International students’ accented English—Communication difficulties and developed strategies

Abstract The purpose of the present investigation is to explore the communication challenges caused by accented English along with strategies of international students in the Australian context. A quantitative approach was employed in order to obtain a comprehensive understanding of the linguistic experience of the students. Participants comprised 182 international students who spoke English as a second language and were undertaking undergraduate and postgraduate programs at three universities in South East Queensland. Results revealed that they experienced accent-related problems particularly in an educational setting. Their preferred strategies to overcome such difficulties were in conflict with what they regarded as the most effective strategies.


Introduction
International education in Australia has expanded rapidly in recent years, with Zhang and Mi (2010) identifying a number of issues that our international students face in their study experience. With ABOUT THE AUTHOR Eunjae Park completed her MA in 2014 and has recently completed her ME and Professional Studies in Research at Griffith University. Her research is centered on understanding L2 international students' linguistic experience related to accented English. Recent studies involve three areas: (1) communicative problems of the students caused by accented English, (2) linguistic discrimination as experienced from the students' perspectives in the context of tertiary education in the Australian context, and (3) strategies to overcome communicative issues and to achieve social harmony. The research reported in this paper may relate to larger issues within higher education and TESOL areas, for example, how to improve university services to accommodate future international students effectively, how second language programs for academic purposes can be designed given the current students' verbal difficulties caused by accentedness, and how to teach pronunciation for these students to improve speech intelligibility.

PUBLIC INTEREST STATEMENT
There few studies on the communication difficulties associated with accented English, faced by international students who speak English as a second language. This study explored the challenges caused by their accented English. By shifting attention to one specific component of spoken language, it was found that for many international students, their accented English was a definite factor which affected their communicative success as well as study experience. The impacts of two variables such as a gender difference and length of stay in terms of accented English were considered in order to deeply explore accent-related issues for the students. Corresponding strategies to overcome such issues were also examined, indicating that pronunciation was a vital area to improve communicative success.
consistently growing numbers of international students coming to Australia for their education, tertiary education has become one of Australia's largest industries (Sawir, 2005). Numerous studies conducted in Australia have reported that international students encounter a broad range of challenges (Kell & Vogl, 2007;Line, Thomas, Jones, & Robertson, 2000;Sawir, Marginson, Deumert, Nyland, & Ramia, 2008), including cultural differences, alienation, social exclusion, financial needs, homesickness, racial discrimination, and language barriers. Notably, the most common and overwhelmingly dominant perceived problem of the students was language barriers, which can give rise to academic difficulties, social exclusion, and psychological anxiety (Line et al., 2000;Zhang & Mi, 2010).
Adequate language proficiency for communication is clearly a vital preparation for international students as it plays an important role in their success in study experiences. However, Jones (1999) noted that not only are communication skills an issue, but also miscommunication. This is because incorrect or poor sound production may give rise to greater comprehension problems than speech which contains lexical or syntactical errors (Gilakjani, 2012). Many studies have found that accentedness itself can result in the students' academic performance and social networks being adversely affected (Eisenchlas & Tsurutani, 2011;Lacina, 2002;Munro, Derwing, & Sato, 2006;Nakane, 2006).
In academic settings, problems caused by different accents have led to the students encountering difficulties not only in understanding their lecturers, but also each other (Campbell & Li, 2008;Khawaja & Stallman, 2011;Sicat, 2011). Further, students who speak English as an additional language (EAL) usually have a foreign sounding accent, which Crawford (2000) argues is quite different from the norms of native English speakers, and can lead to accent stereotyping. This can prove a substantial obstacle to the students' social integration, consequently leading to their social isolation (Lacina, 2002).
Second language (L2) learner-centered research identified that L2 students may employ diverse strategies such as repeating, paraphrasing, speaking more clearly, writing down, and spelling out in order to compensate for communication breakdowns (Derwing & Rossiter, 2002;López, 2011;Osburne, 2003). These studies examined English learners at language institutions in an ESL or English as a Foreign Language (EFL) context. Oxford (1996) and Rossiter (2005) reported that the more proficient L2 learners are capable of using a broader range of strategies, and their overall strategy use is at a higher level. While their focus has been on students at language institutions, the issues faced by higher education L2 students are also of particular interest. L2 tertiary students are generally regarded as being at a higher level of language proficiency than those at language schools as they are required to reach a minimum English entry requirement. However, little is known about the communication strategies employed by L2 international university students or the effectiveness of their strategies.
Given the potential communication and related social difficulties caused by accentedness, this study focuses on communication difficulties associated with accented English that international students face, and the communicative strategies that they employ to enable them sustain interaction with others and resolve communication breakdowns.

