Introduction

Nowadays, linguo-intercultural sensitivity is a very important topic. It includes many scientific disciplines and branches: sociology, linguistics, psychology, anthropology, culturology, etc. We cannot imagine contemporary communication without English language. It is one of the main international languages; therefore, it is learned as a foreign/second language in countries where it is not lingua materna. Hence, there is a connection between English language competencies, skills, and its usage with intercultural sensitivity. We called this kind of relationship linguo-intercultural sensitivity (or flexibility).According to Kovalainen and Keisala (2002), developing this kind of sensitivity is necessary when contemporary trends are present, such as: increased transnational mobility, migration of skilled professionals, studying abroad, international collaboration of scholars, etc. Riemer (2007) wrote that awareness of cultural components (i.e. intercultural sensitivity) leads to maximizing understanding of messages between sender and receiver who belong to different cultures. Combined with good communication skills, it can result in more positive studying, academic, and working experiences.

Some researchers (e.g. Blue, Kapoor, & Comadena, 1996) use cultural values in defining and measuring linguo-intercultural sensitivity. However, this is only one aspect of this concept and we also have to consider its communication components. Other researchers and theorists (e.g. Piller, 2007) underline that languages have structural differences and those interlocutors from various cultures can have specific cognitive schemes, which are related to expressing and understanding opinions, attitudes, and verbal messages.

In his study, Lázár (2003) found that intercultural competence is implicitly taught in the following courses: Sociolinguistics, TEFL, English literature, Psychology, Folklore Studies, British and American Studies, Phonetics, Theory of Education, and History. As we can notice, intercultural skills and competencies (namely sensitivity) are not taught in the courses which are totally dedicated to this subject. However, every component of intercultural sensitivity (which is linked to skills of English language usage) has to be explained to students and students need to develop it with a help of their teacher/trainer.

Dorn and Cavalieri-Koch (2005) listed six intercultural skills: behavioral flexibility (adaptation to communicational and interactional context), communicative awareness (this is the ability to understand communicative conventions of interlocutor and to respond and behave in accordance with them), respect for otherness (a person has to respect the cultural background of his/her interlocutor), empathy (a person should be able to understand emotions, attitudes, and cultural values of his/her interlocutor), tolerance of ambiguity (dealing with ambiguous or unclear situations constructively in order to achieve positive communicational outcomes), and knowledge discovery (this is cultural knowledge which is used in intercultural interactions).

The most famous theoretical approach to linguo-intercultural sensitivity is Developmental Model of Intercultural Sensitivity, proposed by Milton J. Bennett (1986). This theorist considers intercultural sensitivity as a continuum from ethnocentrism (avoiding cultural differences) to ethnorelativism (seeking cultural differences) (Bennett, 2004). Based on their places or positions on this continuum, we have six orientations: denial (lack of interest in cultural difference), defense (discrimination of different cultures), minimization (minimizing differences/finding similarities between our own culture and other cultures through the mechanisms called physical and transcendental universalism), acceptance (experiencing our culture as one of the numerous worldviews or cultures), adaptation (if we have this orientation, we can behave and respond, regarding to the rules and customs of other cultures), and integration (constructing and accepting intercultural identity). People who have one of the first three orientations (denial, defense, and/or minimization) are prone to the ethnocentrism. On the other hand, people who accept one of the last three orientations (acceptance, adaptation, and/or integration) are thinking and behaving in accordance to the ethnorelativism.

Similarly, Hammer and Bennett (2001) constructed and validated Intercultural Development Inventory (IDI), which is designed to measure the following orientations toward cultural differences (i.e. intercultural sensitivity dimensions): denial/defense, reversal (some sort of defense orientation, defined as a sense of superiority of one’s own cultural worldview compared to other cultural backgrounds), minimization, and acceptance/adaptation to cultural differences (Hammer, Bennett, & Wiseman, 2003). As we can see, this inventory is based on the Developmental Model of Intercultural Sensitivity (DMIS), discussed in the previous paragraph.

