The Socio-linguistic Profi les, Identities, and Educational Needs of Greek Heritage Language Speakers in Chicago

The present study aims to further the research on heritage language speakers (HLSs) by providing the socio-linguistic profi les and identities of an uninvestigated community of heritage speakers, namely the Greeks of Chicago, thus offering data for a less-studied HL, Greek. The participants were fi fty-four (N=54) fi rst, second, and third-generation Greek HLSs. The socio-linguistic data were collected through an online survey, while identifi cation with Greek culture as well as ethnic attachment and practice of Greek traditions were investigated through the content analysis of data from the Greek Heritage Language Corpus. The results of the study are discussed with respect to how they can improve our knowledge of the educational needs of Greek HL learners. This research-based knowledge can be employed for addressing the academic needs of HL learners through educational programs. The authors propose an agenda for a more linguistically and culturally responsive education program for HL learners, in general, and Greek HL learners in diasporic communities, in particular.


Introduction
Classrooms around the world are increasingly characterized by the presence of heritage speakers. Heritage languages (HLs) are spoken by simultaneous or sequential early bilinguals, commonly referred to as heritage speakers (HSs), whose one shared characteristic is a restricted knowledge of a home language mainly because of insufficient linguistic input. As a result, HSs can understand the language of the home and probably speak it to some degree but are more proficient in the dominant language of their host country (Polinsky, 2011). Previous research has shown that HLs are often related with identities and that HSs' identities are negotiated constantly, shaped by multiple variables, and are dynamic and socially constructed (Berard, 2005;Block, 2007;Valdés, 2001).
As far as Greek HLSs (GHLSs) are concerned, even though there are a considerable number of them in the U.S., Australia, and Canada, they are still in uncharted territory. Only recent research (Aravossitas 2016;Gardner-Chloros, 2009;Gavriilidou & Mitits, 2019;Karatsareas, 2018Karatsareas, , 2019 has documented the linguistic competency and sociolinguistic characteristics of Greek or Cypriot-Greek HSs, so there is an urgent need to collect more data from GHLSs around the world in order to delineate their characteristics, investigate their identities, and address specific educational requirements stemming from their needs. Otherwise, as research on other HLs has shown, heritage Greek is faced with a loss of language contact and a pending language death with every subsequent generation of HLSs. Thus, the purpose of the present paper is to extend previous research on HLs by offering data for a less-studied HL, Greek, and a HSs' community that has not been investigated yet -GHLSs of Chicago, in order to cast light on predictable (but variable) sociolinguistic profiles of GHSs. Based on previous research highlighting various parameters closely related to HSs' socioliolonguistic profiles, our specific aim is to investigate their age of arrival and possible contact with Greece, language preferences in various contexts, the formal instruction they might have received, exposure to Greek, self-assessed proficiency, attitudes toward Greek, and identity characteristics in order to discuss how these may connect with and what they imply about their educational needs when learning Greek as an HL.
generation HSs often undergo the demanding process of choosing an identity, which may result in the rejection of the HL and culture, more rarely in the rejection of the majority language and culture, or finally in a bilingualbicultural identity (Lambert, 1975). In third-generation HSs, language attrition is predominant (Carreira & Kagan, 2011) and quite often accompanied by a monocultural identity. Furthermore, previous research has shown that that younger learners are more at risk to lose proficiency in their HL (Carreira & Kagan, 2011).
On the other hand, there is no previous research investigating the direct effect of contact with the HSs' home country. It may include travel to visit relatives, study abroad programs, volunteering, or internships, and could be classified as what is usually called 'heritage tourism'. Actually, visiting a country where the HL is spoken has many benefits. In addition to improving their language skills, HSs who travel to countries where their HL is spoken gain self-confidence, deepen their cultural knowledge, and strengthen their heritage identity. It seems that travel to heritage countries, with or without formal study, is one of the best ways to boost language skills in an HL. But this hypothesis needs to be supported with empirical data.

