oral, 2004;Doğan and Çoban, 2009;Başbay et al., 2009;Bulut, 2009;Demirtaş et al., 2011;Bulut, 2011;Aslan and Yalçın, 2013;Edwards, 2014).Sade et al. (2007), in their study on attitudes towards online learning and Pierce et al. (2007, as cited in Kara, 2010), in their study on learning mathematics through technology found that attitudes of students affected their behavior during the learning process.Positive attitudes towards learning have been reported to cause greater attempts.

Relationships between Educational Stress and Attitudes to Learning
Educational stress resembles occupational stress due to work life.It can be affected by various factors.Relations between educational stress experienced during the learning process and various factors have been the subjects of several studies.Its relations with such factors as success, learning strategies (Laakkonen and Nevgi, 2014) and gender (Bonneville-Roussy et al., 2017) were dealt with in the literature.Yıldırım et al. (2017) examined the relations between educational stress and ways of coping with stress and self-respect, social support and general health status in 517 nursing students and found that educational stress and ways of coping with stress are significantly related to self-respect and social support and are affected by general health status.Aktürk (2012) carried out a study using Scale for Attitudes to Learning in 200 prospective teachers to reveal the relation between attitudes to teaching and reasons for wanting to become a teacher and academic performance.The researcher found that the participants wanting to be a teacher for internal reasons were more open to learning, had higher levels of expectations from learning and had lower anxiety about learning.In addition, a significant positive relation was detected between academic performance and the subscale nature of learning in Scale for Attitudes to Learning.Wang et al. (2015) performed a study to investigate learning pressure, learning attitudes and achievement in Macau undergraduates.Their study comprised 135 Chinese volunteers from two public universities and one private university in Macau.Out of 135 volunteers, 55 were male and 80 female, and 39 were first-year, 33 second-year, 34 third-year and 29 fourth-year undergraduates.They reported that learning pressure had a significant, moderate, and positive correlation with learning attitudes and had a negative correlation with academic achievement.
The relation between educational stress and attitudes as an emotional factor has also been the focus of interest in the literature.Izgar (2015) conducted a study on students at an education faculty (n=208) and on students taking pedagogical formation courses (n=107) to deal with both educational stress and learning attitudes.In his study, there was a significant difference in scores for attitudes to learning in favor of male students.However, there was not a significant difference in educational stress between genders.The researcher attributed this to the fact that all the students had severe stress due to KPSS regardless of gender.It seems to be important to search the relations between educational stress and various factors in order to elucidate problems experienced in education systems.Therefore, this study, using data collection tools similar to those in Izgar's study, was performed in a different sample and at a different university and time.

Aim and research questions
Cognitive, affective and psychomotor knowledge and skills of prospective primary school teachers become important considering their effects on their occupation after graduation.These will offer education to primary school students by using the abovementioned knowledge and skills, and thus equipping the students with the knowledge their teachers have and are affected by their teachers' attitudes and stress.
The aim of the study is to examine prospective primary school teachers' attitudes towards learning and educational stress levels.The term "prospective primary school teachers" refers to the "thirdand fourth-year students" included in the sample.This is the first study performed only on students studying primary education in an education faculty to determine prospective primary school teachers' attitudes to learning and educational stress levels.Prior research has not mostly focused on both attitudes to learning and educational stress.In fact, there have been only two studies about the relation between these variables, conducted by Izgar (2015) and Wang et al. (2015).Although Izgar's study used the same methodology as the current study, his study included both students in an education faculty and students not studying education but taking pedagogical courses.Wang et al.'s study comprised of first-year, second-year, third-year and fourth-year university students.The research questions of the present study are as follows: 1) What are the students' attitudes to learning and educational stress in general?2) Is there a difference in attitudes to learning and educational stress between the female and the male students?3) Is there a difference in attitudes to learning and educational stress between the third-year and the fourthyear students?4) Is there a difference in attitudes to learning and educational stress between the students in the four sections they were assigned into at the beginning of the term?5) Is there a relation between the students' scores for their attitudes to learning and those for their educational stress?