Accented English for tertiary international students
Accent is recognized as a major cause of miscommunication (Derwing & Munro, 2009). Despite the fact that other issues such as lack of knowledge of slang and colloquial expressions were identified as one of the concerns of international students (Campbell & Li, 2008;Sherry, Thomas, & Chui, 2010), Gilakjani (2012) emphasized that unintelligible sounds may cause greater problems than speech with lexical or grammatical errors. This is because accent-associated issues could weaken communicative competence of L2 speakers (Gilakjani, Ahmadi, & Ahmadi, 2011;Hinofotis & Bailey, 1980;Lin, 2014). Therefore, researchers affirm that accent problems are the most salient issues of spoken English language (Gilakjani et al., 2011;Hinofotis & Bailey, 1980;Lin, 2014). Further, the impact of accent does not appear to be restricted to communication, but also has been linked to stereotypical views that may lead to inaccurate judgment of students' intrinsic academic ability (Eisenchlas & Tsurutani, 2011).
As many studies indicated, it appears that the challenges caused by accented English for these students are twofold: (1) the accent on the part of students causing communication difficulties and (2) the accent on the part of other people such as lecturers and other pupils, which caused comprehension difficulties for the students (Campbell & Li, 2008;Khawaja & Stallman, 2011;Malau-Aduli, 2011;Sicat, 2011). In a qualitative study, Sicat (2011) claimed that Asian international students studying at a university in the Philippines faced difficulties not only because of their inability to understand a second language, but also in understanding different accents; that is, the lecturer's accent as well as each other's accents. Furthermore, in another qualitative study conducted in Australia, Khawaja and Stallman (2011) reported that L2 tertiary students faced a major issue in the lack of understanding of Australian regional accents, which often hindered their studies. Interestingly, these students stated that they experienced difficulties in being understood because of their accent (Kell & Vogl, 2007;Khawaja & Stallman, 2011), identifying that it was embarrassing not only for them, but also for the listeners who did not understand them.
Societal attitudes toward accentedness seem to play a role in these issues, encompassing the presence of stereotypical views and undesirable beliefs within the student community. L2 speakers are often considered less educated, less trustworthy, and poor in intelligence (competence) compared with native speakers of English (Kinzler, Shutts, DeJesus, & Spelke, 2009;Lev-Ari & Keysar, 2010;Tsurutani, 2012). Eisenchlas and Tsurutani (2011) noted that L2 students' academic ability is often underestimated owing to their accented speech. They suggested that due to differential expectations of lecturers, L2 students may get fewer opportunities to participate than other students (Eisenchlas & Tsurutani, 2011). Such differences, leading to lesser engagement in learning activities, may be reflected in their final results (Munro, Derwing, & Sato, 2006;Nakane, 2006). In addition, Lacina (2002) noted that such language discrimination can delay the students' adjustment. Interactions with other students are important in forming friendships and becoming part of a supportive learning environment; however, establishing social networks can be challenging due to the presence of accent stereotyping (Lacina, 2002).
In conclusion, communication difficulties caused by accented English are by no means unimportant, having both direct and indirect effects on learning activities and the broader learning environment. In effect, such difficulties can influence the students' overall study experience and social harmony in a number of negative ways. This study examines how the role of accent exacerbates their overall communication difficulties.