Language Preference and Use in Different Situational Contexts
The study of language preference and use of an HL in different situational contexts allows for predictions about the linguistic ability of the different generations of HSs and that was the main reason for incorporating this variable in the present research. It is evident that the majority of second and third-generation HSs in the U.S. are, if not English dominant, English preferent. Some of them may continue to function in two languages, especially for communicating with family or community members of the first generation. Valdés (1999) noticed that HL use is restricted to everyday discourse; it is narrower in range and is a low variety, appropriate mainly for casual, informal everyday interactions. Other researchers (Chevalier, 2004;Shin, 2013) pointed out that the restricted use of an HL in informal settings in home and local communities is often the result of monolingual educational policies, which also lead to monocultural identities. These restricted situational contexts in which HLs are used, compared to the use in home countries, lead to a number of idiosyncratic changes and attrition in all linguistic levels of the HL. This attrition results in HSs transferring their HL "in a 'mutilated' form to the next generation" (Valdés, 1999: 10). Functionally, though, HSs may be able to carry out conversations on everyday topics with ease and a relevant fluency, and may even be able to understand spoken language that includes the use of humor.

Formal Instruction in an HL
The correlation between the years of formal instruction and proficiency in an HL has been duly acknowledged in previous research and it was decided to further investigate this variable in the present study as more years of instruction may ensure a higher level of proficiency (Kagan, 2005;Kagan & Dillon, 2008). Some studies have focused on the effect of specific instruction programs (Potowski et al., 2009) on HSs' proficiency in an HL. What is important to note here is that instruction in an HL aims not only to enhance HSs' linguistic ability but also to achieve community advancement, since it addresses broader academic and identity challenges and may strengthen cultural ties between HSs and their home and community. Furthermore, it cultivates a positive attitude toward the HL as it can counteract the hegemony of the majority language (Martinez, 2016).
Various types of programs for HL teaching have emerged in recent years: school based-programs in primary and secondary schools, higher-education programs in private and public universities, and finally communitybased programs which offer after-school, Saturday-school, or summer school instruction that is hosted in museums, cultural, religious, or community centers (Beaudrie, 2016).
Actually, Greek Orthodox churches in the U.S. and more specifically in Chicago started establishing parochial schools in the 1900s, some of which are full-day schools with a bilingual English and Greek curriculum, while others are afternoon and Saturday schools with only a Greek-language curriculum. According to data reported in Michopoulos (2008), the total enrollment of Greek ancestry pupils was 6,000 pupils in day schools and 30,000 pupils in afternoon or Saturday Schools. While the vast majority of Greek children attend Chicago public schools (except for those enrolled in Greek day schools), practically all Greek children attend afternoon (following public school attendance) and Saturday schools, where they learn the rudiments of the Greek Orthodox faith along with Greek language and culture (Kunkelman, 1990). In these schools, teachers are usually volunteers, with no previous teacher training, teaching methods are often outdated, and traditional and pedagogic material is often obsolete; as a result, these schools are not always attractive to young Greek HSs. However, since the U.S. educational system does not prioritize bilingualism in different PROFILES, IDENTIFIES AND NEEDS OF GREEK HERITAGE SPEAKERS heritage languages, specifically in Greek, such programs fill this void and play a crucial role in Greek language and culture maintenance (Nikolidakis, 2005). It should be noted that there are also numerous cultural organizations and unions promoting Greek cultural heritage and traditions. Thus, such programs in combination with the education received by HSs in mainstream schools help them acquire a unique bicultural identity and a sense of belonging to two cultures, and also assign equal status to the two cultures and languages.

Exposure to and the Use of an HL
Exposure is a predictor of possible bilingual or HL development, but we may also add maintenance, as reported in previous studies (De Houwer, 2007;Dixon et al. 2012), which is why it was deemed necessary to include this parameter in the present study. The length and manner of HSs' exposure to an HL may vary according to socioeconomic parameters and this variation can have a great impact on mastering it. Research has shown that HSs do not have much exposure to their HL outside the home (Carreira & Kagan, 2011) and that parental involvement can be crucial for HL and culture maintenance. For instance, Park and Sarkar (2007) found that one way that parents helped their children maintain Korean as an HL in Montreal was through encouragement of the internet use in Korean. Prevoo et al. (2014) reported that reading to children in the HL may boost children's' vocabulary in that language and expose them to a broader range of concepts and perspectives. Furthermore, it is the parents' wish to expose their children to the family HL that motivates them to enroll their children in community-based programs for HL instruction (Liu, 2013). It should be noted that activities such as reading literature in the HL, navigating the internet, watching TV, reading newspapers, etc., may help HSs maintain contact with their HL and culture.

