This chapter elaborates on the methodology of the research. It includes the explanation of the framework, research hypothesis, research procedure, the methods applied, target population and sample. This study made use of quantitative research approach to examine the relationships between depth of cultural exposure, cultural intelligence (CQ) and foreign language communication anxiety (FLCA).
Research Framework
The research framework was developed in the accordance with the review of the literature. Based on the literature review, it has been shown that there is a strong relationship between depth of cultural exposure and cultural intelligence which are the independent and dependent variable respectively in this study. The framework also utilized foreign language communication anxiety as its moderator. The other crucial information to the framework is education and number of countries visited are used as control variables.
Figure 3.1. Research framework.
26
Research Hypotheses
Based on the proposed questions and literature review, the hypotheses below were examined:
Hypothesis 1: Depth of cultural exposure is positively related to cultural intelligence.
Hypothesis 2: Foreign language anxiety significantly and negatively moderates the relationship between depth of cultural exposure and cultural intelligence.
Research Procedure
The procedure of this research is shown below. Please see the Figure 3.2.
Figure 3.2. Research procedure.
27
Target Population and Sample
The target population of this research are Taiwanese students who have had an experience of studying abroad regardless of the program and target country, except for Mainland China where the main official language is Chinese. The type (exchange, degree-seeking, visiting students, etc.) was not considered. The students who graduated from any type of school were applicable too, however, the duration of staying abroad for pursuing education could not be less than one semester which is around five months, the length of one academic term may differ depending on a country or level of study, for example, high school, university.
The data obtained from the Ministry of Education presented there were 63,270 Taiwanese who went abroad for receiving education in 2016 and 67,688 in 2017.
Questionnaire Design
The original language of the scales used in the questionnaire was English, however, due to the fact that the target population in this study were Taiwanese students whose mother tongue is Mandarin Chinese, the questionnaire was translated into Chinese as the main language. The questionnaire was designed in two forms; paper-based copies and online which was created by Google forms. The questionnaire consisted of four sections: depth of cultural exposure, cultural intelligence, foreign language communication anxiety and demographics.
Back Translation
The original scales were in English. Since the target population was Taiwanese people, the survey was created in English and then translated into Chinese to avoid misunderstandings.
In order to ensure the quality and accurateness of the translation in Chinese, the version in Mandarin was checked and corrected by classmates and a professor fluent in both languages.
28
Pilot test
This research was a quantitative one. Before the data collection the pilot test had been conducted. From the survey online 40 valid samples were obtained. The period for data collection was about a month. The Cronbach’s alpha (α) coefficient values of the variables are as follows depth of cultural exposure .63, cultural intelligence .93 and foreign language communication anxiety .96. Table 3.1 shows Cronbach’s α coefficient value was measured to check the internal consistency of a measurement. The Cronbach’s α coefficient of cultural intelligence and foreign language communication turned to be very high presenting a great internal consistency. When it comes to depth of cultural exposure, its α value did not exceed .70, however, it might have been caused by the fact that the sample used to the pilot test was not big enough.
Table 3.1.
Cronbach’s α Coefficient of the Measurement in Pilot Test (N = 40)
Variables Cronbach’s α
1. Depth of cultural exposure .63
2. Cultural intelligence .93
3. Foreign language communication anxiety
.96
Data collection
The data was obtained by the distribution of a self-reported questionnaire, it was compiled by using Google forms and sent to the Facebook groups which gather Taiwanese people studying abroad. Also, the link to the survey was distributed to friends of the researcher and the researcher’s friends who had the experience of studying abroad. The researcher got 213 valid responses.
29
Measurement Depth of Cultural Exposure
The literature review reveals that studies on cultural exposure in the past do not determine this variable in great details. International exposure measurement differs a lot from one research to another and it is frequently simplified, that is also why in order to collect information on the complex character of depth of cultural exposure, this research utilized the items developed by Crowne (2013). The depth of CE was measured by indicating how frequent the participants interacted with the local people while studying abroad. I was calculated on the 5-point Likert scale starting from 1 (never) and ending at 5 (always).