Corresponding communicative strategies
L2 speakers of English employ diverse communication strategies to resolve their communication issues. Communicative strategies may be defined as techniques employed by individuals in order to bridge the gap between their linguistic knowledge and communication goal (Maleki, 2007;Saeidi & Farshchi, 2015). Although many L2 learner-oriented studies of second language acquisition investigated communicative strategies, they tend to concentrate more on their oral skills in general than phonology (e.g. Bialystok, 1983;Dörnyei & Scott, 1995;López, 2011;Paribakht, 1985;Poulisse, 1993).
Two studies in the Canadian context investigated communicative strategies of L2 adult learners at language institutions focusing on their phonological difficulties (Derwing & Rossiter, 2002;Osburne, 2003). Derwing and Rossiter (2002) examined 100 ESL adult learners from diverse first language backgrounds at an intermediate level of language proficiency. Their mixed methods study found that the majority of the participants believed their communication difficulties stemmed from their pronunciation. In addition, over a third of respondents considered that the primary cause of communication breakdown was due to accent (Derwing & Rossiter, 2002). Commonly used strategies identified among this group of learners were "paraphrasing, self-repetition, writing/spelling, volume adjustment, slowing speech rate and speaking clearly" (Derwing & Rossiter, 2002, p. 163).
On the other hand, Osburne's experimental study (2003) focused at the advanced English level on 50 L2 adult learners from diverse first language backgrounds. This study was distinctive due to the experimental approach. A content analysis of the discussions revealed that the most preferred compensatory strategies for oral communication among the advanced group were: (1) memory or imitation, which means the participants copied or remembered what the interviewer said; (2) paralanguage, for example, correction in voice quality, speed, and clarity; (3) local articulatory gestures or single sounds, for example, gestures of tongue and lips for articulatory phonetics or single sounds in a certain word; and (4) pronunciation of an individual word as a whole unit.
While previous studies have focused on the strategies used by L2 learners at L2 language institutions in addressing communication difficulties, there has been little focus on the situation regarding international students at tertiary level in the Australian context. This group is important due to their position associated with studying in Australia; further, they are a distinct group with generally higher levels of speaking skills and also quite explicit communication needs due to their tertiary study. Thus, it is likely that international university students in the Australian ESL context may have different strategies in overcoming their unintelligible sound production.
The study reported in this paper focuses on how accent plays out in English communication of L2 international students in Australia. It looks not only at the communicative strategies that these students use to mitigate communicative problems with their accented English, but also explores how effective the L2 students perceive these strategies to be. This will provide a richer understanding of the students' linguistic experience, and thus support the development of coping strategies among L2 international students in Australia.

Research aim
Language difficulties of L2 international students have been one of the enduring issues with respect to their academic, social, and cultural adjustment. This study investigates one of the major communication issues of these students, followed by corresponding communication strategies they have adopted. Although it was noted that both the accented English of these students and the accents of others were found problematic, we focus on the issue of accent faced by the students and how they mitigated the challenges accented English caused. Thus, the purpose of this study is threefold: to explore (1) the experiences of communication difficulties caused by accented English; (2) the communicative strategies to mitigate these challenges; and (3) the perceived effectiveness of these strategies.

Participants and settings
According to the Department of Education and Training (2016), of the 498,155 international students in Australia, 258,369 (51.8%) were enrolled at universities. The majority of international students in higher education were Asian, with almost half of these from China (27.3%), India (10.8%), and Vietnam (4.4%). In Queensland, 28,097 international students were enrolled at three major universities located in Brisbane (Department of Education & Training, 2015). International students in higher education in Australia have different levels of language competence. The minimum requirement for the majority of Australian universities ranges from 6.5 to 7.0 with no sub-score of less than 6 to 6.5, based on faculty level (Read, 2015). This study aimed to target respondents at the minimum level with an overall IELTS (International English Language Test System) score of between 6.5 and 7, thus representing competent L2 speakers. This study excluded students with an IELTS score of greater than 7. This was because the difficulties of those with a lower IELTS score could be more extreme than those with a higher score and so would provide a stronger focus on the issue. Thus, it was felt less useful to include that group in the sample.

Survey instrument
Survey questions were developed based on previous research literature (Derwing & Munro, 2009;Khawaja & Stallman, 2011;Lacina, 2002;López, 2011;Osburne, 2003;Sicat, 2011), for example, particular contexts in which international students encountered communication difficulties due to accentedness and the diverse communicative strategies employed by L2 speakers at different levels of language proficiency focusing on their sound adjustment. Respondents were asked to report on three areas of language use against assessments using a five-point scale. Firstly, respondents were asked if native and non-native English speakers easily understood their accented English. They were asked to assess their experiences (1: always; 2: most of the time; 3: not sure; 4: seldom; 5: never). Secondly, respondents were asked to comment on the frequency of communication breakdowns in a range of situations, with frequency assessed on a five-point scale (1: very often; 2: often; 3: sometimes; 4: rarely; 5: never, with an additional option of Uncertain). Finally, respondents were asked to evaluate a range of communication strategies they used, the frequency of use, and perceptions of effectiveness. Frequency was assessed by the five-point scale (1: very often; 2: often; 3: sometimes; 4, rarely; 5: never). Perceived effectiveness of their strategies was also assessed on five-point scales (1: very effective, 2, 3, 4, 5: not effective). Both the frequency and perceived effectiveness scales had an option: Uncertain.

