Language Attitudes in Australia: Results from a Nationwide Survey

Abstract

Recent research on attitudes to Australian English (AusE) shows that there is a general increase in its acceptance, legitimacy, and endonormativity. However, a certain “cultural cringe” exists, particularly when “broad” AusE is seen as representative of the variety. A significant gap in the literature is how the perceptions and usage of AusE may change as the population becomes more diverse. This paper presents findings of an online survey of language attitudes towards AusE with 661 respondents across Australia, over a third of whom were born overseas. Overall, there is minimal evidence of a standard language ideology, with 80% of respondents reporting having an accent to some degree. Almost half of respondents report occasionally or frequently changing their accents due to context, interlocutor, or making themselves understood. When asked to rate AusE along six traits on a seven-point scale, the traits of educatedness, professionalism, and attractiveness were consistently centered on neutral. For friendliness and likeability, the majority skewed towards neutral and positive. For the trait of clarity, there was a greater range of responses, but overall, 50% of respondents found AusE to be somewhat, moderately, or really clear. These findings further our understanding of attitudes and ideologies in Australia’s increasingly diverse language ecology.

1. Introduction: Language Attitudes in Australia

Language attitudes are evaluative reactions to different language varieties, reflecting the sequential cognitive processes of social categorization and stereotyping (Dragojevic 2017). Language attitudes tell us about how people feel about their own languages and varieties and the languages/varieties spoken by others. These feelings are inextricably linked with how they feel about the speakers of those languages/varieties and, thus, are reflective and constitutive of identities. Attitudes and identities can shift in response to societal norms, practices, and demographics, and they are, thus, not static but subject to change. This paper provides an overview of contemporary language attitudes towards Australian English (AusE) as captured in a nationwide survey with the aim of representing the views of Australia’s increasingly diverse society. Among 661 respondents, 66% were born in Australia and 34% born overseas across a range of countries and continents: a ratio which is on par with the national average of 27.6% born overseas (ABS 2022). We provide the first known study on attitudes towards AusE which explicitly compares these two groups, offering an up-to-date study on Australia’s linguistic ecology.

1.1. Folklinguistic Understandings of Broad and General Australian English

Recent research on attitudes to Australian English shows that there is a general increase in the acceptance of the legitimacy and endonormativity of this variety (Willoughby 2020). However, a certain “cultural cringe” exists, alongside an ideology that “celebrates Australian culture for its informality, mateship and egalitarianism and rails against a stultifying, classist British past” (Willoughby 2020, p. 224). These attributes tend to be strongly linked to Australian identity, of which the Australian accent or AusE is an important part (see Penry Williams 2019). The complicated relationship that Australia has with Britain is borne out in studies showing that a historical deference and inferiority towards the prestige voices from the “Home Country” (but also the colonizer) has now developed to a situation where American English is predicted to become the new prestigious variety in the Antipodes (Bayard et al. 2001; see also Bradley and Bradley 2001 for an overview of the development of AusE attitudes over time as portrayed in the media and literature).

Stereotypes and folk linguistic perceptions surrounding AusE are ubiquitous, particularly when “broad” AusE1 is seen as representative of the variety as a whole (Willoughby 2020; Kingstone 2019; Penry Williams 2019). Some of these stereotypes are typified in personae or characterological figures such as Steve Irwin2, who are seen, according to Willoughby (2020), as embodying the “broad” AusE speaker, along with the traits of toughness, masculinity, and a certain fearlessness in the face of nature and the outdoors. Supporting this view, Sussex (2004, p. 12) describes stereotypical AusE speakers as “laconic, self-deprecating individuals, usually male, rather taciturn but prone to intermittent bursts of humorous, creative language”. However, despite the body of work on stereotypes surrounding broad AusE, other research focusing on “general” AusE has found it to have become more acceptable over time (Price 2012), particularly for status and solidarity, and power and competence (Bayard et al. 2001; Bradley and Bradley 2001).

While the general trend in language attitude studies is that “prestige accents” score highly for status and “local or regional accents” score highly for solidarity and affiliation (Cargile et al. 1994, p. 224), Australians in a matched guise study were found to rate their own variety as highly as British and American varieties for status but lower than American English for solidarity (Bayard et al. 2001, p. 40). A panel study by Bradley and Bradley (2001) found that among the same group of Australians interviewed in 1980 and again in 1995, a total of 18.5% of the speakers became more positive about AusE over time. However, in a matched guise study, they specifically found that broad AusE did not elicit the high solidarity ratings that one would expect but rather general AusE. This all suggests growing but not complete security in AusE, and that this security is contingent on the variety being neither too broad nor too cultivated or British-like.

In an extensive folk linguistic study in Melbourne, Penry Williams (2019, chap. 7) proposes five main local identities: the ocker (representing chauvinism and boorishness, but also quintessential Australianness), the wog (associated with migrants, particularly the 1950s wave from Italy, Greece, and other parts of the Mediterranean), the bogan (representing the working class, similar to “white trash” in the U.S. context), the Queenslander (referring to a state in the north-east, but specifically more rural parts of that state), and posh types (a negative type meaning stuck-up or snobby). However, these social types represent extreme ends of a spectrum, and the way Australians view their own accents as relating to their own (presumably more complex) identities remains understudied.

1.2. Migration and Changing Attitudes to Australian English

A further question surrounds the impact of recent increases in migration to Australia and how this would affect the linguistic ecology. Newcomers to Australia may feel their sociolinguistic identities to be at odds with, or even challenged by, the personae or tropes as described in Penry Williams (2019). Australia has always been a country of immigrants, but its population is becoming increasingly diverse, with the proportion of overseas-born individuals at an all-time high of 27.6% as compared to the first census of 1911 when the figure was at 17.7% (ABS 2022). The work of Dovchin (2019) argues that language ideologies in Australia must be well understood in order to better explain the processes of identity and belonging, and conversely, the processes of exclusion and prejudice, or “linguistic racism” that can be experienced by migrants, and indeed any other kind of non-mainstream speakers. In a confederate-scripted study by Kidd et al. (2016), it was found that when speakers used AusE hypocorisms, which are known to be emblematic of AusE (e.g., uni for university, servo for service station, and Maccas for McDonald’s), there was significantly higher participant-rated perceived common ground when the speaker who used the hypocorisms also had an Australian accent, compared to a non-Australian (“Asian”) accent. This suggests, supporting Dovchin (2019), that discrimination based on accent exists in Australia.

