To answer this research question, Pearson correlation analyses were conducted to discover the movements among the research variables, i.e., teachers’ self-efficacy, language proficiency and their pedagogic strategies. Davis’s (1971) criteria were used to describe the magnitude of correlations. According to the criteria, r = .70 or higher meant very high; .50 to .69, substantial; .30 to .49, moderate; .10 to .29, low;
and .01 to .09, negligible.
Overall, the results showed that all research variables formed substantial or moderate positive correlations with one another, ranging from .575 to .221, and reached significant levels (P<.001), except on teachers’ pedagogic strategies away from CLT orientations. That is to say, the higher language proficiency a teacher self-perceived, the more efficacious a teacher was and the more confident a teacher felt in engaging students, managing the class, utilizing effective or innovative instructional strategies, and adopting CLT approach in English class. However, the language proficiency and self-efficacy of the participants showed little connection to teachers’ adoption in form-focused approach or avoidance of CLT approach.
Therefore, the results were not displayed in the table.
Among all the significant correlations, the strongest relationship was found between instructional strategies and English proficiency (r = .575). Among the four skills, all the correlations of language skills reached substantial levels with efficacy for instructional strategies, except for listening skills which only moderately
correlated with efficacy for instructional strategies. In another word, the data means teachers’ proficient language skills could increase their capabilities in instructional strategies that included employing versatile methods to meet the individual needs of
71
students, facilitating students’ comprehension, providing suitable challenges for different students, and designing multifaceted assessments. Among the sub-scales of efficacy domains, efficacy for classroom management formed lower correlations with teachers’ language proficiency and their four skills. The results indicate classroom management demanded not so much language abilities from teachers in their English class as student engagement and instructional strategies did. The
weakest correlation was between classroom management and listening proficiency (r
= .221). The results exhibited during the listening activities, classroom management skill was less required.
Compared to classroom management (r = .290 to .221) and student engagement (r = .465 to .406), instructional strategy dimension was found to have a stronger relationship with teachers’ English proficiency (r = .575 to .490). The meaning of the data was that English proficiency across four skills helped a teacher to perform better in instructional strategies, which included orchestrating various teaching approaches to meet different students’ needs, utilizing a variety of assessment more effectively, and answering students’ questions and doubts more satisfyingly, than in other efficacy dimensions, such as dealing with disruptive behaviors in a class and motivating students’ learning. When further compared with the correlation
coefficients among four language skills and the subscales of teachers’ self-efficacy, speaking formed stronger relationships with the research variables than any other language skills. This finding revealed teachers’ oral proficiency played a more important role in building up teachers’ efficacy for student engagement, for classroom management, and for instructional strategies than other language skills did.
Language proficiency in general formed moderate relationships (r = .438) with
72
teachers’ tendency to CLT approach, so did four skills respectively (from r = .401 to .410). All the correlations reached significant level. The data pointed out that teachers’ language proficiency formed a decisive factor to use CLT approach in their language class. The more proficient a teacher was in four skills, the better he/she would employ the spirits of CLT approach in language class. When the discrepancies among the four skills were further compared, the difference of their correlation coefficients was not very distinctive. This phenomenon indicated CLT approach demanded all four skills from teachers to instruct a language class.
As for the relationships among CLT approach and teachers’ self-efficacy, the correlation coefficients all proved to be moderate and statistically significant (from r
= .487 to .344). The results signified teachers’ self-efficacy helped their employment of CLT approach. Among its three domains, teachers’ efficacy for instructional strategies weighed more in deciding their pedagogical choice, CLT approach, than efficacy for student engagement and classroom management. This finding expressed when teachers possessed stronger confidence in their instructional strategies they would be more able to integrate the spirits of CLT approach into their teaching.
Refer Appendix E for the detailed description of statistical correlation coefficients.
Table 4-3-1 summarized the results of Pearson correlation analyses among research variables.
73
Table 4-3-1 Summary of Pearson Correlation Analyses of Research Variables
Self-efficacy EFSE EFCM EFIS CLT
Language Proficiency
.485** .456** .290** .575** .438**
Speaking .477** .460** .294** .556** .401**
Listening .407** .406** .221** .490** .407**
Reading .452** .403** .288** .554** .404**
Writing .459** .451** .275** .534** .410**
CLT .469** .450** .344** .487** ---
Note. ** Correlation is significant at the 0.01 level (2-tailed)
TSE refers to “teachers’ self-efficacy.”
EFSE refers to “efficacy for student engagement”.
EFCM refers to “efficacy for classroom management”.
EFIS refers to “efficacy for instructional strategies”.
CLT refers to “communicative language teaching”.