Jantawej and Inada (2011) investigated intercultural sensitivity of foreign teachers in Thailand. Their findings were the following ones: the teachers from their sample scored high on the Interaction attentiveness, Intercultural engagement, and Interaction confidence subscale. On the other hand, their scores were low on the Respect for cultural differences and Interaction enjoyment subscale. As can be seen, there are cultural differences in perceptions of foreigners and interactions with them.

Cubukcu (2013) examined the intercultural sensitivity levels of pre-service English teachers in Turkey. His findings indicated teachers’ high levels of tolerance for cultural differences and positive teachers’ perceptions of culture teaching (e.g. shared values/beliefs, a sense of empathy, various cultural expressions, daily life and routines of different cultures, etc.).

Ruokonen and Kairavuori (2012) researched the intercultural sensitivity of the ninth graders in Finland. They got the following gender differences: girls claimed that they were more eager to accept people who come from different cultures, and allow them to adapt and integrate themselves in the participants' culture (compared to boys, who did not think so); on the other hand, boys reported high levels of denial, defence, and minimization with regard to people with different cultural backgrounds.

How is intercultural sensitivity related to foreign language learning and teaching? Laopongharn and Sercombe (2009) claimed that increased intercultural communicative competence (ICC), which can be considered as one of the components of intercultural sensitivity, produces more proficient usage of English language in intercultural interactions. These authors provided lots of arguments on which their conclusion is based. They discussed the relationship between culture, foreign and native language, communication, social interaction, and intercultural environment.

Likewise, Alptekin (2002) highlighted that English is a world language which is used in local and international contexts, by both native-nonnative and nonnative-discourse participants. Intercultural communicative competence (such as, in our case, verbal ability and verbal expressiveness, with intercultural awareness), according to this author, has to be taught in ESL/EFL courses i.e. this is a very important topic in English language teaching (ELT).

Knott, Mak and Neill (2013) investigated the benefits of teaching intercultural competences in introductory psychology. This teaching intervention had the following outcomes: increased English knowledge of students, their respect of people who come from different cultures, and high levels of participants' competence while interacting and communicationg with students who have various cultural backgrounds.

Aydogan and Akbarov (2014) researched the linkage between six compontents of intercultural sensitivity, verbal abilities and competencies, and linguo-intercultural flexibility. Their results indicated that English language communication skills/ competencies, emotional attitudes, and linguo-intercultural flexibility are in statistically significant correlations with almost all facets of intercultural sensitivity (interaction engagement and enjoyment, interaction attentiveness and confidence as well as respect for cultural differences).

Based on the previously mentioned considerations of intercultural sensitivity, the following set of research questions have been defined:

Are there correlations between English competence, emotional attitudes towards English, verbal expressiveness, verbal abilities and linguo-intercultural sensitivity (flexibility)?

Do English competence, emotional aspect of attitudes towards English, verbal expressiveness, and verbal abilities have common characteristics with linguo-intercultural sensitivity/flexibility?

Are English competence, emotional attitudes towards English, verbal expressiveness, and verbal abilities good predictors of linguo-intercultural sensitivity?

What does the shape of distribution for lingo-intercultural sensitivity look like?

Likewise, considering the research questions listed above, we proposed these hypotheses (which are defined using language of statistics, in the form of statements):

Correlations between the five mentioned variables related to attitudes towards English and its usage are statistically significant.

English competence, emotional aspect of attitudes towards English, verbal expressiveness, and verbal abilities, taken together, explain statistically significant part (percent) of variance of Linguo-intercultural sensitivity/flexibility.

Verbal expressiveness and verbal abilities are the best predictors of Linguo-intercultural sensitivity.

The difference between the distribution of Linguo-intercultural sensitivity (flexibility) and normal curve is not statistically significant.