Despite this, there has been an acknowledgement of the diversity of Australian accents in mainstream culture going back to the 1980s, when the Special Broadcasting Service (SBS), who describe themselves as “Australia’s most diverse broadcaster”, announced that its news anchors were no longer required to have “native speaker accents” (Bonner 2011, p. 100; Willoughby 2020, p. 229). This acknowledgement is corroborated somewhat by a study by Starks and Willoughby (2015), who found that 10% of Australian students think migrants should learn AusE and 40% think they should learn AusE slang, suggesting a certain freedom or openness surrounding ownership of AusE, at least on a lexical level. A modified matched guise study by Eisenchlas and Tsurutani (2011) found that Australian university students studying foreign languages rated their foreign-accented lecturers highly in many personality dimensions under the broader categories of competence, integrity, and attractiveness, indicating “students’ greater readiness to accept foreign accents” (Eisenchlas and Tsurutani 2011, p. 216). Notwithstanding, there is a well-known “monolingual mindset” in Australia (Hajek and Slaughter 2015), and little research to date has been conducted on language attitudes in contemporary Australia among its overseas-born population.

1.3. Research Questions

Overall, previous research on attitudes to AusE finds that, firstly, there is a conflicted relationship with British English as the “inherited” or “coloniser” language in Australia, which has resulted in progressively more positive attitudes to general (not cultivated) AusE over time. However, a “cultural cringe” persists, particularly for broad AusE, and other varieties, such as American English, may be ranked higher even for traits such as solidarity, where preference for one’s own variety would be expected. Secondly, the linguistic ecology in Australia appears to be changing in the face of increasing diversity, with a greater range of accents appearing on the nationwide SBS broadcaster for radio and television, and some growth in acceptance of the idea that (a) non-Australian accents are acceptable in Australia; and (b) non-Australians are licensed to take on AusE, at least on a lexical level. Nonetheless, Australians have been found to have a preference for AusE lexical items, such as hypocorisms, to be spoken with an Australian accent, such that migrants face a challenge to be accepted if they wish to pass as authentic speakers of AusE. Indeed, linguistic racism and discrimination continue to be issues in Australia, although broadly speaking, work in the area of perceptions of AusE by non-Australian-born populations is limited. This paper responds to calls by Willoughby (2020) for further understandings of language attitudes towards AusE by its increasingly diverse population, and it is guided by the following research questions:

• What are the attitudes and ideologies towards AusE held among Australian-born and overseas-born residents of Australia?
• How do Australian-born and overseas-born respondents differ in their ratings of AusE along the six traits of educatedness, professionalism, attractiveness, friendliness, likeability, and clarity?

2. Materials and Methods

This paper draws on data from an online survey designed and distributed using Qualtrics, with ethical approval from (Western Sydney University). The survey was publicized on mailing lists, on the first author’s social media account, via a radio interview with the first author, and via an article in Diskin-Holdaway and Escudero (2021).

We present results from 661 respondents who filled out the survey between 1 November 2020 and 12 August 2021. The survey had 42 questions in total (excluding screening/eligibility and consent questions), although some questions were only applicable to those who had provided certain responses previously, so no one respondent completed all 42 questions (see Appendix A for examples of how respondents were redirected to certain questions). We present a subset of 27 questions, five of which were essential demographic questions, and 22 of which were selected for their relevance to the research questions.3 These questions have been renumbered from 1 to 27 in Appendix A for readability. The first message that appeared on screen after clicking on the survey link was “Thank you for taking part in this survey about accents in Australia”, so participants were aware of the focus of the study from the outset.

The survey had an average completion time of 12 min and could be completed on any kind of device (i.e., desktop, laptop, tablet, mobile phone). A total of 73% of respondents completed the survey in full, and 27% of respondents completed a portion of the survey, ranging from 4–94% (average 35%). The responses of those who did not fully complete the survey are included in our analysis if they responded to the relevant questions. In most cases, we indicate the number of respondents for each individual question.

Beginning with an overview of the responses to five demographic questions, we then present an analysis of ten survey questions about accent, for which participants were given no explicit instruction as to how to answer. The questions about accent were designed to encourage respondents to think about their own linguistic (in)security (Labov 1966), whether they had a standard language ideology (Lippi-Green 1997), and whether they change or modify their accent(s) as part of accommodation (Giles et al. 1991) or through wanting to integrate linguistically.

We also present an analysis of six seven-point Likert scale questions, which were presented in blocks in a randomized order, and six follow-up, open-ended questions asking respondents to explain their ratings. Participants were shown the following message on-screen before starting the Likert scale questions: “The following questions will ask you what you think about the Australian accent”. These questions were influenced by matched guise studies (Lambert et al. 1960; see also Campbell-Kibler 2007 and Villarreal 2018 for further matched guise studies, and Giles and Billings 2004 for an overview of attitudes and speaker evaluation studies), which are designed to gain insight into language attitudes, stigma, and prestige typically in a more experimental setting.

Matched guise studies typically focus on the two broad categories of status and solidarity (Dragojevic 2017). The six traits of educatedness, professionalism, attractiveness, friendliness, likeability, and clarity in the present study match 4 of the 13 traits in another matched guise study on AusE using Likert scales (Bayard et al. 2001), which include intelligent, competent, cheerful, and warm. This enhances the reliability of the research instrument. Attractiveness has been included in other language attitude studies, including another on AusE ratings on foreign-accented English using a 7-point semantic differential scale (Eisenchlas and Tsurutani 2011). Clarity was added to the present study as it was felt that this would be an important dimension for those born overseas but living in Australia and encountering AusE. Linguistic (in)security, standard language ideology, linguistic accommodation, and language attitudes are fundamental concepts in sociolinguistics that give insight into language variation and change, language acquisition and maintenance, and language and identity. To the best of our knowledge, these concepts have received limited exploration in the context of AusE and no previous exploration among overseas-born individuals in Australia.