METHODS
The study is based on the relational model.This model deals with presence of a relation between two or more variables and its degree (Karasar, 1991: 81).In this study, the relation between students' attitudes to learning and educational stress was examined.

Sample and its characteristics
The study population included all the students in the Department Of Primary School Education at the university where the researcher worked.The reason for selection of this population was that it was easily accessible.Convenience sampling was used, and the study sample included the third-year students, who started practicum  1, the number of the female students was higher than that of the male students.This difference was also shown in other studies performed by Hamurcu (2006Hamurcu ( , 2010) ) and Pamuk et al. (2014) in the same study setting at different times.It may be that teaching as a profession is more popular with females.The reasons for the high number of female students could be examined in further studies.
The sample included a total of 189 students, of whom 136 were female and 53 were male.Out of 49 students in 4A, 17 were male and 32 were female.Out of 32 students in 4B, nine were male and 23 were female.Out of 52 students in 3A, 12 were male and 40 were female.Out of 56 students in 3B, 15 were male and 41 were female.

Data collection tools
Data were collected with Educational Stress Scale for Adolescents and Scale for Attitudes to Learning.Educational Stress Scale for Adolescents was developed by Sun et al. (2011) to measure levels of stress resulting from academic factors.It is a five-point self-report Likert scale and has five subscales and 16 items.The subscale pressure from study involves four items, workload three items, worry about grades three items, self-expectation three items and despondency three items.The scale was translated into Turkish and its validity and reliability for the Turkish population were tested by Akın et al. (2012).Construct validity of the scale was tested on 300 university students.According to the explanatory factor analysis made after achievement of the construct validity, Kaiser Meier Olkin value was 0.81 and the Bartlett's Sphericity test result was as follows: χ²: 3488.103.Sixteen items and five subscales explained 83% of the total variance.The internal consistency analysis made to determine the reliability of the scale showed that Cronbach's alpha was 0.86 for the scale and 0.87 for pressure from study, 0.93 for workload, 0.90 for worry about grades, 0.90 for selfexpectation and 0.91 for despondency (Akin et al., 2012).Cronbach's alpha was 0.77 for the scale in the present study.Scale for Attitudes to Learning was developed and its validity and reliability were tested by Kara (2010).It is a five-point Likert scale and has four subscales and 40 items.The subscale nature of learning involves seven items, expectation nine items, openness eleven items and anxiety thirteen items.The scale is composed of 29 positive items and eleven negative items.The construct validity of the scale was tested on 285 university students.According to the factor analysis for repeated measures, Kaiser Meier Olkin value was 0.79 and Bartlett's Sphericity value was as in the following: χ²: 3101,363.The internal consistency analysis, made to test the reliability of the scale, showed that Cronbach's alpha was 0.73 for the scale, 0.77 for nature of learning, 0.72 for expectations, 0.78 for openness and 0.81 for anxiety (Kara, 2010).Cronbach's alpha was 0.68 for the scale in the present study.

Analysis of data
Data obtained were analyzed with Statistical Package Program for Social Sciences for WINDOWS 17.0.Frequency, mean, standard deviation, mode and median were utilized for analysis of the data.Since the data did not have a normal distribution, the nonparametric tests, Kruskal Wallis, Mann-Whitney U test and Chisquare test and Pearson correlation analysis were employed for comparisons.p<0.05 was accepted as significant.