Language Proficiency in an HL
Proficiency in an HL can be defined narrowly or broadly. It can refer to reading, writing, speaking, and listening skills in an HL or to basic linguistic abilities such as vocabulary or grammar (Zyzik, 2016). Some scholars claim that proficiency in an HL does not necessarily mean knowledge of the culture carried by the language and viceversa (Hoffman, 1991). Others believe that "the further one progresses in bilingual ability, the more important the bicultural element becomes, since higher proficiency increases the expectancy rate of sensitivity towards the cultural implications of language use" (Beardsmore, 1986: 24). Different research protocols have been used to assess the language proficiency of HSs such as self-report assessments (Kagan & Friedman, 2003), vocabulary knowledge (Polinsky, 2008a), speech rate (Polinsky, 2008b), cloze-tests , picture naming , etc. (for the proficiency assessment of HSs, see Ilieva & Clark, 2016). Regardless of the instrument used to assess proficiency, the majority of studies have concluded that HSs demonstrate strong oral skills, while other skills lag behind (Ilieva & Clark, 2016). Valdés (1999: 11) observed that HSs "may seem quite superior in some respects and quite limited in others", e.g. HSs' reading and writing skills may be restricted or even nonexistent (Valdés, 2000). Similarly, Carreira and Kagan (2011) found that most HL learners who had acquired English after acquiring their HL could understand and speak the HL better than they could read and write it. Zyzik (2016) noted that HSs' language proficiency is limited to basic-level cognition which includes listening and speaking, and high frequency lexical items and frequent grammatical expressions commonly used in everyday interactions. In the same vein, Swender et al. (2014) in their study of Russian and Spanish HSs found that their sample demonstrated an inability to express abstract thinking and reasoning and lacked extended discourse and precise vocabulary.

Attitudes towards the HL
Research (Cho et al., 1997;Park & Sarkar, 2007) has shown that positive attitudes towards HLs are interrelated with the desire to learn an HL, successful HL learning, and/or different patterns in HL maintenance (Polinsky, 2018). It was, thus, deemed necessary to examine this variable. Research on socio-cultural aspects of HL learning has emphasized that attitudes toward an HL are not static but emerge from interactions and social practice (Zyzik, 2016). Furthermore, HSs tend to express a growing interest in maintaining their HL, leading to an increased enrollment in HL instruction classes (Lee & Kim, 2008;Yang, 2003). Carreira and Kagan (2011: 48) investigated HL learners' attitudes towards their HL and found that most of the students had "largely positive feelings and experiences" with their HL. Similar results were reported by Polinsky (2011), but also by Karatsareas (2018) who found a positive attitude among the members of the Greek-Cypriot Community in London towards their HL.

Heritage Speakers' Identities
Since language has a symbolic value, is a carrier of cultural capital, and determines one's conception of the world and one's relation to otherness and position in society, it is in an intimate relation with one's identity and is actually a strong indicator of a person's or group's identity (Lee, 2002). HSs connect with at least two linguistic communities and cultures and are in constant negotiation within a multicultural environment (Crawshaw et al., 2001). Whether they develop a dual linguistic and cultural identity or not depends on (a) how able they are to adopt hybrid-situated, non-conflicted identities in the frame of the dominant and heritage socio-cultural discourses (He, 2006) and on (b) how much the educational and linguistic policies adopted aim at strengthening dual identities or just simply promote monolingualism, thus weakening HL learner programs.
Thus, based on the above review of the literature and the need for investigations into the group of GHLSs, the following research questions were posed: RQ1. What is the socio-linguistic profile of GHLSs living in Chicago? RQ2. What are the GHLSs' defining elements of their Greek identity? RQ3. What are the educational needs of GHLSs based on RQ1 and RQ2?