3. Results

3.1. Demographic Questions

In Q1, respondents were asked for their year of birth, which allowed us to calculate their age at the time of filling out the survey. Of the 72% (n = 476) who filled out this question, the mean age was 45 (SD = 13). In Q2, they were asked how they self-identified with respect to gender, again with 72% giving a response, and of this, 66% responded female, 32% male, and 2% other categories such as transgender or non-binary. Q3 asked whether they were born in Australia or another country, and again 72% responded to this question, among which 66% reported being born in Australia, and 34% overseas, which is somewhat above the national average of 27.6% (ABS 2022). The places of birth listed were diverse, including countries such as the Philippines, Belgium, and Iran.

Respondents were requested in Q4 to provide their postcode (94% response rate), which allowed us to ascertain which states they resided in: 56% were in Victoria, 27% in New South Wales, 6% in Queensland, 5% in Western Australia, 4% in South Australia, and 1% in both Tasmania and the Northern Territory. Respondents were asked in Q5 to select any and all of the languages they spoke, with an 83% response rate. Of those, 43% selected English only, and 29% selected English and another language or languages. This is somewhat above the national average of 22.3% of people in Australia who speak a language other than English at home (assuming that this is a somewhat reliable measure of bilingualism), according to ABS (2021). The remaining 29% represents relatively high numbers of speakers of French, Spanish, and different varieties of Chinese. Languages included in the “Other” category by participants included Irish, Maltese, Neapolitan, and Sanskrit.

3.2. Questions about Accent

3.2.1. Do You Speak English with an Accent?

Q6 of the survey asked respondents “Do you speak English with an accent?” (n = 617;

Table 1. Responses to the question “Do you speak English with an accent?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Definitely Not	Not Really	Somewhat	Mostly	Definitely
Overall (n = 617)	9%	10%	23%	10%	47%
Australia (n = 315)	10%	10%	26%	8%	46%
Overseas (n = 161)	3%	9%	24%	14%	49%

). Responses indicated that 80% of all respondents felt that they had an accent to some degree (Table 1). Furthermore, 77% of respondents to this question had also provided their place of birth, and these results were analyzed by those born in Australia and those born overseas. The responses appeared fairly similar, although there was an increase of 3% for “definitely” and 6% for “mostly” among overseas-born as compared to Australian-born respondents, along with a decrease of 7% under the “definitely not” category. Following Pearson’s chi-squared test, the responses of the overseas-born group were found to be significantly different from the responses of the Australia-born group X²(4, n = 476) = 11.55, p < 0.02).4

Overall, it seemed that respondents were mostly conscious of the fact that they had some kind of accent, even if that accent was the mainstream one in the community. Among the 19–20% who reported not really or definitely not having an accent, 89 out of the 106 responses to Q8, to which they were redirected (“Why do you think you do not speak English with an accent?”), could be described as centering on seven main themes which emerged naturally from the data. First, 34 respondents seemed to ascribe to a standard language ideology or a “bias towards an abstract, idealized homogeneous language” (Lippi-Green 1997), believing that their lack of accent was due to heritage, lineage, and even monolingualism, suggesting a purist understanding of what it means to be accentless:

My parents and extended family are all native English speakers. Everyone in my family is also monolingual.

(Participant #542, born in Australia)

I have English speaking parents, grandparents and great grandparents and I have lived in Australia my entire life.

(Participant #510, born in Australia)

Second, 17 respondents explained that others had noticed this trait about them; however, most of these responses alluded to the fact that they felt they had an Australian accent (not no accent at all):

Everyome [sic] tells me I dont [sic] have an accent/think I am Aussie when they meet me.

(Participant #559, born in Germany)

Third, 12 overseas-born participants felt they had “no accent” because they had “lost” the accent of their country of origin but admitted to having an Australian accent, which again appeared to not “count” as an accent:

Ppl [sic] on the telephone do not know they are speaking to a Chinese person, and have Anglicised my name in error. I have done all my schooling in Australia and can hear my Australian accent.

(Participant #566, born in Hong Kong)

Fourth, nine participants felt as though they had no accent because they sounded like the people around them:

The way that I speak does not sound significantly different from the majority of English speakers in my community. English is my first—and only functional—language. If I listen to Australian radio or TV, I sound like most speakers.

(Participant #517, born in Australia)

Fifth, seven participants explicitly stated that they felt they had no accent because they did not have a “foreign” (non-Australian or non-Anglo Australian) accent:

I selected ‘not really’ as I suppose I have an Australian accent; but I am not an Australian with an accent other than Australian (e.g., British, Canaian [sic], Lebanese etc.).

(Participant #504 born in Australia)

Sixth, there were six respondents who explicitly recognized the privilege of speaking an unmarked, prestigious variety:

Well… I guess I have an Australian accent. But I think I speak English without an accent because I went to a private school and come from an upper middle class background where we learned to speak a certain way, although we do often speak with an Australian accent or use Australian slang.

(Participant #530, born in Australia)

Seventh, four respondents stated that they could “hear themselves” as either not having an accent, or as having a “mainstream” accent:

I have heard myself speak (e.g., mobile-phone made videos) and I think that my accent is not dissimilar to the broader population around me (e.g., local community; the workplace; television).

(Participant #546, born in Australia)

3.2.2. How Much Do You Like Your Accent?

For respondents who had indicated in Q6 that they felt they spoke English with an accent, they were redirected to Q9 asking “How much do you like your accent when you speak English?” (n = 481). The most popular response for this question was “neutral” at 34% (

Table 2. Responses to the question “How much do you like your accent?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Dislike	Moderately Dislike	Somewhat Dislike	Neutral	Somewhat Like	Moderately Like	Really Like
Overall (n = 481)	1%	4%	9%	34%	15%	25%	14%
Australia (n = 251)	<1%	2%	8%	31%	18%	29%	12%
Overseas (n = 141)	2%	9%	8%	34%	11%	18%	18%

). Over half of the respondents were positive about their accent, with 54% somewhat, moderately, or really liking their accent. A smaller proportion were negative about their accent, with 14% somewhat, moderately, or really disliking their accent. These figures suggest that high levels of accent awareness (80%, as reported in the previous section) may not necessarily correlate with “accent cringe” or linguistic insecurity, although this question did not give respondents the option to expand on their answers.