Table 3.2.
Items of Depth of Cultural Exposure
No Items
1 How often did you visit local shops?
2 How often did you visit local food markets?
3 How often did you visit local restaurants?
4 How often did you meet local residents?
Note. Adopted from “Cultural exposure, emotional intelligence, and cultural intelligence: an exploratory study,” by K. A. Crowne. 2013, International Journal of Cross Cultural Management, 13(1), p. 5-22.
Cultural Intelligence (CQ)
Cultural intelligence (CQ) indicates a person’s capability to handle culturally different situations (Earley, 2002). This study made use of the cultural intelligence scale (CQS) that was compiled by Ang and the colleagues (2007). The very scale consists of twenty items that are grouped into four dimensions: meta-cognitive CQ, cognitive CQ, motivational CQ as well as behavioral CQ. There are four items measuring meta-cognitive CQ, for instance, “I adjust my
30
cultural knowledge as I interact with people from a culture that is unfamiliar to me”; cognitive CQ has six items and one of its question included is “I know the cultural values and religious beliefs of other cultures”; there are five items found in motivational CQ, “I enjoy living in cultures that are unfamiliar to me”; and behavioral CQ involves five items, for example “I use pause and silence differently to suit different cross-cultural situations”.
The questionnaire was filled in by the respondents on the basis of their honest opinions while living abroad. This part of the questionnaire was measured by means of a 5-point Likert scale starting from 1 (strongly disagree) and ending at 5 (strongly agree). The lowest score means that an individual has problems with adapting to a new cultural environment, while the highest sore shows that the very person easily adjusts to an unknown culture, understands its values and norms and at the same time is able to express proper verbal and nonverbal responses.
Table 3.3.
Items of Cultural Intelligence
Dimension No Items
Metacognitive CQ
1 I am conscious of the cultural knowledge I use when interacting with people with different cultural backgrounds.
2 I adjust my cultural knowledge as I interact with people from a culture that is unfamiliar to me.
3 I am conscious of the cultural knowledge I apply to cross-cultural interactions.
4 I check the accuracy of my cultural knowledge as I interact with people from different cultures.
(continued)
31 Table 3.3. (continued)
Dimension No Items
Cognitive CQ
5 I know the legal and economic systems of other cultures.
6 I know the rules (e.g., vocabulary, grammar) of other languages.
7 I know the cultural values and religious beliefs of other cultures.
8 I know the marriage systems of other cultures.
9 I know the arts and crafts of other cultures.
10 I know the rules for expressing nonverbal behaviors in other cultures.
Motivational CQ
11 I enjoy interacting with people from different cultures.
12 I am confident that I can socialize with locals in a culture that is unfamiliar to me.
13 I am sure I can deal with the stresses of adjusting to a culture that is new to me.
14 I enjoy living in cultures that are unfamiliar to me.
15 I am confident that I can get accustomed to the shopping conditions in a different culture.
Behavioral CQ
16 I change my verbal behavior (e.g. accent, tone) when a cross-cultural interaction requires it.
17 I use pause and silence differently to suit different cross-cultural situations.
18 I am conscious of the cultural knowledge I use when interacting with people with different cultural backgrounds.
19 I adjust my cultural knowledge as I interact with people from a culture that is unfamiliar to me.
20 I am conscious of the cultural knowledge I apply to cross-cultural interactions.
Note. Adopted from “Cultural intelligence: its measurement and effects on cultural judgement and decision making, cultural adaptation and task performance,” by S. Ang et al. 2007, Management and Organization Review, 3(3), p. 335-371.