RESULTS
Obtained results are dealt with and discussed in accordance with the research problems.To deal with the first research question "What are the students' attitudes towards learning and educational stress", the data are presented in Table 2.
Table 2 shows mean scores for subscales of the scales.Since the data collection tools are five-point Likert scales, they have four ranges and each range corresponds to the score of 0.80.Depending on the number of the items in the subscales, ranges can be calculated.
The lowest and the highest scores for Scale for Attitudes to Learning are 40 and 200 respectively.The students got the mean score for the scale was 151.21, corresponding to "mostly agree".They had a positive attitude to learning in terms of the nature of learning,  (Akın, 2012: 105).In the present study, the students got the mean score of 49.14 for the scale, which corresponds to indecisiveness.However, they got 10.7, a high score for the subscale self-expectation.
The second research question of the present study was whether the students' attitudes to learning and educational stress differed in terms of their gender.Table 3 presents a comparison of the scores for attitudes to learning and educational stress between genders according to the analysis with Mann-Whitney U test.
As shown in Table 3, there was a significant difference in scores for attitudes to learning in general and the subscale expectation between genders (p<0.05).Similarly, the difference in scores for the subscales of educational stress, pressure from study and selfexpectation between the genders was significant (p<0.05).
The third research question of this study was whether there was a significant difference in attitudes towards learning and educational stress in terms of the year of study.According to the analysis with Mann-Whitney U test, the differences between the third-year and the fourth-year students in attitudes to learning and educational stress are shown in Table 4.
As presented in Table 4, there was not a significant difference in attitudes to learning between the third-year students and the fourth-year students.However, concerning with educational stress, there was a significant difference in self-expectations between the third-year and the fourth-year students p<0.05.This difference resulted from the higher scores of the fourthyear students for self-expectations.
The fourth research question of the present study was whether the students differed in their attitudes to learning and educational stress in terms of their sections.The students were receiving education in four different sections.They were assigned into these sections according to the last digit of the numbers in their student IDs at the beginning of the term.Kruskal Wallis-H test was performed to reveal possible differences.Table 5 shows the mean rank scores of the students in four sections and Table 6 reveals a comparison of the mean rank scores between the four sections; that is, 4A, 4B, 3A and 3B.
As shown in Table 5, there were small intragroup and intergroup differences in the mean rank scores for attitudes to learning and educational stress.To determine whether these differences were significant, Kruskal-Wallis analysis was performed.Chi-square and p values obtained through this analysis are shown in Table 6.
As seen in Table 6, no significant difference was found between the sections in terms of attitudes to learning and educational stress.The small differences shown in Table 5 were found to be insignificant.The students receiving education in four sections did not differ in their attitudes to learning and educational stress they experienced.
The fifth research question was whether there was a relation between the scores for attitudes to learning and those for educational stress.Pearson correlation analysis was made to determine the presence of this difference and the results of the analysis are presented in Table 7.
As demonstrated in Table 7, significant relations were detected between several subscales of Scale for Attitudes to Learning and those of Educational Stress Scale for Adolescents.The correlation coefficients 0.70-1.00indicate a strong correlation, 0.70-0.30a moderate correlation and 0.30-0.00 a weak correlation.Negative correlation coefficients show an inverse relation (Büyüköztürk, 2002: 31-32).Accordingly, as shown in Table 7, the following results were obtained: No significant relation was found between the total scores for Educational Stress Scale for Adolescents and those for Scale for Attitudes to Learning (r= 0.363; p= 0.392).However, there was a significant relation between the scores for Educational Stress Scale for Adolescents and those for the subscales of Scale for Attitudes to Learning and between the scores for Scale for Attitudes to Learning and the scores for the subscales of Educational Stress Scale for Adolescents.
There was a significant, positive weak relation between the score for the subscale self-expectation in Educational Stress Scale for Adolescents and the score for Scale for Attitudes to Learning (r=0.166;p=0.023).
A significant, positive weak relation was also found between the score for the subscale despondency in Educational Stress Scale for Adolescents and the score for Scale for Attitudes to Learning (r=0.148;p=0.043).
No significant relation was found between the score for the subscale nature of learning in Scale for Attitudes to Learning and the score for Educational Stress for Adolescents.
A significant relation was found between the score for the subscale expectation in Scale for Attitudes to Learning and the score for Educational Stress Scale for Adolescents and its three subscales.There was a moderate negative correlation between expectation and educational stress in general (r= -0.229; p=0.002), pressure from study (r= -0.222; p=0.002), workload (r= - A significant relation was detected between the score for the subscale openness in Scale for Attitudes to Learning and the scores for Educational Stress Scale for Adolescents and its three subscales.A negative moderate relation was found between the score for openness and the scores for Educational Stress Scale (r= -0.352; p=0.000) and its subscales pressure from study (r= -0.324; p=0.000) and despondency (r= -0.334; p=0.000).A negative weak relation was detected between the score for openness and the score for the subscale workload (r= -0.266; p=0.000).
There was a significant positive relation between the score for the subscale anxiety about learning in Scale for Attitudes to Learning and the scores for Educational Stress Scale for Adolescents (r=0.468;p=0.000) and its four subscales.Anxiety had a significant moderate relation with pressure from study (r=0.407;p=0.000) and despondency (r=0.518;p=0.000) and a significant, weak relation with workload (r= 0.248; p=0.001) and selfexpectation (r=0.243;p=0.001).