When divided into country of birth, slightly more Australian-born respondents were on the positive end of the scale about their accent (59%) as compared to the overall mean of 54% and the overseas-born respondents (47%). Slightly more (19%) of the overseas-born somewhat, moderately, or really disliked their accents as compared to the Australian-born (>11%) and the overall mean (14%). The results for “neutral” did not differ greatly across the three measures. The greatest difference between responses was 11% (18% overseas versus 29% Australian-born) presenting in the “moderately like” category. Due to the low numbers in some of the cells, the responses were binned into “dislike”, “neutral”, and “like” and subject to a chi-square test, finding that the responses of the overseas-born group were significantly different from the responses of the Australia-born group X²(2, n = 392) = 7.19, p < 0.03.

3.2.3. Do You Ever Try to Change Your Accent?

For the respondents who had indicated that they felt they spoke English with an accent in Q6, they were also redirected to Q10: “Do you ever try to change your accent when speaking English?” (n = 480;

Table 3. Responses to the question “Do you ever try to change your accent?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Never	Rarely	Occasionally	Frequently
Overall (n = 480)	25%	29%	41%	5%
Australia (n = 251)	25%	29%	41%	4%
Overseas (n = 141)	27%	24%	39%	10%

). The rationale for this question path was that respondents would only report changing their accent if they believed they had one. However, it is possible that those reporting no accent could have reported changing their accent here, particularly because the results later revealed that “no accent” for many respondents actually meant having an Australian accent (see Section 3.2.1).

When broken down by place of birth, the results were practically identical for the Australian-born group as compared to the overall mean. The results for the overseas-born respondents were somewhat different, however, with an increase of 6% as compared to the Australian-born group when it came to “frequently” changing one’s accent. There was also a notable 5% decrease in the “rarely” response category. However, the responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(3, n = 392) = 5.61, p = 0.13).

When asked to describe the ways in which they changed their accent in Q11, there were reports of linguistic accommodation, coupled with a sense that this accommodation may not be a reflection of one’s true, authentic accent, e.g.,

I speak with a broader accent when I am speaking with people with broader accents. I try to fit in with them. It’s only when I really relax that I actually speak the way I did when I was growing up.

(Participant #145, born in Australia)

Overseas-born respondents also reported accommodation, but more often to make themselves understood:

I work with tradies, it is too much fun putting on an accent but very challenging. Need to do it often to make myself understood.

(Participant #83, born in Peru)

A minority of overseas-born respondents reported actively trying to change their accents to make them more Australian:

Hearing what words I’m saying other people are struggling with, and slowly trying to relearn them in an Australian way. There are also Youtube videos of people teaching the nuances of Strayan so I’m working on that as well.

(Participant #169, born in Israel)

3.2.4. Types of Accents and Wanting an Australian Accent

Respondents who had indicated that they spoke English with an accent in Q6 were redirected to Q7 asking “How would you describe the accent you have when you speak English?” with four options for responses. A total of 485 respondents (73.4%) gave an answer to this question, with the following breakdown: “Australian accent” (58%), “accent influenced by another language” (16%), “other English-speaking accent” (11%), and “mixture of accents” (15%). Respondents were invited to specify for the second, third, and fourth options. For “accent influenced by another language”, responses were diverse, and included Russian, English, and Croatian. For “other English-speaking accents”, responses included Indian, Malaysian, Irish, and American. For “mixture of accents”, responses included the following:

I don’t have a fully Australian accent perhaps due to my Mum’s Lithuanian accent.

(Participant #54, born in Australia)

I grew up in the northeastern [sic] US, so that is my “base” accent. I have lived in Australia for a decade, though, so it is heavily inflected with an Australian twang. (My American family and friends all note it now.)

(Participant #147, born in USA)

Respondents who indicated that they spoke with an accent other than an Australian one in Q7 (i.e., picked option 7b, c, or d) were redirected to a further question about the Australian accent, Q12, which asked “Earlier, you responded that you do not speak English with an Australian accent. Would you like to have an Australian accent?” (n = 191). The majority of these respondents were born overseas (64%), followed by Australian-born (23%), and those who did not specify their place of birth (13%), and their results are presented in

Table 4. Responses to the question “Earlier, you responded that you do not speak English with an Australian accent. Would you like to have an Australian accent?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Yes	No, Happy with Own Accent	Be Able to Switch	Mixture	Completely Different	No Accent
Overall (n = 191)	10%	50%	20%	8%	3%	9%
Australia (n = 43)	9%	53%	21%	5%	7%	5%
Overseas (n = 123)	12%	46%	20%	8%	2%	11%

(see Appendix A for the full wording of the question and responses). It was of note that 43 respondents born in Australia did not consider themselves to have an Australian accent, which shows that AusE is not something necessarily acquired by being born in the country.

The breakdown of responses shows that even among this cohort of respondents who did not feel that they had an Australian accent, half were happy with their accent the way it was and just 10% would like an Australian accent. These responses corroborate the finding presented in previous sections, which is that the sample had relatively high levels of linguistic security, although respondents were not given the option to expand on their answers here, except for “completely different”. It is worth noting that a sizeable 20% said they would like both an Australian accent and their own, and to have the ability to switch between them, indicating an awareness of the desire and necessity to accommodate to interlocutors and code-switch depending on context and situation. A small proportion (3%) said they would like a completely different accent, including “received pronunciation” (Participant #315, born in England) or “more of an international English accent” (Participant #131, born in Australia).

The respondents who reported not having an Australian accent in Q7 (i.e., did not select option 7a) were redirected to Q13: “Do you ever try to adopt an Australian accent when speaking English?” (n = 303—note that there are higher numbers here as this redirect included respondents who had skipped Q7). Almost 65% of respondents stated that they rarely or never tried to do so. The remaining 35% stated that they occasionally (27%) or frequently (8%) tried to adopt the Australian accent. Respondents did not have the option to elaborate on their answers here.

