Rita Divéki: Developing Global Competence in the Hungarian EFL Classroom
›
3. Research Design and Methods
›
3.4. Studies 3 and 4 – The Questionnaire Studies
›
3.4.2. Data Collection and Analysis
›
3.4.2.2. Study 3 – The Main Study in the Secondary School Context

3.4.2.2. Study 3 – The Main Study in the Secondary School Context

Jegyzet elhelyezéséhez, kérjük, lépj be.!

The questionnaire was administered online in June-July 2021, through a call for participation via email, the distribution of the link to the questionnaire by colleagues and posting the link to the questionnaire on a social media platform. Convenience and snowball sampling methods were used to identify the participants. As the questionnaire was powered by Google Forms, the responses could be easily downloaded as a spreadsheet. After cleaning and coding the data, the spreadsheet was loaded into SPSS 22 and the initial analysis began.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Table 3.8 The Reliability Analysis of the Scales in the Main Study in the Secondary School Context
Scales
No of items
Cronbach’s alpha
1. Teachers’ likeliness to include certain global issues in their lessons
22
.93
2. Teachers’ likeliness to include certain local issues in their lessons
15
.93
3. Teachers’ likeliness to include certain intercultural issues in their lessons
17
.93
4. The frequency of the inclusion of global, local, and intercultural topics in the EFL class
5
.82
5. Teachers’ preferences in the integration of global content
6
.84
6. The importance of the inclusion of global, local, and intercultural issues
4
.81
7. Aspects influencing the inclusion of global content
30
.75
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

As explained above in 3.4.2.1, the validity of the instrument was already established, nevertheless, some steps had to be taken to establish the reliability of the questionnaire as well. Similar to the pilot, this started with the calculation of the Cronbach’s alpha reliability coefficient of the scales. The results of this test can be seen in Table 3.8.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

After the reliability analysis, an unrotated PCA was run on the scales which did not deal with teachers’ likeliness to include certain issues in their classes. In the case of scales 4 and 6, the items constituted one dimension. In the case of the teachers’ preferences in the integration of global content scale (scale 5), differently from the pilot, the answers constituted two distinct (but close) dimensions, as teachers thought differently about the discussion of local issues in EFL classes than about the inclusion of global and intercultural issues. Table 3.9 shows this difference in perception.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Table 3.9 Teachers’ Preferences of Dealing with Global, Local or Intercultural Content in their Classes – The Results of the Principal Component Analysis
Item
Code
Component
Dim.
1
2
1. How much do you like dealing with global issues if you bring the topic into class?
PREFERENCE_01GLOBAL_TINITIATES
0.79
–0.22
1.
2. How much do you like dealing with global issues if the students bring the topic into class?
PREFERENCE_02GLOBAL_SINITIATES
0.87
–0.10
3. How much do you like dealing with local issues if you bring the topic into class?
PREFERENCE_03LOCAL_TINITIATES
0.63
0.66
2.
4. How much do you like dealing with local issues if the students bring the topic into class?
PREFERENCE_04LOCAL_SINITIATES
0.65
0.68
5. How much do you like dealing with intercultural issues if you bring the topic into class?
PREFERENCE_05INTERCULTURAL_TINITIATES
0.75
–0.44
1.
6. How much do you like dealing with intercultural issues if the students bring the topic into class?
PREFERENCE_06INTERCULTURAL_SINITIATES
0.82
–0.33
Note. Dim = Dimension.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In the case of last scale, the aspects influencing the inclusion of global content in EFL classes, separate tests were run to first test the reliability of each construct, and then to see whether the construct really measured the same dimension. Out of the seven constructs, four worked properly, and one construct (MATERIALS) had to be divided into two (MATERIALS and COURSEBOOK). After these two constructs were separated, dimension reduction showed that they effectively constitute separate components, and their reliability coefficient value also reached the .6 threshold. In the case of two constructs (TIME and TEACHER COMPETENCE), one item had to be deleted from each so that the others would constitute one component.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Table 3.10 The Finalisation of the Constructs for the Main Study in the Secondary School Context
Original construct
Changes made
New construct
Name
No
α
Comp
Name
No
α
Comp
TIME
4
.65
2
1 item deleted (FACTOR02_TIME1)
TIME
3
.66
1
GROUP
4
.68
1
–
GROUP
4
.68
1
TEACHER COMPETENCE
4
.58
1
1 item deleted (FACTOR10_TEACHERCOMP2)
TEACHER COMPETENCE
3
.70
1
PD
4
.73
1
–
PD
4
.73
1
MATERIALS
6
.14
2
Divided into two components
COURSEBOOK
3
.84
1
MATERIALS
3
.67
1
STUDENTS
4
.29
2
Construct deleted
TEACHER ATTITUDE
4
.18
2
Construct deleted
Note. α – Cronbach’s Alpha, no. – number of items, comp – components, PD – professional development.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