Methodology

Participants/subjects

We conducted this research with 194 students who were at Sarajevo international high schools and universities. These students were selected randomly and participated voluntarily in this research. Mean age of our subjects was M = 18.05, with standard deviation of SD = 2.12. Their age ranged from 14 to 27 years old. The mean and standard deviation of age by gender are displayed in Table 1.

Table 1: Participant's Age and Gender

As we can see (Table 1), the arithmetic mean of males' age is a bit higher than that of females (M = 18.67 vs. M = 17.75). However, the range of females' age is broader, than that of males (from 14 to 27 vs. from 16 to 22 years old).

In Figure 1 below, we displayed the distribution of our sample by gender.

Figure 1. Gender Distribution Among Participants

As can be noticed (Figure 1), our sample mostly consisted of women (n = 131, or 67.5% of the total sample). There are also 45 males (23.2%) and 18 participants (9.3%) did not provide information about their gender.

Measures

Our main instrument for this research was Attitudes towards English and its Usage Scale (ATEUS, Aydogan & Akbarov, 2014). It has 30 statements, in the form of five-point Likert scale. ATEUS consists of five subscales. These subscales with their Cronbach's alpha coefficients (α) of internal consistency (reliability) are:

English competence (statements: 1 to 6; α = .878),

Emotional attitudes towards English (statements: 7 to 12; α = .825),

Verbal expressiveness (items: 13 to 18; α = .920),

Verbal ability (items: 19 to 24; α = .857), and

Linguo-intercultural sensitivity/flexibility (items: 25 to 30; α = .771).

The reliability of the total scale is α = .932. We can conclude that ATEUS and its subscales have appropriate reliabilities.

Procedure

We conducted this research on December 2013 among students of international schools in Sarajevo, Bosnia and Herzegovina. It took them approximately 10 minutes to fill out the questionnaire. During the research, we took care of ethical recommendations and informed our participants that the survey is voluntary and anonymous.

After the data were collected, they were entered into SPSS 16.0 for Win, to conduct appropriate statistical analysis.

Findings

First, we displayed descriptive statistical values for our variables, which will be used in the coming analyses (Table 2). Note that the meaning of labels is as following: Min – minimal result (score), Max – maximal result, M – arithmetic mean, and SD – standard deviation.

From Table 2, we can read that Emotional attitudes subscale has the highest arithmetic mean (M = 25.25), whereas Verbal expressiveness has the lowest arithmetic mean (M = 23.43). The most variable results are those of Verbal expressiveness (SD = 4.72) and the less variable results are on Linguo-intercultural sensitivity subscale (SD = 3.84).

To test our first hypothesis, we calculated Pearson's coefficients of linear correlation between the results on five subscales of ATEUS.The results are given in Table 3.

From Table 3, we can read that English competence is in positive and statistically significant correlations with all other variables: Emotional attitudes (r = .410, p < .001), Verbal expressiveness (r = .737, p < .001), Verbal abilities (r = .733, p < .001), and Linguo-intercultural sensitivity (r = .272, p < .001). English competence shares the greatest amount of its variance with Verbal expressiveness (coefficient of determination is: .7372 = .5432, which is 54.32% of common variance).

Emotional attitudes toward English are also in a statistically significant correlation with Verbal expressiveness (r = .489, p < .001) and Verbal abilities (r = .430, p < .001). The results on this subscale are not in a statistically significant correlation with Linguo-intercultural sensitivity (r = .130, p > .05). Emotional attitudes share the greatest part of its variance with Verbal expressivenes (coefficient of determination .4892 = .2391, i.e. 23.91% of shared variance).

Verbal expressiveness is also in a statistically significant correlations with Verbal abilities (r = .670, p < .001) and Linguo-intercultural sensitivity/flexibility (r = .339, p < .001). It shares the greatest amount of variance with English competence (as it was already said).