DISCUSSION
In the present study, the students were found to have a positive attitude towards learning.In contrast with this finding, Izgar (2015: 393) reported that the students got lower scores for their attitudes to learning.In Izgar's study, the mean score was 91.30 in the students at the Education Faculty and 87.49 in the students taking Pedagogical Formation courses, with a significant difference (p<0.05).The researcher ascribed this difference with the idea that the students at the Education Faculty were more enthusiastic with becoming a teacher.
In the current study, the students received low scores for educational stress; however, they agreed that selfexpectations created educational stress.It can be suggested that they did not have high levels of educational stress, compatible with the results reported by Izgar (2015: 393).In Izgar's study, the students at the Educational Faculty had a mean score of 48.8.
Causes of stress have not been elucidated completely and have been classified differently in the literature.Gupta (1981) categorized them into environmental, organizational and personal factors (Gupta, 1981, as cited in Buluş, 1999: 67).Izgar (2008) identified six factors as stated in Introduction; that is, physical, psychological, social, mental, and work related and temporary factors (Izgar, 2008, as cited in Izgar, 2015: 387-388).The reason for the changes in classification of these causes is that stress is created by interplay of numerous factors.Gender is one of these factors.While some studies did not show a relation between gender and stress (Chan, 2002;Cam, 2004;Durna, 2006;Erdoğan et al., 2009;Izgar, 2015), others revealed a significant relation between them (Aysan, 1998;Sökmen, 2005).These conflicting findings might be due to differences in setting and time of studies, characteristics of samples and multiplicity of factors causing stress.
In the present study, the female students got significantly higher scores for expectations from learning, pressure from study and self-expectations.This suggested that the female students had a higher level of educational stress due to their expectations from learning, pressure from study and self-expectations.However, Durna (2006: 374) in a study on 378 university students did not find a significant difference in stress severities between male and female students.In addition, Izgar (2015: 393) did not show a significant difference in educational stress between the male and the female students (t=-0.36; p= 0.72; p>0.05).The researcher attributed this to the fact that both male and female students were getting prepared for KPSS, which would take place soon.
Although several studies showed that gender did not have a significant influence on attitudes to learning (Saracaloğlu, 2000), other studies revealed a significant difference between genders (Izgar, 2015;Wan and Lee, 2017).The present study revealed that the female students had higher self-expectations.It may be that they were more willing and assiduous to receive education and have an occupation.In contrast with the finding in the present study, Izgar (2015: 392), in his study on 182 female students and 133 male students found a significant difference in attitudes to learning in favor of the male students (t=-2.03;p=0.04; p<0.05).The researcher suggested that the male students were more willing to learn.The conflict between the current study and Izgar's study might have been due to differences in stress related factors.Like Izgar, Wan and Lee (2017) showed that male students had significantly more favorable attitudes to science in terms of the subscales selfconcept in science, enjoyment in science, learning in and outside the classroom and future participation.
The year of study did not have a significant influence on attitudes to learning.This indicates that the year of study was not predictive of attitudes to learning.However, it had a significant influence on expectations.In fact, the fourth-year students had higher expectations about their academic success and their life in the future.Therefore, they experienced more severe educational stress.Considering that the subscale self-expectation has items about hopes for the future, the finding is indicative of their stress and anxiety about their life in the future.
In the current study, self-expectations and despondency had a positive weak relation with attitudes to learning.This suggests that self-expectations and despondency slightly affect attitudes to learning.The nature of learning had no significant relation with educational stress.This indicates that it did not produce educational stress.
Educational stress in general, pressure from study, workload and despondency had moderate, negative correlations with expectation.This finding is suggestive of a moderate, negative effect of educational stress in general, pressure from study, workload and despondency on expectations related attitudes.
Educational stress in general, pressure from study and despondency had a significant negative moderate relation with openness.These findings are suggestive of a moderate effect of educational stress in general and its subscales pressure from study and despondency.Workload had a significant negative weak correlation with openness.This suggests presence of a mild influence of workload on openness related attitudes to learning.