Of those who answered “occasionally” or “frequently” to Q13 (n = 106), they were redirected to Q14 asking, “How successful have your attempts been to adopt an Australian accent?” Just 11% reported a lack of success, and the remaining 89% of respondents reported being somewhat (37%), moderately (26%), or very successful (26%) at adopting the Australian accent. Again, of those who had answered “occasionally” or “frequently” to Q13, they were redirected to Q15 where they were invited to “explain your success or lack of success”. Here, they provided explanations such as the strong interference of their first language and going to extended efforts to change their accent. Others, as with Participant #135, found it hard to consistently produce AusE when under pressure to perform:

It takes too much conscious effort to keep thinking about saying words in an Australian accent and I forget the Australian way to say words when I am on the spot.

(Participant #135, born in New Zealand)

3.3. Likert Scale Ratings of the Australian Accent

3.3.1. Educatedness

After rating their own accents, respondents were asked to rate what they thought about the Australian accent in general along six different traits using a seven-point Likert scale. The scores for 500 respondents on the trait of educatedness are presented in

Table 5. Responses to the question “To what extent do you think the Australian accent sounds educated?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Uneducated	Moderately Uneducated	Somewhat Uneducated	Neutral	Somewhat Educated	Moderately Educated	Really Educated
Overall (n = 500)	2%	6%	19%	48%	9%	13%	3%
Australia (n = 315)	2%	6%	20%	50%	7%	13%	2%
Overseas (n = 161)	3%	7%	17%	45%	12%	12%	4%

. Most responses were “neutral”, at almost 50% of the sample, noting that, throughout this section of the survey, the seven-point scale offered the potential drawback of a middle ground, which may have influenced respondents’ answers. For the overseas group (n = 161), the trend was almost identical, except for “neutral”, rated at 45%. These percentages suggest that, overall, AusE is neither considered to sound particularly educated, nor particularly uneducated. However, the overseas-born respondents were more likely to rate AusE as sounding somewhat, moderately, or really educated, at 28% of the sample, as compared to the Australian-born respondents, at just 22%. However, the responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 476) = 5.92, p = 0.43). Since there were low numbers in some of the cells, responses were binned into “uneducated”, “neutral”, and “educated”; however, these differences remained insignificant at X²(2, n = 476) = 2.15, p = 0.34).

The comments provided by respondents to justify their ratings offer further insight into the preference for “neutral” for this trait. Several respondents seemed to select “neutral” to signal “it depends”, responding that there was no one Australian accent, and that people’s degree of educatedness depended on their class, upbringing, profession, etc. A minority indicated discomfort with having to rate AusE in this way, stating the following:

Again, I am not sure what you want here. You are asking me about stereotypes and I do not have such stereotypical thinking.

(Participant #219, born in Hungary)

The next most selected response after neutral was “somewhat uneducated” at 17–20%. One participant justified their choice of this rating by comparing AusE with “posh English accents” but also indicated a certain discomfort in having this view:

I feel bad saying it sounds uneducated… I definitely don’t think that an accent says anything about how intelligent or educated someone is! But I guess I’ve grown up with this impression that posh English accents signify being privately/well educated and everything else as not (not that I have a posh English accent by the way!!).

(Participant #434, born in New Zealand)

Another participant rated AusE as “somewhat uneducated” by comparing it to the two “gold standards” in the English-speaking world, British and American English. However, this view came from a participant born in the Philippines, where there is a lot of exposure to American English:

It doesn’t sound as polished as the British accent and Australians don’t sound as articulate as American. But I believe this is mainly due to the media’s influence.

(Participant #233, born in Philippines)

An Australian-born participant (#530), rating AusE as “somewhat uneducated”, commented on a “pronounced” Australian accent (which may be referring to broadness) as indexing a yobbo. This has been described as a term that predated the term bogan, although it is, perhaps, even more strongly linked to “lower SES, social disruption and being ‘rough’” (Penry Williams 2019, p. 186), as well as lack of education:

A pronounced Australian accent often sounds like a yobbo, an uncultured person, and that goes along with an assumption of a lack of education.

(Participant #530, born in Australia)

3.3.2. Professionalism

Professionalism was rated similarly to educatedness, and the results are presented in

Table 6. Responses to the question “To what extent do you think the Australian accent sounds professional?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Unprofessional	Moderately Unprofessional	Somewhat Unprofessional	Neutral	Somewhat Professional	Moderately Professional	Really Professional
Overall (n = 502)	2%	6%	17%	46%	13%	14%	4%
Australia (n = 315)	1%	6%	17%	47%	11%	14%	4%
Overseas (n = 160)	2%	6%	15%	42%	16%	14%	4%

. Ratings were similar across Australian and overseas-born respondents, with the greatest difference presenting in the “somewhat professional” category, which was selected by 11% of Australian-born but 16% of overseas-born respondents. The responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 475) = 4.2, p = 0.65). Differences between responses remained insignificant when data were binned into “unprofessional”, “neutral”, and “professional” X²(2, N = 475) = 1.66, p = 0.44).

Overall, as with educatedness, most responses were centered on “neutral”. The overseas-born had 5% fewer responses in this category as compared to the Australian-born. The “neutral” response seemed to, again, be used to indicate “it depends”, or sometimes “it doesn’t matter”, e.g.”

I dont [sic] think of accents in that way. Any accent can sound professional and depends on what is being spoken about and to who.

(Participant #68, born in India)

Another respondent rated AusE as “somewhat unprofessional” but then reflected on their own biases in a follow-up comment:

Well, there is a working class twang to it, but I like that! So… to British ears for example, it might sound uncouth. But what the hell! We have to uncouple ideas of intellect being represented by a posh accent.

(Participant #285, born in Australia)

There were also some extreme views, with Participant #57, for example, who rated the Australian accent as “really unprofessional” and claimed that this may be connected to the fact that AusE is very informal, including in its word choice (this is, perhaps, an indirect reference to hypocorisms; see Kidd et al. 2016):

This is also related to the fact that I find it sounds very unsophisticated and thereby unprofessional. This is due as well to word choice, the Australian language is often very informal which also does not help either.