Verbal abilities are also in a statistically significant correlation with Linguo-intercultural sensitivity (r = .362, p < .001). This subscale shares the greatest part of its variance with English competence subscale (.7332 = .5373, that is 53.73% of common variance).

Linguo-intercultural sensitivity/flexibility has the biggest overlapping part of variance with verbal abilities (.3622 = .1310, i.e. 13.10% of shared variance). Looking at the results discussed above, we can say that the most part of the first hypothesis is proved.

To test the second hypothesis, we conducted multiple regression analysis (MRA), using the standard model. The data are displayed in Table 4 and Table 5 below.

Note that in Table 4 are: SS – sum of squares; df – egrees of freedom; MS – mean square (variance); F – F ratio (test); p – the significance of F – ratio; R – coefficient of multiple correlation; R2 – coefficient of multiple determination (when multiplied with 100, it shows the part of explained variance of criterion, by

predictors); and Adj. R2 – adjusted (corrected) coefficient of multiple determination.

Considering the values in Table 4, we can conclude that our model with four predictors (subscales of ATEUS) explain the statistically significant part of variance of Lingo-intercultural sensitivity/flexibility (F (4; 189) = 8.795, p < .001). Taken together, our predictors are moderately correlated with criterion (R = .396) and explain 15.7% of its variance (R2 = .157). If we look at the adjusted coefficient of multiple determination, the explained variance is a bit smaller i.e. 13.9%. Hence, we proved our second hypothesis.

To test the third hypothesis, we analysed beta-coefficients for every predictor. We expected that students are more lingo-interculturally sensitive/flexible if they have good verbal expressiveness and abilities.

The values displayed in Table 5 are labeled as: B – non-standardized regression coefficient; SEB – standard error of the coefficient B; β – standardized regression coefficient; t – value of t-statistic; and p – its significance.

As we can see (Table 5), English competence is not a statistically significant predictor of Linguo-intercultural senisitivity/flexibility (β = -.104, t = -0.925, p > .05), nor are Emotional attitudes (β = -.082, t = -1.057, p > .05).

However, statistically significant predictors of Linguo-intercultural sensitivity are Verbal expressiveness (β = .252, t = 2.363, p < .05) and Verbal abilities (β = .305, t = 2.949, p < .01), as we expected. Therefore, we also proved our third hypothesis.

In order to test the normality of distribution for Linguo-intercultural sensitivity/flexibility, we conducted Kolmogorov-Smirnov test. Its result showed that there is not a statistically significant difference between this distribution and the normal curve (K-S Z = 1.227, p > .05).

Figure 2.The Distribution of Scores on Linguo-intercultural Sensitivity Subscale

In Figure 2, we can notice that there are several deviations from the normal curve, which could be considered with the exact significance of Kolmogorov-Smirnov Z statistic (it was p = .099, which is the value close to the marginal value od .05). But, the result is not significant, hence, we can accept our final hypothesis and say that the distribution of results on this subscale do not differ significantly from the normal (bell-shaped) curve.

Discussion and Results

First, our results suggest that our participants scored above average on all subscales of ATEUS (Table 2). This is probably the case beacuse we applied self-report measures. The other explanation is based on the high levels of self-esteem among students. Participants consistently estimate themselves in positive manner (they think that they are self-worthy). Hence, self-esteem influenced their assessments.

Secondly, only emotional attitudes did not correlate with linguo-intercultural sensitivity/ flexibility (Table 3). We can explain this finding by the nature of emotional component of attitudes. This is a component which is very wide and includes lots of different reactions to various social situations and inner emotional states of people.

As it was expected, self-reported verbal abilities and expresiveness moderately highly correlated with English competence. In addition, verbal expressiveness and verbal abilities highly correlated with each other.

We also got low to medium correlations between English competence, verbal abitity, and verbal expressiveness on one side and linguo-intercultural sensitivity on the other side (Table 3). That is, participants who believe that their communication abilities, skills, and competencies are of high value, also believe that they are more linguo-interculturally sensitive or flexible.