After the necessary items were deleted, the reliability coefficients were recalculated in each case and only those constructs remained which reached the .6 threshold, and which only had one component. It must be noted that according to Dörnyei and Csizér (2012), when it comes to analysing the results, “we should aim at coefficients in excess of .7” (p. 84), which turned out to be the case with three constructs (TEACHER COMPETENCE, PD, COURSEBOOK). By adding more items to these constructs, these numbers could be made even better. Table 3.10 presents in detail what steps had to be taken to ensure that each construct would work properly.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

The distribution of the scores in each construct was also calculated, using descriptive statistical procedures. As it was suggested by Larson-Hall (2016), when it comes to evaluating the normality of our data, we can use the skewness number and if it is under .1, there is “no cause for concern” (p. 78). As it can be seen from Table 3.11, the skewness and the kurtosis of each scale was under 1, meaning that “normality was not violated” (p. 78). Given that the data had normal distribution, parametric tests could be performed on the dataset.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Table 3.11 Normality Test Results for the Constructs
Scale
Mean
SD
Skewness
Kurtosis
Statistic
Std. Error
Statistic
Std. Error
PD
2.42
.93
.83
.18
.06
.37
COURSEBOOK
2.62
1.11
.31
.18
–.67
.37
GROUP
3.13
.82
–.26
.18
.04
.37
TIME
3.39
.91
–.56
.18
–.12
.37
MATERIALS
3.57
.89
–.29
.18
–.32
.37
TEACHER COMPETENCE
3.64
.74
–.59
.18
.91
.37
Note. Std. Error = Standard Error; SD = Standard Deviation.
 

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Data were then submitted to various statistical procedures. To find out how likely teachers are to bring different global, local, or intercultural issues into their classrooms, descriptive statistical tests were run. To be able to confidently say which topics teachers would bring into their classes and which ones they would not, inferential statistical tests, in this case, independent samples t-tests were run. To see whether there are any differences in teachers’ likeliness to bring in such issues into their classes based on age, geographical location, or the type of school they work in, one-way ANOVA (analysis of variance) tests were run.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

To measure the frequency of inclusion of global content, teachers’ preferences of dealing with such issues and the importance they attribute to dealing with these topics, descriptive statistical tests were performed. To answer research question 2.2, to be able to compare the answers of the different groups of participants based on their demographic characteristics, ANOVA tests were run. To test what aspects influence teachers’ decision to incorporate global content in their classes, correlation tests were run. A p-value of .5 was used to determine the significance of the results. The answers to the open-ended (clarification and short-answer) questions were coded separately and analysed using content analysis.
 
chevron_left
chevron_right
Tartalomjegyzék navigate_next
Keresés a kiadványban navigate_next
A kereséshez, kérjük, lépj be!
Könyvjelzőim navigate_next
A könyvjelzők használatához
be kell jelentkezned.
Jegyzeteim navigate_next
Jegyzetek létrehozásához
be kell jelentkezned.
    Kiemeléseim navigate_next
    Mutasd a szövegben:
    Szűrés:
    Kiemelések létrehozásához
    MeRSZ+ előfizetés szükséges.
      Útmutató elindítása
      delete
      Kivonat
      fullscreenclose
      printsave