(Participant #318, born in the Netherlands)

3.3.3. Attractiveness

A similar trend to educatedness and professionalism was observed for attractiveness (

Table 7. Responses to the question “To what extent do you think the Australian accent sounds attractive?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Unattractive	Moderately Unattractive	Somewhat Unattractive	Neutral	Somewhat Attractive	Moderately Attractive	Really Attractive
Overall (n = 503)	2%	6%	11%	45%	16%	15%	6%
Australia (n = 315)	2%	5%	12%	46%	17%	13%	5%
Overseas (n = 161)	4%	7%	7%	42%	15%	18%	7%

), with “neutral” ratings at around 45%, but slightly more positive attractiveness ratings overall, with a combined overall 37% of somewhat, moderately, or really attractive ratings. The Australian-born were somewhat less likely to rate the accent as attractive (total 35% for these three ratings) versus the overseas cohort at a combined 40%. However, the responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 476) = 8.65, p = 0.19). Differences between responses remained insignificant when data were binned into “unattractive”, “neutral”, and “attractive” X²(2, n = 476) = 1.22, p = 0.54).

Some participants rated the Australian accent as “really attractive” because it reminded them of home, indicating that it is not so much the accent itself, but the memories and experiences it invokes:

It sounds like home. When you arrive at the airport from overseas and you hear that accent, you can’t help smiling.

(Participant #549, born in Australia)

Some overseas-born participants rating AusE high for attractiveness described it as “cute”, “sophisticated”, “sexy”, and “there’s just something about it”. Others expressed an affinity for any kind of accent that expresses a sense of place:

I love accents in general (having my own “regional” accent variety of Hebrew), and Aussie sounds great.

(Participant #169, born in Israel)

3.3.4. Friendliness

The scores for 502 respondents on the trait of friendliness are presented in

Table 8. Responses to the question “To what extent do you think the Australian accent sounds friendly?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Unfriendly	Moderately Unfriendly	Somewhat Unfriendly	Neutral	Somewhat Friendly	Moderately Friendly	Really Friendly
Overall (n = 502)	0%	1%	2%	21%	18%	29%	29%
Australia (n = 315)	0%	1%	1%	19%	19%	31%	29%
Overseas (n = 160)	1%	1%	3%	22%	14%	28%	31%

. These scores present quite a different picture from the ratings for educatedness and professionalism, with most responses skewing towards the positive, with over 20% at neutral and just 3% overall rating the accent as somewhat or moderately unfriendly. Just one respondent (in the overseas cohort) rated the accent as really unfriendly. However, there were fewer respondents in the overseas cohort rating the accent as somewhat, moderately, or really friendly (total 73%) as compared to the Australian-born cohort (79%) and the overall cohort (76%). The responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 476) = 4.94, p = 0.55). However, differences between responses approached near significance when data were binned into “unfriendly”, “neutral”, and “friendly” X²(2, n = 476) = 4.83, p = 0.09). Nonetheless, these findings indicate that, overall, the Australian accent is viewed as quite friendly.

The ratings for friendliness support the idea that AusE is informal and, therefore, makes people sound approachable, with respondents justifying their choices by stating the following:

Because of our irreverence and our love of shortening words.

(Participant #549, born in Australia)

Participant #549 also refers to hypocorisms here, and there were other comments to the effect that Australians either “elongate” or “abbreviate” their words and sounds, making them appear more friendly. Other respondents relied on their views of Australians as a general population to make this judgement about AusE, stating the following:

I think this again reflects that I generally think Australians are friendly.

(Participant #294, born in Australia)

3.3.5. Likeability

A similar overall trend emerged for likeability as did for friendliness (

Table 9. Responses to the question “To what extent do you like the Australian accent?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Dislike	Moderately Dislike	Somewhat Dislike	Neutral	Somewhat Like	Moderately Like	Really Like
Overall (n = 506)	1%	2%	6%	23%	18%	28%	22%
Australia (n = 315)	1%	2%	6%	21%	18%	29%	22%
Overseas (n = 161)	2%	2%	6%	25%	15%	29%	21%

). Neutral responses ranged from 21–25%, with the remainder of responses on the positive end of the scale, with 68% of all respondents somewhat, moderately, or really liking AusE.

There were some small differences between the groups of respondents. The overseas group had fewer positive responses overall (65% versus 69% for Australian-born) and 4% more neutral responses than the Australian-born group. This mirrors the patterns for friendliness, where Australian-born respondents were rating AusE higher than overseas-born respondents. However, the responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 476) = 2.4, p = 0.88). Differences between responses remained insignificant when data were binned into “dislike”, “neutral”, and “like” X²(2, n = 476) = 1.38, p = 0.5).

Overseas-born respondents rating AusE low for likeability described it as “unsophisticated”, “coarse”, “ugly”, and “bogan”. However, others described acquiring a taste for it, calling it “earthy and appealing”. Some stated they had lowered their rating because they liked some Australian accents more than others and others explained they had rated it “neutral” because there was not enough nuance to the question:

Which one? The Crocodile Dundee one or the accents of my work colleagues? I might like some Au [sic] accents and dislike others; don’t have enough information to answer this question.

(Participant #229, born in Croatia)

3.3.6. Clarity

The scores for 507 respondents on the trait of clarity are presented in

Table 10. Responses to the question “To what extent do you think the Australian accent sounds clear?”, broken down by place of birth (Australia or overseas). Figures rounded to nearest whole number and may not add up to 100.

Place of Birth	Really Unclear	Moderately Unclear	Somewhat Unclear	Neutral	Somewhat Clear	Moderately Clear	Really Clear
Overall (n = 507)	2%	8%	19%	20%	16%	24%	10%
Australia (n = 315)	2%	8%	18%	20%	15%	24%	12%
Overseas (n = 161)	2%	9%	22%	19%	17%	26%	4%

. This presents a different pattern from the other five traits. Here, there is a more even spread across all responses, particularly in the middle five ratings from “moderately unclear” to “moderately clear”. This suggests some disagreement and lack of consistency across respondents as to whether AusE is clear or not, although overall, 50% of respondents rated AusE as somewhat, moderately, or really clear.