Verbal expressiveness and abilities were only two statistically significant predictors of linguo-intercultural sensitivity (Table 5), which indicates that students' general beliefs of their verbal (communication) capacity and skills are crucial for starting and maintaining intercultural interactions.

Our predictors, as said before, explained 15.7% of variance of linguo-intercultural sensitivity, which means that there is 84.3% of unexplained variance. The rest of criterion's variance can be explained by other relevant variables such as: social self-esteem, extraversion, collective self-esteem, the difference between interlocutors' specific cultural backgrounds, etc.

Finally, in our sample, linguo-intercultural sensitivity was distributed (almost) according to normal curve, which implies that most of the people have average levels of linguo-intercultural flexibility as well as smaller number of people score below or above average. However, by inspecting Figure 2, we can notice that, in fact, more participants scored above average, than below average. This indicates that our subjects are interculturally aware and try to make transnational connections with other people who have different cultural worldviews.

Our study, like some other studies, has its advantages and disadvantages. The main advantage is the research of a very important topic among international students and including analysis such as other relevant variables (communicational/verbal capacity of students). The main disadvantage is the unbalanced number of males and females in the sample, which decreased the generalization potential of our research.

Recommendations for the future research in this field

The measures used in this research were a sort of self-report techniques. However, other researchers should apply more objective measures of students' English competences, skills and knowledge in order to test students' verbal abilities. For example, they can engage three independent estimators (English teachers) who will test students' verbal skills, competences, and expressiveness. Moreover, the main researcher will calculate the average value of estimators' grades given to students. After that, these grades will be correlated with linguo-intercultural sensitivity/flexibility.

The second recommendation is the following one: other researchers should test gender differences in linguo-intercultural sensitivity (by using t-test for independent samples) or they can investigate the levels of linguo-intercultural sensitivity by comparing participants who came from different cultures (e.g. Turkish vs. American students, Bosnian vs. French students, etc.).

The third recommendation for the future research in this scientific field is the comparison between the high school and university students, with regard to their levels of verbal expressiveness, verbal abilities, English competences, attitudes towards English language, and linguo-intercultural sensitivity/flexibility.

Regarding to the proposed hypotheses, we can conclude the following:

Almost all subscales are mutually intercorrelated and these correlation coefficients are statistically significant (except in the case of Emotional attitudes and Linguo-intercultural sensitivity/flexibility, where the correlation was insignificant and equal to r = .130). English competence was correlated the most with Verbal expressiveness (r = .737, p < .001), Emotional attitudes correlated the most with Verbal expressiveness (r = .489, p < .001), Verbal abilities had the strongest relationship with English competence (r = .737, p < .001), and Linguo-intercultural sensitivity had the strongest relationship with Verbal abilities (r = .362, p < .001).

English competence, Emotional attitudes towards English, Verbal expressiveness, and Verbal abilities explain (taken together) statistically significant part of the variance of Linguo-intercultural sensitivity/flexibility (more precisely, they explained 15.7% of the criterion's variance). Their coefficient of determination was R = .396.

Verbal expressiveness and Verbal abilities are statistically significant predictors of the mentioned criterion, as we stated in our third hypothesis. Verbal abilities were better predictor (β = .305, t = 2.949, p < .01), compared to verbal expressiveness (β = .252, t = 2.363, p < .05). Hence, participants who think they have very good or excellent verbal abilities have high levels of linguo-intercultural sensitivity/ flexibility as well.

The distribution of the results on Linguo-intercultural sensitivity/flexibility is not significantly different from the normal curve (Kolmogorov-Smirnov Z = 1.227, p > .05). Therefore, we can accept the notion that there are lots of participants who had average scores on this subscale. Additionally, there was smaller number of participants whose scores were very low or very high (this finding is in accordance with the model of normal curve).