32
Foreign Language Communication Anxiety (FLCA)
In this research in order to measure foreign language communication anxiety which is
“the feeling of tension and apprehension specifically associated with second language [or third, or fourth, etc.] contexts, including speaking, listening, and learning” (MacIntyre & Gardner, 1994, p.284) a measure developed by Guntzviller and colleagues (2016) was applied. It is suitable for various settings where a foreign language may be used to communicate. Even though, the measure is suggested as a new one, it strongly avails itself of the concepts and wording introduced in the measure of FLCA to a classroom context developed by Horwitz and others (1986), a medical setting by Guntzviller, Jensen, King and Davis (2011). The questionnaire questions were adapted by Guntzviller and others (2016) to match to settings that are not related to a specific context, but to a general one and it consists of twenty-two items.
The responses were fitted into a 7-point Likert scale beginning at 1 (strongly disagree) and ending at 7 (strongly agree). The lowest number means that an individual’s scores low on foreign language communication anxiety while a person suffers from high levels of foreign language communication anxiety if he or she ends up choosing the highest number.
Table 3.4.
Items of Foreign Language Communication Anxiety
Dimension No Item
10 I feel very self-conscious when I speak the language in front of other people.
11 I do not feel confident when I speak in the language.
(continued)
33 Table 3.4. (continued)
Fear of making mistakes (M)
3 I worry about speaking in the language, even if I am well prepared for it.
7 I get nervous when I am asked questions in the language that I have not prepared in advance.
18 I am afraid native speakers are ready to correct every mistake I make.
19 I worry about making mistakes when speaking the language.
Anxiety about understanding (U)
5 I get nervous when I do not understand every work in the language.
12 It frightens me when I do not understand what the other person is saying in the language.
13 I feel anxious if I cannot understand everything the other person is saying in the language.
14 I get embarrassed when I do not understand what a native speaker is saying in the language.
Feeling of incompetence (I)
6 I fear that people will laugh at me when I speak the language.
15 I keep thinking that other people are better at languages than I am.
16 I always feel that other people who also learned the language speak it better than I do.
17 It embarrasses me to voluntarily speak in the language.
Distinctions
21 Even though I do not usually get anxious when communicating with others, I do if I have to speak in the non-native language.
22 Speaking in the language makes me unusually anxious.
Note. Adapted from “Foreign language communication anxiety outside of a classroom: scale validation and curvilinear relationship with foreign language use,” by L. M. Guntzviller, R. N.
Yale & D. J. Jensen. 2016, Journal of Cross-Cultural Psychology, 47(4), p. 605-625.
34
Control Variables Education and number of places visited abroad
According to Alon and others (2018) ones of the most important factors contributing to rising of CQ levels are number of countries a person has spent in for more than a half of a year (McNulty & Tharenou, 2004) as well as their education (Crowne, 2008) .
Validity and Reliability
The validity and reliability were tested by internal consistency and stability. The Cronbach’s alpha was made use for to examine the internal consistency. Construct validity was manipulated by confirmatory factor analysis (CFA). Average variances extracted (AVE) was evaluated as well. The construct validity was estimated by means of the average variance extracted (AVE) and composite reliability (CR). The AVE is a measure of the variance amount that is shown by a construct with regard to the amount of variance caused by measurement error.
It is said that the measurement expresses a sufficient convergent validity when the AVE score is above .50 (Fornell & Larker,1981). The CR value is comparable to the Cronbach’s α coefficient in terms of the general item reliability in the model. Its value should be above .60 (Bagozzi & Yi, 1988).
Cronbach’s Alpha Coefficient Test
Cronbach’s α coefficient is often used to show the consistency or interrelations between the items in a measurement. It is said for Cronbach’s α value to be higher than .70 to indicate high consistency. Table 3.5 presents the values of variables used in the present study. The Cronbach’s α coefficient of the depth of exposure was the lowest scoring .44. The values of Cronbach’s α coefficient of the last two variables shows a great consistency .85 and .96 for cultural intelligence and foreign language communication anxiety, respectively.