The ratings in the overseas-born group were fairly similar to those of the Australian-born, with one difference of note being a substantial 8% lower rating for “really clear” at 4% of respondents versus 12% for Australian-born. There was also a 4% difference in the “somewhat unclear” rating (18% for Australian-born versus 22% for overseas). This suggests, unsurprisingly, that some of the overseas-born group found AusE to be somewhat less clear than the Australian-born respondents. However, the responses of the overseas-born group were not found to be significantly different from the responses of the Australia-born group X²(6, n = 476) = 8.86, p = 0.18). Differences between responses remained insignificant when data were binned into “unclear”, “neutral”, and “clear” X²(2, n = 476) = 1.43, p = 0.49).

Taking a closer look at respondents’ justifications for their clarity ratings, several overseas-born respondents stated that Australians “eat their words”, “swallow sounds”, and have “ambiguous vowels” and a “strong nasal tone”. Others stated that AusE takes getting used to, as it was not a variety they were typically exposed to before moving to Australia. The views and justifications of Australian-born participants with more negative ratings were not that different, with many stating that they had difficulty making themselves understood to non-Australians, and that they “run their words together” and “mumble”. Australian-born respondents who rated the accent as “really clear” gave understandable justifications such as the following:

I have no problems understanding Australian accents as I’ve heard it since I was a child.

(Participant #121, born in Australia)

Others acknowledged that AusE was likely difficult for non-Australians to understand:

I think we tend to mumble and to sort of swallow our vowels sometimes (as someone who comes from Melbn)- this seems to be hard for people whose first language is not English.

(Participant #16, born in Australia)

3.4. Correlations of Likert Scale Ratings

In order to test whether respondents’ ratings for certain traits were correlated, a series of Pearson product-moment correlations were run (

Table 11. Pearson’s product-moment correlation matrix of Likert scale ratings (renumbered 1–7 from negative to positive ratings). Figures show r correlation coefficients rounded to two decimal places. Correlations run on 495 ratings from each of the six traits (2970 total ratings) to run a paired comparison. A total of fifty ratings were removed from respondents who did not provide ratings for all six traits.

	Clarity	Educatedness	Professionalism	Likeability	Attractiveness	Friendliness
Clarity	--
Educatedness	0.44	--
Professionalism	0.48	0.73	--
Likeability	0.35	0.41	0.47	--
Attractiveness	0.34	0.53	0.55	0.62	--
Friendliness	0.28	0.31	0.34	0.48	0.42	--

). These showed no strong correlations between any of the trait ratings.5 However, there was a moderate positive linear correlation between ratings for professionalism and educatedness (r = 0.73), as well as for professionalism and attractiveness (r = 0.55), and attractiveness and educatedness (r = 0.53). Furthermore, there were moderate positive correlations between likeability and attractiveness (r = 0.62), educatedness (r = 0.41), and professionalism (r = 0.47). Clarity correlated moderately only with educatedness (r = 0.44) and professionalism (r = 0.48). Friendliness correlated moderately only with likeability (r = 0.48) and attractiveness (r = 0.42).

4. Discussion

4.1. Attitudes and Ideologies towards AusE

This paper presents a selection of findings from an online survey with 661 respondents residing in Australia, of which approximately one third was born overseas. The first research question concerned the attitudes and ideologies towards AusE held among these two groups of respondents. The findings for the question “Do you speak English with an accent?” showed 80% of respondents reported having an accent to some degree, and that the overseas-born respondents had significantly different responses to the Australian-born, likely due to their higher “mostly” and “definitely” ratings. The figure of 80% was higher than expected, as often, speakers with a mainstream or “standard” accent view themselves as having no accent at all—the category of “accent” being ascribed only to the Other (see Lippi-Green 1997). The sample as a whole seemed to be quite sociolinguistically aware, with many respondents acknowledging or conceding that the Australian accent also counts as an accent. Those who reported little or no accent had reasons that centered on seven different themes, including the fact that they were “native speakers”, were “monolingual”, or had “lost” their accent after living in Australia for a long time. The fact that the majority felt that they had an accent could be related to the widely reported “cultural cringe” (see e.g., Willoughby 2020) and feelings of inferiority vis-à-vis more prestigious varieties of English, such as American English (Bayard et al. 2001).

However, the findings for linguistic (in)security in response to the question “How much do you like your accent?” showed that over half of the respondents were positive about their own accent, and almost 35% had neutral feelings about it. This suggests fairly high levels of linguistic security, coupled with high language awareness, meaning that the respondents were comfortable with their accents but not to the point where they were in the privileged position of not noticing it anymore. It may be the case that a standard language ideology is more prevalent in places where speakers already speak a globally prestigious variety (see e.g., Lippi-Green 1997 for American English and Preston 1999 on north/mid-Western varieties of American English).

Among the overseas-born group, there was a significant decrease in positive ratings for their own accent, indicating that linguistic insecurity was somewhat higher for this group. In a further question about changing one’s accent, 10% of the overseas-born group reported “frequently” changing their accent, either to accommodate but also to make themselves understood. Some comments suggested that overseas-born respondents liked AusE and would like to emulate it, but that it was challenging to do so. Nonetheless, over 89 respondents reported some amount of success in adopting AusE. However, in a follow-up question, only 11% said that they would like to have an Australian accent, and it was more common to want to have a number of interchangeable accents (20%), or an accent “mixture” (8%). This supports work by Starks and Willoughby (2015) and Eisenchlas and Tsurutani (2011) showing an increasing acceptance of non-Australian-born people to either adopt AusE, switch between AusE and their native accent, or simply be accepted with their native accent. Furthermore, there were few reports of the kind of linguistic racism reported in Dovchin (2019). This suggests that there was a certain ambient acceptance of linguistic diversity in the sample, which is perhaps not surprising considering how diverse Australian society has become.