Participants
The present study investigated fifty-four (N=54) GHLSs. The participants were included in the study by convenience sampling, after contacting the Greek community in Chicago. Informed consent was obtained from all participants. There were twenty-three (N=23, 42.6%) male and thirty-one (N=31, 57.4%) female participants whose ages ranged from <12 to 55+ (see Table 1), with all education levels from primary education to postgraduate degree holders represented (see Table 2). All procedures performed in the study involving human participants were in accordance with the ethical standards of the institutional ethics committee of Democritus University of Thrace (60589/2111/31-8-2018).

Instrumentation
The data collection instrument used to answer RQ1 was the 29-item Greek Heritage Language Background Questionnaire (GHLBQ), adapted to and validated in Greek, English, and Russian within the frame of the project Greek Heritage Language Varieties. Its design was based on a survey that investigated the reading skills of HL learners (Jensen & Llosa, 2007) and parts from other surveys available from the National Heritage Language Resource Center (Carreira, 2009;Gignoux, 2009;Lyutykh, 2012;Montrul, 2012;Torres, 2012). The design and validation procedure of GHLBQ followed the adaptation protocol described in Gavriilidou and Mitits (2016).

PROFILES, IDENTIFIES AND NEEDS OF GREEK HERITAGE SPEAKERS
The GHLBQ contains 29 questions in total. More specifically: (a) three questions (Q 1-3) on demographic information about the participants (age, gender, and the level of education), (b) six questions (Q 4-9) on biographical information related to the contact with the languages they use, (c) five questions (Q 10-14) on language use and preference, (d) one question (Q 15) on heritage language instruction, (e) seven questions (Q 16-22) on previous exposure to written/spoken heritage and dominant languages, (f) four questions (Q 23-26) on self-rated language proficiency, and (g) three questions (Q 27-29) on their attitude towards the languages they spoke.
The Cronbach's alpha coefficient for the 46 five-point Likert scale items grouped in questions 23-26 for assessing self-rated language proficiency was .969, suggesting an excellent degree of internal consistency. The survey was administered online to GHLSs in the spring of 2019.
To investigate RQ2, the authors relied on data from the Greek Heritage Language Corpus (GHLC), which is a speech corpus comprising 90 hours of recordings of speech from Greek HSs living in Moscow, St. Petersburg, and Chicago. More specifically, the study investigated the 85,000 words found in the Chicago sub-corpus. 1 Finally, answers to RQ3 were based on the sociolinguistic profile of the sample as well as on supplementary data from the Chicago sub-corpus of the GHLC.

Data Analysis
To investigate RQ1, SPSS v.24 Multiple Dichotomy Analysis was used for the multiple response sets and descriptive statistics were used in order to compute the frequencies of the responses containing nominal data from the GHLBQ. Content analysis was adopted for the analysis of the data obtained from the GHLC in order to answer RQ2 and RQ3.

Results
The results of the quantitative analysis for RQ1 are presented in the following section, while content analyses of the GHLC data for RQ2 and RQ3 are included in the discussion section.

Age of Arrival and Contact with Greece (Qs 4-9)
It was found that the majority of the participants (N=43, 79.6%) were born in the U.S., while 11 of them (20.4%) reported being born in Greece. Those participants who were not born in the host country arrived there at an early age <14, with the exception of two cases who arrived between the ages of 14 and 18. All of them had attended school in Greece before emigrating. However, the school years they completed varied (see Table 3). There were also cases of those who were born in the U.S. but then moved to Greece for a period of time before returning. As for their visits to Greece, only 3 out of 54 have never traveled there, while a large number of the respondents have visited Greece three to five times (n=26, 48.1%) and one-third (n= 17, 31.5%) go there at least once a year. Lastly, when it comes to the language that they learned to read in first, the participants report an almost even distribution among English, Greek, and both languages (n=16, 29.6%; n= 19, 35.2%; n=19, 35.2% respectively).