Procedures
Prior to commencement of data collection, this study was approved by the Griffith University Human Ethics Committee. This study used two different forms of surveys, an online survey and a paper survey in order to maximize response rate.

Sampling
Participant recruitment was conducted at three major universities in Brisbane. The majority enrolled at these universities were undertaking Bachelor's degree, followed by Master's by Coursework and Higher Degree Research (Griffith University, 2014; Queensland University of Technology, 2016; University of Queensland, 2015).
To achieve a narrow band of lower English proficiency, participants who professed to have an IELTS score of 6.5 and 7 were invited to participate in this study. Students are very familiar with their IELTS score as this is used as an entry language requirement at universities. Its disclosure in this study would have no negative impacts, thus it was felt this measure provided an efficient screening tool. The sample was selected using a snowballing sampling technique for the completion of data collection and recruitment of participants. The researcher contacted potential participants personally and asked them to recommend the survey to their friends and peers. Invitations and the survey were distributed via email and personally. In addition, in order to ensure sufficient numbers of participants, invitations were distributed through EnglishHelp Higher Degree by Research (HDR) students by the Griffith English Language Institutes (GELI). This study included female and male L2 international students, who were undertaking undergraduate and postgraduate programs.

Data analysis
A total of 188 responses from online and hard copy surveys were gathered including some incomplete responses where informants answered most of the questions. However, six responses were eliminated as no attempt was made to answer any of the questions in the survey. For undertaking statistical analysis, the Statistical Package for the Social Science (SPSS) was used. Data were collected in two different ways: 57.7% (n = 105) of participants responded online and 42.3% (n = 77) of participants responded by hard copies. The responses collected through the paper survey were entered into SPSS in order to analyze data efficiently. Descriptive statistics such as frequencies and means were calculated. Subsequently, data were analyzed in further detail with the use of demographics (e.g. gender, age, level of education, and length of stay in Australia) of the participants to decide whether there was any relationship among the variables. Descriptive analysis identified overall perceptions of the participants' communication experiences, together with communication difficulties caused by their accented English and the communicative strategies they use to alleviate those challenges.

Demographic profiles of the participants
Varied responses were received from a total of 182 students and these are summarized in Table 1. A far greater proportion of participants were female (65.9%, n = 120) than male (34.1%, n = 62).
This study included a wide range of nationalities, summarized in Table 2. The respondents were from 30 countries with the majority having Asian backgrounds. In particular, the largest number of participants were from Korea (20.3%, n = 37), followed by China (14.8%, n = 27), Japan (7%, n = 12), and India (6%, n = 11). This was fairly consistent with the national figures.

Constraints of communication
In the survey, the students were asked about their perceptions of whether native and non-native English speakers could understand them easily. This identified perceptions of their overall communication experience with regard to their accent. Gender difference showed an interesting contrast, especially when they communicated with L2 speakers. It is interesting to note that these competent users of English were understood better by L1 speakers than L2 speakers. Figure 1 firstly summarizes selfreported communication experience of the participants when they communicated with L1 speakers.
The vast majority of participating informants responded that their accented English was understood by L1 speakers most of the time (F = 70%, M = 64.5%). On the other hand, a few students were unsure to what extent their accented English was understood by L1 as 20% of females and 25.8% of males selected "not sure". Not many of the respondents regarded their accented speech as always intelligible, nor did they consider that they were hardly ever understood by L1 speakers. The findings of communication difficulties with other L2 speakers are presented in Figure 2. Despite the fact that those who participated in this project are defined as competent users of English, they reported experiencing more communication difficulties with L2 speakers than with L1 speakers. In particular, for male participants, the level of uncertainty of their intelligibility with L2 was far higher than for female participants.
As Figure 2 demonstrates, there was a high level of uncertainty in this assessment with 66.1% of male participants selecting "not sure" for their intelligibility with L2 compared to 39.2% of female participants. These levels are far higher than were reported in Figure 1. Over half of the female respondents (55%) and 30.6% of the male students indicated that their accented speech was still intelligible to L2 speakers most of the time, which was also far less than that in Figure 1.
Overall, it was found that the participants presumed that their accented English was more intelligible to L1 speakers than L2 speakers. In particular, it was male participants who felt more uncertain about their intelligibility of their accented English than female participants when communicating with other L2 speakers.
It was assumed that participants who had been in Australia for a longer period may have encountered fewer difficulties than those who recently arrived. Thus, communication difficulties on the basis of frequency were analyzed based on their length of stay. Interestingly, no marked differences in communication difficulties were found. Table 3, participants with a longer length of stay in Australia (5+) still faced difficulties and those difficulties were not markedly different from those who had been in Australia for less than a year. In particular, in educational settings, many reported communication difficulties. The frequency of communication difficulties for these two groups of participants with maximum (5+) and minimum stays (<1) in Australia was within the same range: participating in tutorials and discussions (M = 2.7, M = 2.9, respectively) and understanding lectures (M = 3.0, M = 2.9 each). In addition, Figure 1. The participants' selfreported intelligibility with L1 speakers.