4.2. Likert Scale Ratings of AusE along Six Traits

The second research question concerned how Australian- and overseas-born participants rated AusE along six different traits, following a well-established methodology in the matched guise literature (Lambert et al. 1960). It is worth noting from the outset that some of the respondents had negative reactions to being asked about an accent in such a general manner and gave feedback in their responses that they felt uncomfortable drawing on stereotypes or speaking about accents and varieties as if they were monolithic entities. Indeed, the matched guise method tends to give respondents voices to listen to, followed by ratings, and in hindsight, this may be a better way to elicit these kinds of ratings, as respondents then do not feel as though they are being asked to rate a variety as a whole (see e.g., Bayard et al. 2001 and Bradley and Bradley 2001 for matched guise studies on AusE). Nonetheless, participants responded well to these questions, with the vast majority choosing to respond to all six questions, even though they were not obliged to. They were also forthcoming in providing follow-up explanations for their ratings. Interestingly, there were no significant differences between the Australian and overseas-born respondents for any of the Likert scale ratings (although near significance for friendliness), suggesting an alignment between these populations when it comes to rating the Australian accent in general (differing to e.g., how they rate their own accents).

The findings show that AusE overall was not rated particularly high for educatedness. Instead, most respondents erred towards “neutral” in this case, although it is not clear whether they simply chose this response as a default choice. The follow-up comments for these elucidated important insights, as respondents explained that they gave a neutral rating because it depended on which Australian accent they were being asked to rate. This supports findings by Bradley and Bradley (2001) who found that Australians were only more likely to give AusE high ratings if it was general AusE rather than broad AusE. Others stated that the reason for their neutral rating was that some Australian accents sounded more educated than others, so this was the “average” across those different accents. Participants who rated AusE low for educatedness often drew on the “bogan” or “yobbo” stereotype in their follow-up comments, stating that the Australian accent was synonymous with a certain kind of social type that is known for not being educated. This social type has been well-reported in the literature (e.g., Penry Williams 2019); it is synonymous with broad AusE, and tended to emerge somewhat more among comments of the Australian-born cohort. However, this view may be changing over time, with one participant claiming that perceptions of intellect and “posh accents” need to be “uncoupled” and that a working class (read: broad) Australian accent could also be considered educated. The fact that the overseas cohort did not mention “bogans” or “yobbos” (although one mentioned “tradies” or tradespeople as representative of a broad/working class variety) suggests that certain kinds of stereotypes and indexicalities may not be as accessible to overseas-born people in Australia, meaning they have something of a “fresh start” when making up their mind about AusE. It could also be that since AusE is a “native” or L1 variety of English, it automatically appears more educated to L2 ears.

A trait related to educatedness is that of professionalism, and in the present study, ratings for these two traits were similar and correlated. However, it is of note that the ratings for professionalism were somewhat higher, although Australians’ penchant for using hypocorisms and generally using informal language emerged as reasons to not rate AusE on the high end of the scale for this trait. This differs somewhat from the rating for attractiveness, which was rated more highly. Indeed, AusE was viewed as attractive by both Australian and overseas-born participants, with some citing feelings of fondness for home and belonging when hearing the accent. This is to be expected in ratings of one’s own accent, particularly when that accent/variety is not the globally prestigious one (Cargile et al. 1994). In other words, the ratings for attractiveness may be linked to a certain covert prestige for AusE. This links with the findings for likeability, which also showed somewhat higher ratings than for educatedness. Interestingly, features such as “shortening words” and national attributes such as “irreverence” tended to influence respondents in opposite ways: for some, it made them like the accent more; for others, it made them like it less.

The findings for ratings of clarity of AusE showed that there was a range of views, although 50% of respondents rated is as somewhat, moderately, or really clear. Understandably, some overseas-born respondents rated the accent as less clear than Australian-born respondents, but their responses were not significantly different to those of the Australian-born. The reason for this is that the Australian-born population appeared to be quite apologetic about the “mumbly” character of AusE, reporting that they had difficulty making themselves understood amongst foreigners and that they tended to “swallow their words”. This could be linked to the historical “cultural cringe” in Australia, which is well described in Bradley and Bradley (2001), where there has been a long tradition of prescriptivism (influenced by a view that British English is the model), along with a sense of shame about its “perverse” character. For the clarity trait and throughout, respondents did mention “broad” AusE in comparison to less marked sub-varieties, suggesting that the broadness continuum (see Cox 2012) continues to be a point of reference for the topic of accent in Australia.

A number of moderate (but not strong) correlations were found between the ratings, which reflect, but do not perfectly align with, the standard dimensions of status and solidarity in language attitude studies (Cargile et al. 1994). Unsurprisingly, there were correlations between (a) professionalism and educatedness (status); and (b) friendliness, likeability, and attractiveness (solidarity). It is of note, however, that attractiveness was also moderately correlated with professionalism and educatedness, as well as with likeability. This suggests that all four of these traits are linked in people’s minds. Since the professional and educational domains tend to be the mainstay of standard language, it appears as though standardness has an attractive and likeable quality in AusE (linking to Bradley and Bradley 2001 on a preference for general AusE). However, when it came to clarity, this trait was only moderately correlated with educatedness and professionalism. This is supported by respondents’ apparently mixed feelings surrounding the clarity of AusE, with many acknowledging that the accent may not be as intelligible to non-AusE speakers. These findings suggest that a respondent was able to rate AusE low for clarity but higher for educatedness and professionalism, without necessarily seeing a contradiction between the two. This reflects the somewhat conflicted attitude emerging among the Australian-born in the sample of a growing acceptance coupled with an underlying insecurity around AusE, and, in particular, broad AusE and its associated negative stereotypes.

5. Conclusions and Future Directions

This present study aimed to contribute to the language attitudes literature on AusE and, in particular, attempted to fill a gap by including the voices of overseas-born people in Australia, who now form a significant part of the overall population at 27.6% (ABS 2022) and were well represented at 34% of our sample. In future research, we will analyze the remaining questions in the survey, which include questions about respondents’ children and gender-based variation in AusE, which were not covered in the present study due to time and space limitations. We will also consider the variables of respondent gender, age, and length of residence in Australia in these further analyses. In doing so, we hope to provide a fuller picture of how attitudes and ideologies exist in the contemporary linguistic ecology of Australia.