74
75
CHAPTER 5
DISCUSSION
The previous chapter addresses the answers to the three research questions of the present study statistically. In this chapter, further ratiocination was done to work on the possible reasons to the results of statistical analyses presented in chapter four.
The phenomena exposed were given possible explanations in the sequence of research questions. First, a portrayal of the statistical descriptive analyses of teachers’ self-efficacy, teachers’ language proficiency, and pedagogic strategies are given. Second, the phenomena of the causal relationship between antecedents and teachers’ perceived self-efficacy are delineated. Finally, a further depiction of the relationships among teachers’ self-efficacy, language proficiency, and pedagogic strategies is portrayed.
Levels of Non-Native EFL Teachers’ Self-Efficacy in Taoyuan
Based on Bandura’s (1997) theory, teachers’ self-efficacy refers to teachers’beliefs in their capabilities of achieving specific tasks concerning effective teaching and learning. The 24-item Teachers’ Sense of Efficacy in Teaching English Scale was modified from Tschannen-Moran and Woolfolk Hoy’s (2001) survey to reflect specific tasks related to teaching elementary school English in Taoyuan, which involved three dimensions: Student Engagement, Classroom management, and Instructional Strategies.
Overall, the self-efficacy in teaching English of NNETs in Taoyuan was at a moderately high level since its mean score is above the average. That is to say, the
76
teachers felt confident in instructing an English class and believed they could have some influences in classroom management, student engagement and instructional strategies. Comparatively, the teachers felt more certain about classroom
management and instructional strategies. The results imply teachers were good at establishing a system of rules well enough to cope with students’ disruptive
behaviors and to keep activities running effectively. In addition, they had confidence in orchestrating differential approaches to meet students’ various needs and in coming up with assorted assessments to gauge students’ performance. Meanwhile, the teachers were less satisfied with student engagement, which means they were less competent to stir students’ learning motivation, to get through the students who were failing and to cultivate students’ critical thinking and creativities.
The results of the current study are inconsistent with the ones of previous studies that showed the EFL teachers who either taught at middle school or at college expressed stronger competence in instructional strategies than the other domains (Chacón, 2005; Chang, 2010; Eslami & Fatahi, 2008). Other research (Lee, 2009; Yavuz, 2007) done in the EFL primary settings of Korea and Turkey
corresponded with the current research showing that the EFL teachers presented higher efficacy in classroom management than the other domains. These findings, more or less, indicated that teachers’ self-efficacy changed due to the various
climates among different teaching levels, which was in accord with previous studies (Tschannen-Moran & Hoy, 2007). Students’ behaviors and mentality change with age. Education focus shifts with various education institutes. The discrepancies among the studies only ascertain the context-bound feature of teachers’ self-efficacy (Bandura, 1997).
In accord with the previous research (Chacón, 2005; Chang, 2010; Eslami &
77
Fatahi, 2008; Tschannen-Moran & Hoy, 2007), the efficacy in student engagement appeared weaker than the other domains. If interpreted into the Taiwanese milieu, these findings somehow reflected the phenomena of many current language classrooms in primary schools. Many EFL teachers stated in the previous research (Liu, 2008; Chang, 2010) that the gaps of different levels’ students in a class were getting bigger and bigger due to students’ varied family backgrounds, which made it difficult for teachers to engage students and caused students’ inertia in learning. As revealed in Bayraktar’s (2011) research, teacher’s self-efficacy and student’s motivation exhibit a reciprocal relationship of strength. Many studies (Adedoyin, 2010; Chang, 2010, to name just a few) also pointed out that teachers’ high
self-efficacy would shoot up students’ self-efficacy and motivation, which results in the improvement of students’ learning and achievement. Therefore, some proper arrangement should be done urgently to deal with the plight that teachers faced, such as classifying students into varied classes by their levels. Otherwise, the teachers would lose most of the students’ interest in class, which would backfire with the consequence of teachers’ self-efficacy decrease. More pedagogical seminars and teaching trainings should also be designed and held to help teachers to cope with such predicaments in language class.
The differing degrees of confidence level among teachers reveal the domain- and task-specific features of the teachers’ self-efficacy, proposed by Bandura (1997) that one’s self-efficacy beliefs are specific to the domain, the level of difficulty within the domain, and the context. The result of current study agreed with the research (Chacón, 2002, 2005; Chang, 2010; Eslami & Fatahi, 2008; Lee, 2009) that found participants’ levels of efficacy varied across tasks. According to the
research(Liu, 2008), many other complex factors were interfering teachers’ beliefs
78
about their competence in carrying out English teaching tasks, such as student characteristics, difficult teaching moments, and social expectation. Therefore, it is dangerous and misleading to classify teachers in the extremes with a high or low sense of efficacy because their levels of self-efficacy might vary across different tasks and under various contexts.