Figure 2. The participants' selfreported intelligibility with L2 speakers.
the result shows participants with 2-3 years of residence in Australia faced more difficulties than those with a minimum stay, when participating in tutorials and discussions (M = 2.5).
Results regarding difficulties within social settings showed a similar pattern. It was not recently arrived participants who faced the most difficulties due to their accented English. Participants with 5 years of stay did not appear to experience remarkably fewer difficulties than other groups, as their responses were ranged within the same categories. However, as the discrepancy in numbers, participants who had been in Australia for 2-3 years appeared to have more difficulties than the other groups. One consideration not assessed here could be that the level of exposure also changes over time, thus with increasing time in Australia, students may broaden their social experiences and thus be more exposed to situations where difficulties occur.

Communication difficulties caused by accented English of the students
The participants were asked to indicate how often they experienced communication difficulties because of their accented English. Given their level of language proficiency, it was expected that responses would not indicate "rarely" or "never" significantly. In fact, a few number of students answered that they rarely or never encountered any problems with their accentedness. However, as Table 4 illustrates, the majority of participants did experience communication issues "sometimes" or "often" depending upon the situations. While shopping was the least problematic context, participating in tutorials and discussions resulted in substantial communication issues.
The overall experience of communication difficulties in the given situations shows that while students sometimes encountered communication issues in social settings, they often experienced problems caused by their accent in an educational setting. Over half of the participants answered "sometimes" to social interaction (52.7%), followed by understanding lectures (52.3%) and job interview (52.3%). Presumably, shopping was the least demanding context for the students as there was no considerable difference in the data for "sometimes" and "rarely". These occupied 39.5 and 39%,  respectively. It is interesting to note that a predominant trait was found in one of the educational settings: 43.4% of the respondents answered that they often faced difficulties when they participated in tutorials and discussions.
In conclusion, we found that participants who are defined as competent users of English in this study, infrequently encountered communication difficulties caused by their accented English. Nevertheless, they confronted accent-related communication difficulties frequently when participating in speech-involved learning activities in their educational situations.

Favorite and effective strategies for repairing communication difficulties
Participants were asked to assess both how often they used each strategy listed in the survey and how effective each of these was. This study identified contrasting views between actual communicative strategies employed by the students and their assessment of the degree of effectiveness of each strategy. Figure 3 summarizes these results.
The mean scores on frequency and effectiveness of each strategy identified an obvious contrast, which demonstrated differences between frequently employed strategies during conversation by the students and their effectiveness. The most effective strategies chosen by the participants were pronouncing more precisely (M = 4.2), followed by paraphrasing (M = 4.1) and speaking more clearly (M = 3.8), although the participants did not employ such strategies as frequently despite their effectiveness. On the other hand, the most popular strategies selected by the participants were repeating (M = 3.8), followed by paraphrasing (M = 3.6) and speaking more clearly (M = 3.6).
The least popular strategy was writing down or spelling out (M = 2.6), even though the participants regarded this strategy (M = 3.3) as fairly effective. The least effective strategy turned out to be speaking louder (M = 2.4), but they employed this strategy more frequently than writing down or spelling out.
Taking these into account, the significant differences in the mean scores were identified as contrasting results between the students' favorite strategies and the effectiveness of the strategies. It seemed that they were well aware of the significance as well as effectiveness of pronunciation to compensate for communication breakdowns caused by their accentedness. In contrast, their actual conversational strategies did not include clearer pronunciation.