Language Preference and Use in Different Situational Contexts (Qs 10-14)
Questions 10-14 investigated language use and preferences in everyday communication. More precisely, these questions focus on determining which language(s) the participants prefer to use and with whom, as well as whether those preferences changed at different periods of their lives. The results showed (a) a decline in the use of Greek as HSs age and (b) an extended use of Greek only at home or with family.
When asked what language they spoke most of the time, only two (3.7%) opted for Greek while the largest number (29, 53.7%) chose English only, followed by both English and Greek (N=23, 42.6%). However, 70.3% of the participants reported using both English and Greek, and 16.7% only Greek when speaking to their family members as opposed to merely 13% who use English at home. When asked which language(s) they spoke with their friends, the participants reported the host country language first (n=39, 72.2%), the combination of dominant and HL second (n=13, 24.1%), and Greek (N=2, 3.7%) third.
The study produced very interesting findings with respect to language use preference changes over time (see Figure 1).

Figure 1
Language preferences There is a dramatic decline in the speakers' use of Greek as they age. While 74.1% (n=40) of respondents said they used Greek predominantly before the age of 5, only 3.7% (n=2) aged 18+ reported the same. At the same time, the number of those who use both languages on a daily basis seemed to increase by age before it drops after the age of 18.

Formal Instruction in Greek as an HL (Q 15)
The type and amount of formal instruction refers mainly to community-based instruction with the majority (N=39, 72.2%) of the participants receiving more than four years of instruction in Greek as an HL (see Table 4).

Exposure to and Use of Greek as a Heritage Language (Qs 16-22)
Next, the study investigated the exposure to Greek that the participants have had throughout their lives, mainly to the written language, which is generally underrepresented in the upbringing of HSs but also to certain tasks and activities that require the comprehension and use of spoken Greek.
Question 16, What type of Greek print did you have as you were growing up, revealed that the GHLSs grew up with a variety of Greek print media at home, with children's books (N=46, 85.2%), books (N=44, 81.5%), dictionaries (N=37, 68.5%), religious literature (N=36, 66.7%), and calendars (N=30, 55.6%) the most widely represented. They were followed by newspapers (N=27, 50%) and to a lesser percentage by flyers, magazines, digital media, and comics (between 35.2% and 24.1%). However, the above numbers drop when compared to what Greeklanguage print they have at their home at present (Q17), except for digital media, the use of which has risen from 27.8% to 48.1 % (see Figure 2).

Figure 2
Greek-language print As parental involvement plays an important role in HL development, the participants were asked to select what type of print they were read to as children by their family members. Their responses unveiled that those were mostly children's books (N=46, 85.2%), literature (N=29, 53.7%), religious literature (N=17, 31.5%), and comics (N=10, 18,5%).
Accessing the internet in Greek is another opportunity for the GHLSs to get additional language input. Although a general increase in the exposure to digital media in Greek was documented by the previous question, GHLSs do not appear to take advantage of this particular tool since only 10 of them (18.5%) use it frequently, whereas 35 (64.8%) and nine (16.7%) do so rarely or never respectively. In general, the time our participants spend reading non-school related material in Greek varies between not at all (n=5, 9.3%) and more than two hours (n=6, 11.1%) with the majority of them reading for less than 15 minutes (n=23, 42,6%). Table 5 shows, in descending order, which activities involving contact with Greek the participants have been engaged in recently. Their exposure to and use of Greek comes mostly from their oral communication in informal contexts, their contact with the popular cultural elements such as music and TV, and their participation in Greek cultural/religious celebrations and events.  Figure 3 for the frequencies of the responses.
When asked to self-rate their English language skills, the respondents provided answers for the four skills that depicted an equal distribution of responses as the vast majority feel they have mastered them. Writing (1.9% intermediate, 7.4% advanced, and 90.7% native-like) and reading (1.9% intermediate, 5.6% advanced, and 92.6% native-like) were rated lower than listening and speaking (9.3% advanced and 90.7% native-like). See Figure 4 for the frequencies of the responses.
The acquisition of the language skills was further investigated by asking the respondents to rate how difficult it is for them to read numerous genres. The genres that most respondents find difficult to read are academic/ technical papers, poetry, novels, theatrical plays, non-fiction, and textbooks, while flyers, dictionaries, letters, and emails were rated as being easy to understand by most of them.
Next, when speaking Greek, the respondents find it relatively easy to very easy to accomplish most of the tasks they were asked about, such as telling a joke, a fairy tale, a story, using polite language, or being rude. Talking about current events or debating an idea show a moderate level of difficulty, while only giving a formal presentation was rated as challenging. Finally, the GHLSs are more confident in their listening skills and judge most of the tasks (eavesdropping, understanding humor, songs, TV, and movies) as relatively easy to very easy to accomplish, except for formal talks and news reports, which have a more equal distribution of answers from difficult to very easy.