35 Table 3.5.
Cronbach’s α Coefficient of the Present Research (N = 213)
Variables Cronbach’s α
1. Cultural Exposure: Depth of Exposure
.44
2. Cultural Intelligence .85
3. Foreign Language Communication Anxiety
.96
Composite reliability (CR)
Composite Reliability (CR) measures the consistency of the items. According to Bagozzi and Yi (1988), the value of CR ought to be more than .60. All of the dimensions of cultural intelligence as well as foreign language communication anxiety met the threshold level of CR. Metacognitive CQ scored .83, cognitive CQ .82, motivational CQ .84 and behavioral CQ .78. Look at Table 3.8 for the reference. The CR values for foreign language communication anxiety were as follow: .92 physical anxiety .78 for fear of making mistakes, .91 for anxiety about understanding, .82 for feelings of incompetence and finally .85 for distinctions from general communication apprehension. For the results, check the Table 3.9. For the results of CR for depth of cultural exposure look at Table 3.7 which shows that the variable did not exceed the necessary threshold
Average Variances Extracted (AVE)
Average Variances Extracted shows the values of the variances of a construct. Fornell and Larker (1981) suggested that its value should not be less than .50. Except for fear of making mistakes (.48) all the dimension of FLCA met the AVE threshold scoring .65, .72, .53 and .59 for physical anxiety, anxiety about understanding, feelings of incompetence and distinctions
36
from general communication apprehension, respectively. Look, Table 3.9. When it comes to the dimensions of CQ the lowest values were noted for behavioral CQ which was .41 and cognitive CQ (.44), both of them did not meet the criterion, however, the rest of them did.
Metacognitive CQ scored .56 and motivational CQ had the value of .52. For the reference, look at Table 3.8. AVE for depth of cultural exposure did not exceed the value of .50. Table. 3.7
Confirmatory Factor Analysis (CFA)
Before the hypotheses testing, confirmatory factor analysis (CFA) was carried out. CFA is a statistical tool to examine the relationships among latent variables. In this research, the latent independent variable was depth of cultural exposure; the latent dependent variable was cultural intelligence; the moderating variable was foreign language communication anxiety.
The CFA results of 𝑥2/df, root-mean-square error of approximation (RMSEA), goodness-of-fit index (GFI), the comparative fit index (CFI), Tucker-Lewis non-normed index (TLI), incremental fit index (IFI) as well as NFI are thought to show the rightness of fit in the measurement models. The model fit is good when RMSEA is below or equal to .08 (Hu &
Bentler, 1999). Table 3.6. shows the criterion of goodness fit indices and its references.
Table 3.6.
The Values of Good Fit Indices
Fit indices Criteria Required References
χ²/df < 3 Carmines & Mclver, 1981; Kline, 2005; Ullman, 2006
RMSEA < .80 Hu & Bentler, 1999; Schumacker & Lomax, 2004
GFI > .80 Bollen, 1990
Fan, Thompson, & Wang, 1999 Bentler & Bonnet, 1980
Bentler & Bonnet, 1980
37
CFA results of depth of cultural exposure. From the outset, the four items measuring the x variable were examined by confirmatory factor analysis in AMOS. CR and AVE were presented in the following Table. 3.7.
Table 3.7.
Confirmatory Factor Analysis of Depth of Cultural Exposure (N=213) Variable No.
Error variance
Factor loading
T value CR AVE
Depth of Cultural
Exposure
1 0.28 0.69 2.90*
0.49 0.21
2 1.39 0.38 8.56***
3 0.93 0.37 8.55***
4 0.55 0.30 9.42***
Note. *p < .05, ***p < .001
Figure 3.3. Confirmatory Factor Analysis of Depth of Cultural Exposure.
38 Table 3.8.
Confirmatory Factor Analysis of Foreign Language Communication Anxiety (N=213) Variable Dimensions No. Error
variance
39
Figure 3.4. Confirmatory Factor Analysis of Cultural Intelligence.
40 Table 3.9.
Confirmatory Factor Analysis of Foreign Language Communication Anxiety (N=213)
Note. ***p < .001
Variable Dimensions No. Error
variance
41
Figure 3.5. Confirmatory Factor Analysis of Foreign Language Communication Anxiety.
42
43