Discussion
The current study had three central research foci, firstly, to better understand the overall communication difficulties of international students due to accented English; secondly, to focus on the communicative strategies used to resolve communication breakdowns; and lastly, to explore perceptions of effectiveness of these strategies. Given participants' reasonable language proficiency, they were capable of managing communication with others; however, the findings of this study show that accented English played a role in communication as well as academic studies.
Overall communication experience with both L1 and L2 speakers was compared, identifying that L2 communication was more difficult for the participants than with L1. The impacts of gender difference as well as length of stay in Australia were explored. Gender difference showed a marked difference, whereas a longer length of time spent in Australia seemed to have a minor influence. While participants indicated a great awareness of the significance of pronunciation as a critical area to mitigate issues with their accents, their preferred strategies did not include such skills.
Although many studies concentrate on intelligibility of non-native English speakers for native interlocutors (e.g. Hahn, 2004;Munro & Derwing, 1999), international intelligibility (L2 speakers for L2 interlocutors) has not been paid much attention (Pickering, 2006). The current study revealed that L2 international students in the Australian context had more issues with L2 speakers than L1 speakers due to their accentedness. In some instances, foreign accented English appears to be understood by native listeners, including some local dialects (Munro & Derwing, 1999). On the other hand, when it comes to L2 communication, it can be complex because they may encounter difficulties in understanding accented speech which does not contain expected sounds (Munro, Derwing, & Morton, 2006). This finding contributes to our understanding of the breadth of communication issues faced by L2 students, difficulties both with being understood and also understanding other less standard accents.
This study considered a link with gender, in particular, in light of accentedness. Interestingly, the results indicated a clear gender difference, with male students reporting that they struggled more with their accented English than females. According to Horwitz (2001) and Tsui (1996), speech intelligibility and comprehensibility are associated with L2 anxiety and reticence, especially with L1 speakers. This is because a fear of intelligibility and comprehensibility can provoke L2 speakers' performance anxiety. Thus, it is likely that accented English is one of the anxiety-provoking areas of communication for L2 speakers; further, it is probable that male L2 speakers may feel more anxious about their accented English than females, given the uncertainty they indicated.
With regard to time in Australia, a longer length of stay tends to have great advantages in some linguistic areas such as level of vocabulary knowledge and overall language competence (Pérez-Vidal, 2014). In view of that, it is likely that both a reduction of distinctive foreign accent and also better speech intelligibility occur with a long duration of residence in Australia. Despite this, this study found no such effect. Similarly, some studies that investigated foreign accent and the effect of range from two years to over five years concluded that the impact of length of stay was unimportant as it did not ensure less foreign-sounding accents (Oyama, 1976;Tahta, Wood, & Loewenthal, 1981). From this study, it appeared that the interaction between duration of time in the country and issues with accentedness was not direct, with, for example, opportunities to practice or improve accented English and understand other accents also a contributor to communication. Thus, the overall level of exposure to the target language or the amount of continued practice may also need to be considered along with time in Australia.
Given the language proficiency of the participating students, it was assumed that their communication issues might not be at a critical level. The outcomes of the study suggested that the participants did not frequently experience difficulties due to their accented English, but rather sometimes as the frequency measurement demonstrated. Despite this, in comparison with other social settings, L2 students in Australia experienced accent-related issues more in an educational setting and specifically when participating in tutorials and discussions. This aligns with the findings of Khawaja and Stallman (2011), who recognized the problems with L2 students' accents in educational settings, especially in a lecture format. However, as this study compared two typical types of educational settings, lectures and tutorials and discussions, it was found that problems in tutorials and discussions outweighed those in lectures. This is likely to be associated with the need to more actively participate in tutorials than in lectures. Thus, this study informs our understanding of the potentially negative effect a foreign accent may have on students' active participation and subsequent academic progression.
One interesting result of this study is the mismatches between the participants' favorite and most effective communication strategy. According to the communication strategy data, participants showed greater reliance on repetition, followed by paraphrasing and speaking more clearly, in order to overcome accent-related issues. In fact, repetition was not an influential strategy to resolve communication breakdowns as its effectiveness was lower than frequency of use. With regard to the effectiveness of the strategy data, the most effective strategy was related to pronunciation, followed by paraphrasing and clarity of speech. It is possible that an ability to paraphrase is closely linked to communicative success, given the evaluation of its frequency of use, as well as its perceived effectiveness by the participants. As already addressed in previous work (Derwing & Rossiter, 2002;López, 2011), paraphrasing is one of the popular strategies among ESL and EFL speakers, regardless of their language proficiency. Not only can an ability to paraphrase lead to successful communication but also an instruction of such strategy enables the use of various paraphrase types and access to a wide range of alternatives (Rossiter, 2001).
Participants seemed to be aware of the significance of pronunciation as well as its effectiveness to overcome challenges caused by their accented English. Other studies have also explored pronunciation, finding that many L2 students indicated pronunciation-related perception of difficulties (Derwing & Rossiter, 2002). Further, they were concerned that their English might not be understood by other English users, unless they improve their accents (Tokumoto & Shibata, 2011). Thus, while high language proficiency can play a role in better communicating with others and overcome communication breakdowns, for tertiary L2 students, pronunciation is still an area which needs further improvement to mitigate issues with their accented English.
Exploring L2 students' communication experience in relation to their accented English can give better understanding of both theoretical and practical issues. The fact that their non-native English affects their communicative competence and their academic progression should not be overlooked. The findings of this study can inform Australian education providers as well as university teaching staff of the importance of strategies to accommodate these students. In particular, increasing the awareness of university teaching staff regarding current L2 students' experience is an essential step as it is this group that has a direct contact with these students. Given the students' issues with their accented English and its impact on their study experience, they may have fallen victim to stereotypical evaluation. This is because listeners may find it difficult it separate distinctive accents and communicative competence of speakers (Creese & Kambere, 2003). Thus, university staff should be encouraged to view these issues differently and make more social adjustments to courses, decreasing potential bias that occurs with L2 students. By creating more understanding environments, students would engage with other students and better manage their learning.
Being able to speak English as a second language is a valuable asset for many L2 speakers as this can act as an influential mechanism for social integration and economic betterment (Park, 2011). However, L2 speakers with a foreign accent can be devalued to some extent in English-speaking contexts. Bourdieu and Thompson (1991, p. 55) stated that "speakers lacking the legitimate competence are de facto excluded from the social domains in which this competence is required, or are condemned to silence". Although L2 students have met the minimum language requirements, and as such should be comparatively competent and capable of managing their study in their second language, their accent can still constrain their communication and thus academic progression. The feeling of social exclusion, leading to embarrassment and even a decision to remain silent was seen in this study, and also noted by Nogita (2010), who also found that sometimes L2 students express a fear of being mocked and feel embarrassed when they attempt to pronounce English words accurately. An appreciation of the significance of such issues by university teaching staff is important if they are to support such students in achieving the academic outcomes of which they are capable.
To improve problems with accented English on the part of students, English programs for future tertiary students should include pronunciation instructions for better speech intelligibility. Many potential L2 tertiary students may focus on reaching IELTS requirements and writing skills rather than intelligible speech with good pronunciation. In language education, generally, pronunciation instruction is emphasized more in lower language proficiency classes than in advanced ones. However, pronunciation instruction should not be marginalized reflecting the needs of the current L2 international students. Language teachers can help students improve their pronunciation without focusing too much on the reduction of their own linguistic backgrounds, understanding their variant of English (Mallinson, Charity Hudley, Strickling, & Figa, 2011).