Figure 3
Greek language skills

Figure 4
English language skills

Attitudes towards Greek (Qs 27-29)
An investigation of GHLSs' attitudes towards their HL was done with the last set of questions. When asked what language(s) they preferred to speak, the majority preferred their dominant language (English) whereas about one-third of the respondents favored speaking a mixture of both (see Table 6). It was found that the overwhelming majority of participants reported having their family's support in maintaining Greek (98.1%). Similarly, all of them would want to teach their children Greek (100%). Although the scale tips towards English as a preferred language (particularly among the 2nd and overwhelmingly among the third generation), the results showed a very positive attitude toward Greek.

Discussion
The purpose of the present study was to investigate the sociolinguistic profiles of GHSs (RQ1) and to discuss how these may connect with their identities (RQ2) and what they imply about their educational needs when learning Greek as an HL (RQ3).

RQ1: What is the Socio-Linguistic Profile of GHLSs Living in Chicago?
The results of the first set of questions (Questions 4-9) of the GHLBQ allowed for the initial profiling of our participants concerning the HS generation they belonged to and the type of bilingualism they had developed. Thus, the first generation were successive bilinguals in Greek and English, the second generation were either sequential or simultaneous (English dominant) bilinguals, while the third generation were simultaneous, English-dominant bilinguals.
By comparing the participants' responses with their language backgrounds (Questions 10-14), we found a trend with respect to which generation of HSs they belonged to. The first generation prefer and maintain Greek as the language they still feel more proficient in. The second generation (sequential or simultaneous Englishdominant bilinguals), who have been given opportunities to stay in contact with their HL continuously and systematically, prefer to speak a combination of both languages. The third generation (simultaneous Englishdominant bilinguals), whose contact with Greek is mainly through formal education, prefer English. Karatsareas (2019) reported a similar linguistic situation with Cypriot-Greek heritage speakers in London. Also, data presented in this paper confirm previous findings (Cho et al., 2004;Montrul, 2008Montrul, , 2016 showing that there is a shift in dominance from the HL to the majority language in the transition from the first to the second generation and a possible loss of the HL by the third generation. Our study also documents this dramatic decline. It occurs particularly after the age of 18 as a result of interrupted formal schooling in Greek but also because of language preferences only in restricted contexts (family meetings, conversations with friends, socializing, etc.). As our results do not confirm an extended use of Greek at home or with family, this can help explain the shift in the language preference from Greek to English over time and from generation to generation. The finding of restricted use of Greek at home may be attributed to the fact that parents and children are often pressured to speak only English at home, a practice that "not only contributes to heritage language loss, but also is likely to lead to a less enriched language environment with fewer opportunities for interactions about interesting topics incorporating sophisticated vocabulary, ideas, and concepts, given that many immigrant parents have limited English proficiency" (Lee & Oxelson 2006: 462-463).
The results of Question 15 about formal instruction in Greek indicated that GHSs who have never attended day or afternoon/Saturday schools belong to the first-generation GHLSs who had studied in Greece before moving to the United States. From our results, it can be concluded that GHLSs systematically receive formal education in their HL. However, when interviewed, GHSs commented that when they were younger, they were forced by their parents to attend such schools, and as a result they lacked motivation to learn Greek, while later they regretted doing so.
When it comes to the quantity and variety of written Greek they have been exposed to since their early years (Questions 16-22), it was found that it depends on parental involvement, whose importance has been documented in related studies (Li, 2006). In other words, it is the close family that provides both spoken and written contact with the HL and the quality of linguistic input is a determinant of how proficient a HS may become. It was also found that accessing the internet in Greek is quite restricted. This finding confirms Carreira and Kagan (2011), who claimed that most HSs are not electronically literate in their HL as they use the internet in English rather than in their HL. Another interesting finding was that the participants did not have reading experiences in Greek outside school. This finding highlights the fact that HSs' knowledge of Greek is not supported by extracurricular reading that could improve their academic vocabulary and writing skills in Greek.