Hamurcu 101
The relation between educational stress and anxiety related to attitudes to learning was significantly positive.This finding shows presence of a linear relationship between anxiety about learning and educational stress and that as one increases so does the other (Büyüköztürk, 2002: 32).It can be suggested that anxiety is influenced by some aspects of educational stress.
In Izgar (2015: 394) study, the score for Scale for Attitudes to Learning had a significant negative weak relation with the scores for the subscales worry about grades (r=-0.16)and self-expectation (r=-0.23,p<0.01) in Educational Stress Scale for Adolescents.It means that as the scores for worry about grades and self-expectation increased, the scores for attitudes to learning decreased, which conflicts with the results of the present study.The conflict between the findings in Izgar's study and those of the present study might have resulted from the differences between the samples.Especially, inclusion of the students taking pedagogical courses in Izgar's study might have had an impact on the difference.In fact, as emphasized before, significant differences were found between the students in the education faculty and those taking pedagogical courses in terms of attitudes to learning and educational stress.This might have an influence on the correlation analysis of the scores for both scales.

CONCLUSION AND RECOMMENDATIONS
This study, directed towards revealing relations between attitudes to learning and educational stress, show that although the students' attitudes to learning varied with subscales of educational stress, the students either mostly agreed or were indecisive about their attitudes and stress.Their low scores for educational stress indicate that they had low levels of stress.Since high levels of stress can be effective in academic performance, their low scores for educational stress seem to be favorable.The comparisons between genders revealed differences in attitudes to learning and educational stress and their subscales in favor of the female students.Although the year of study did not create a difference in attitudes to learning, the fourth-year students had a higher self-expectation.The sections of the students did not cause a difference.Attitudes to learning were associated with some aspects of educational stress.
In light of the results of this study, the following recommendations can be made: 1) The study was performed on a small sample of the students in a single Education Faculty.Therefore, further studies should be conducted in larger samples.2) Data were collected only through quantitative research tools like scales.It can be recommended that qualitative data be gathered with such tools as interviews and focus group interviews.
3) Detection of relations between attitudes to learning and educational stress can provide guidance for arrangement of the learning environment.Taking account of effects of attitudes on academic success, appropriate interventions directed towards minimization of worries about grades and workload and supporting expectations should be offered to reduce academic stress of students.The units and specialists responsible for reduction of educational stress in students at universities could be appointed to conduct these interventions.4) Stress and anxiety can affect academic performance.This may prevent students from receiving sufficient scores to pass KPSS in Turkey, required to get a job in state organizations.Therefore, the Turkish Ministry of Health and other policy makers should pay attention to the relation between academic performance and stress and anxiety.The results of the present study can help them be aware of the problems experienced by prospective primary school teachers, take appropriate precautions and provide appropriate support for these prospective teachers.

Table 1 .
Characteristics of the students included into the sample (n=189).
classes, and the fourth-year students, who were getting prepared for KPSS.A total of 244 third-and fourth-year students completed the data collection tools at the end of the academic year of 2016-2017.After elimination of the measures with missing responses, data from 189 students were analyzed.The response rate was 77.46%.Table1presents characteristics of the students included into the sample.As shown in Table

Table 2 .
Descriptive characteristics of data from Educational Stress Scale for Adolescents and Scale for Attitudes to Learning (n=189).

Table 3 .
Comparison of scores for Attitudes to Learning and Educational Stress between genders.

Table 4 .
Comparison of Attitudes to Learning and Educational Stress between Third-Year and Fourth-Year Students.

Table 5 .
Distribution of mean rank scores for Attitudes to Learning and Educational Stress in four sections.

Table 6 .
Comparison of mean rank scores for Attitudes to Learning and Educational Stress between the sections.

Table 7 .
Results of the correlation analysis for the Relation between Attitudes to Learning and Educational Stress.