Conclusion
The current study investigated widespread and persistent communication difficulties encountered by L2 international students during their sojourn in Australia. By focusing attention on one specific element of spoken English language-accented English-it demonstrates that accent is a significant factor that can have marked impact on students' communicative as well as educational success. The findings indicated that participants were aware of the problems caused by their accents and did use strategies to manage this; however, greater awareness and support from the educators is important for strategies to be effective.
It is important to acknowledge that this approach was not free from limitations. Firstly, it does not allow extrapolation of the views of participants regarding the inconsistency between preferred compensatory communication strategies and the effectiveness of these strategies. Another limitation was that known problems of accent stereotyping and discrimination against foreign-accented speech have not been addressed. Undoubtedly, unfair judgment or incorrect assumptions toward L2 students from non-English-speaking backgrounds may adversely impact adjustment to educational and social settings. Although students with higher language proficiency may be more capable of dealing with communication breakdowns, as demonstrated, linguistic-associated difficulties, particularly accentedness, present challenges for L2 international students. This is worth further investigation as are the strategies that L2 students can adopt themselves in understanding accented speech.
However, this study does provide an important focus on an issue impacting the academic outcomes of L2 tertiary students. By better understanding the specific language difficulties faced by tertiary L2 speakers, strategies to address such issues could enhance the experiences and outcomes of higher education and university services for L2 international students.