PROFILES, IDENTIFIES AND NEEDS OF GREEK HERITAGE SPEAKERS
This finding is in line with Carreira and Kagan (2011), who found that most HL learners who had acquired English after acquiring their HL did not have much exposure to their HL outside the home.
GHSs' self-rated proficiency was also investigated (Questions 23-26). It can be argued that self-assessed proficiency as an approach to measuring the level of language acquisition has its drawbacks, in that it requires linguistic and metalinguistic awareness on the part of HSs. Moreover, speakers tend to assess their knowledge as lower or higher, showing an inverse correlation between a person's self-assessment and their actual proficiency (see Polinsky, 2011). However, studies have shown that HSs' self-assessment correlates very well with independent measures of proficiency (Montrul et al. 2010). Furthermore, asking the participants to selfrate their proficiency in English helped us to further validate our method and confirm the claim that HSs are more proficient in the language of the host country, particularly those belonging to the second and third generations. Thus, in light of the previous findings, we hypothesized that the sample would report a higher self-rated proficiency in listening and speaking and lower in reading and writing, which was confirmed. This finding is in line with Valdés (2000) and Carreira and Kagan (2011), who also found that HSs could understand and speak the HL better than they could read and write it. Actually, the reading and writing abilities of HSs depend on the amount of formal instruction that HSs have received in their HL. However, a different pattern was found when they were asked to self-rate their proficiency in English, where they did not differentiate much between the four skills in English. This finding supports the generally expressed views that HSs are by definition dominant in the host language (Montrul, 2016;Polinsky, 2011).
To further investigate the acquisition of language skills in HG, the respondents were asked to rate how difficult it is for them to read in various genres. It was found that academic/technical papers, poetry, novels, theatrical plays, non-fiction, and textbooks were rather difficult for them. This finding agrees with previous research (Leeman & Serafini, 2016), which showed that HSs are not acquainted with the registers associated with academic or professional settings. Sánchez-Muñoz (2016) attributed these drawbacks in academic language to different social contexts.
It is obvious that the development of language skills observed in our study follow the order of the L1 acquisition process of a child naturally learning the language of the home. Namely, the participants rate their linguistic abilities regressively starting from listening, speaking, reading, to finally writing. Moreover, the phenomenon of the dominant language overtaking the HL (Campbell & Rosenthal, 2000;Polinsky & Kagan, 2007) is documented by very high self-rating skills in English as opposed to Greek. Similarly, oral production (in particular tasks in English) poses no problem, while more limited contact with spoken modalities of Greek lead to lower self-rated speaking ability.
Questions 27-29 investigated GHSs' attitudes towards Greek, which proved to be highly positive. Studies have shown that many parents view HL maintenance as a way to preserve their cultural identity and wish for their children to maintain it (Cho et al., 1997;Guardado, 2010;Hashimoto & Lee, 2011;Lao, 2004;Li-Yuan & Larke, 2008;Wong Fillmore, 2000). Others have suggested that parental attitudes are strong indicators of children's success with HL maintenance (Garcia, 2003;Lao, 2004;Wong Fillmore, 2000). Further findings point to positive parental attitudes being directly associated with more successful maintenance (Hashimoto & Lee, 2011). Our results also agree with Polinsky (2011) and Karatsareas (2018) who report a similar positive stance among their participants towards their HL. Carreira and Kagan (2011: 48) found that most of their students had "largely positive feelings and experiences" with their HL. In our study, Greek is viewed as a symbol of GHSs' ethnolinguistic identity, culture, and history that has to be preserved and there is a subsequent strong desire, especially from the first and second generation, to maintain it.

RQ2: What are the GHLSs' Defining Elements of their Greek Identity?
Data from the GHLC showed that the two most salient indicators of the Greek identity are the Greek language and Greek orthodox religion. First-generation HSs demonstrate high proficiency in Greek and claim they are Greek because they were born and raised in Greece. They relate to Greek more than to American culture, are active members of the Greek community of Chicago, and go to local Greek-orthodox churches every week (there are 26 of them in the Chicago area). They also actively participate in events organized by local Greek communities and follow all Greek customs and traditions (e.g. Easter, Christmas, etc.). They prefer Greek food, wish to preserve the Greek identity of their children by sending them to community-based schools, feel nostalgic about what they left behind when they immigrated, and have a sense of belonging to and love of Greece. As a result, they visit relatives in Greece frequently and some of them say that they hope to return to Greece when they retire. For most of them, cultural heritage is considered as a form of support in their life in the host country.
On the other hand, there is a variation in second and third-generation HSs. The more their parents have engaged them in cultural and social activities associated with the Greek community and culture, the higher sense of appreciation for Greek language and culture the children have. In such cases, second-generation HSs combine Greek and American culture and adopt a bicultural Greek-American identity as can be seen from the extracts from interview 1000 from the GHLC in (1a) and (b)  It is important to highlight here that, since language proficiency in the HL is low for third-generation HSs, the Greek orthodox religion seems to be the defining element of Greek identity for this generation. This finding supports Lee's (2002: 120) claim that "the loss of one marker does not automatically entail the loss of cultural identity and secondly, the loss of one or more identity markers tends to reinforce the remaining identity markers."

PROFILES, IDENTIFIES AND NEEDS OF GREEK HERITAGE SPEAKERS
Addressing the educational needs of GHLSs, particularly the second and third generations, is of the utmost importance for the future of Greek as an HL. The future of any HL is at stake unless a systematic approach to the design of educational material is taken. However, a first step in deciding the content and form of more linguistically and culturally responsive educational resources is to identify the socio-linguistic needs of HL learners. Their motivation to learn/maintain their HL as a part of their identity must also be addressed, since designing an HL program without considering their individual proficiencies and needs would compromise the success of the program (Beaudrie, 2016).
The thorough consideration of the language proficiency levels and instructional needs of GHSs, as they appeared in their sociolinguistic profiles presented in this study, suggest that in order to motivate students, the GHL program should include interesting, inspiring, and flexible activities that prioritize systematic contact with Greece. Moreover, utilizing the COVID-19 pandemic experience with remote learning methods gained by all parties involved (students, parents, and educators), e.g. video conferencing with peers and/or teachers from Greece can help prevent the sudden drop in exposure to the HL. The content of such a program ought to depict which tasks HL learners are capable of completing independently and which (formal talks, etc.) could be achieved only with targeted assistance/scaffolding, and should also include teaching of the less developed language skills of reading and writing by using Open Educational Resources (OER) (e.g. online tasks, textbooks, images, quizzes, videos, podcasts, etc.) that address the text types (genres) that Greek HL learners have not been sufficiently exposed to. Content should also enhance cultural and linguistic diversity, foster sustainability through reuse, and address the needs of policymakers and educators (Bijlsma et al., 2016) with OER.
Furthermore, through the program, students need to build HL learning strategies for becoming effective HL leaners while parents need to be trained to effectively motivate their children, particularly the third generation, to stay in touch with Greek culture by relating it to their particular age groups and interests and not relying solely on the traditional elements brought down by previous generations. Finally, since teachers often feel frustrated and inadequate when teaching HL learners, because they essentially use L2 or L1 teaching methods that are not appropriate for this specific target group, they should be offered systematic, additional training in HL teaching and the use of OERs in order to develop the skills and knowledge needed for teaching this population.

Conclusion
The present study has furthered the research on HSs by offering data for a less-studied HL and providing the socio-linguistic profiles and identities of an uninvestigated community of HSs: Greeks living in Chicago. By comparing the results of studies on various HLs with the ones presented here, the authors hope to have revealed some underlying dimensions important to the fate of heritage languages and shed light on questions such as how socio-linguistic profiles and identity issues can help tailor educational programs for HSs. This researchbased knowledge can be employed in addressing the academic needs of GHLSs through educational programs. One drawback of the study is that identification with Greek culture, ethnic attachment, and the practice of Greek traditions were investigated only through a structured interview, as such questions were not included in the GHLBQ. Assessing the above-mentioned characteristics using a survey, followed by an interview, would help verify respondents' answers. Finally, further research should investigate the relationship between cultural identity and HL maintenance.