Introduction

Recent data indicate that there is a large population of Latinos in colleges around the U.S. Between the years 2005 and 2015, Hispanic enrollment between the ages of 18 and 24 in undergraduate programs increased from 25% to 37%. As a whole, the Hispanic population enrollment in higher education institutions doubled from 1.4 million in 2000 to 3 million in 2014 (^{NCES, 2017}). Even though this student population continues to grow, there are few higher education institutions that offer language courses specifically developed for Spanish Heritage Learners (HLs). Some researchers have reported foreboding statistics about these programs that are not promising in regards to addressing the language education needs of this student population. For example, ^{Beaudrie (2012)} reported that merely 40% of Spanish programs in 422 universities offered courses for this population of learners.

Although the number of classes for Spanish HLs has increased, the majority of them are developed for learners with high oral proficiency in Spanish who have not received formal instruction in Spanish, thus ignoring a large number of HLs that are receptive bilinguals. Often characterized by low proficiency skills in writing and speaking, these learners would like to increase their linguistic abilities and fluency, as well as to develop their identity as heritage speakers and their connections to the heritage culture (^{Beaudrie & Ducar, 2005}).

It is the case that these students are usually motivated to learn Spanish, so they frequently enroll in Spanish courses for beginners designed for second language (L2) learners (i.e., mixed classes) with different configurations: mixed classes with few HLs, a balanced number of HLs and L2 learners, or a large number of HLs (^{Carreira, 2014}). Regardless of the configuration, mixed classes have been deemed as not appropriate for HLs because they have different linguistic and affective needs as well as a desire to connect with the heritage culture that the L2 students do not have (Carreira, 2014; ^{Felix, 2009}; ^{Potowski, 2002}). In spite of having HL and Spanish L2 students with different educational needs in mixed classes, some researchers argue that mixed classes could still help the receptive bilingual HL to develop his or her Spanish linguistic skills (^{Beaudrie, 2009}; ^{Carreira, 2014}) if these courses are centered on topics and experiences related to the Hispanic population in the United States, focus on linguistic forms useful for the HL as well as the L2 learners, implement a macro-based approach in which instruction begins with meanings associated with authentic materials first rather than grammatical forms, and uses differentiated instruction. Although there maybe benefits on the interactions between HL and L2 learners or differentiated instruction in mixed Spanish courses, it is worth noting that HLs linguistic practices and needs should be validated on their own rather than supplanted or supplemented with a more standardized variety of Spanish (^{Rosa, 2019}). Teachers and their practices in mixed classes could perpetuate the perception of HLs as a deficient linguistic group in Spanish, so it is important to be aware of how we are positioning these students in our instructional context.

In order to understand the nature of instruction in mixed classes, a group of studies has explored the students' and teachers' perceptions of different aspects of mixed classes. From the students' point of view, studies like the ones by ^{Bowles (2011)}, Bowles and ^{Montrul (2014)}, ^{Burgo (2019)}, ^{Campanaro (2013)}, ^{Dones-Herrera (2015)} and ^{Edstrom (2007)} report that HLs have positive attitudes toward working in groups with L2 learners in which they can learn from L2 learners and help them with their language concerns. These kinds of interactions position HLs as language experts, which may contribute to an increase in their linguistic confidence (^{Blake & Zyzik, 2013}). Other researchers, however, have found that the Spanish L2 learners could feel intimidated when working with HLs, and some HLs could have the perception that they usually do most of the work in group activities (^{Leeman & Serafini, 2017}). In addition, ^{Burgo (2019)}, ^{Felix (2009)} and ^{Potowski (2002)} noted that HLs could also feel marginalized by their L2 classmates when they feel that the Spanish L2 learners have better control of grammar and writing skills, and embarrassed when instructors and L2 learners have higher expectations for them in regards to their linguistic level, causing them to worry about not being able to meet those expectations.

From the Spanish instructors' perspective, mixed classes pose different challenges on how to address the needs of both L2 learners and HLs. College instructors in studies by ^{Bello (2014)} and ^{Campanaro (2013)} believe that having HL and L2 learners is a positive experience because the L2 learners can be exposed to different Spanish accents and the L2 culture, and HLs can help the L2 learners. However, they recognized that the L2 learners could feel intimidated because they perceive the HLs as more advanced and better at speaking. Moreover, ^{Lacorte and Canabal (2005)} pointed out that college language instructors should be aware of the effect of their own ideology, identities, attitudes and beliefs about FL teaching and HLs regarding the dynamics between the HLs and their instructors and peers.

At the high school level, teachers surveyed in Boyd (2002) and ^{Russell and Kuriscak (2015)} also believe that there may be advantages to the interactions between L2 learners and HLs in mixed classes, but pointed out that the main issues of working with mixed classes include linguistic issues related to the knowledge and use of language (e.g., dialectal variation, spelling and writing skills), social and identity issues (e.g., assuming the HLs know Spanish already), and curricular issues on how to integrate strategies, materials, and assessments that support HLs with varying levels of Spanish language proficiency. Furthermore, ^{Randolph (2017)} added that high school Spanish teachers may interact differently with L2 learners vs. HLs, using advanced Spanish and establishing linguistic authority with HLs while using teacher talk with L2 learners.

Although the research reviewed above has provided hints on the possible roles HLs play in mixed classes, no study has explored in depth the specific roles as assigned by teachers, especially in beginner classes and how they support the heritage language instruction goals and meet their needs for cultural connection at this level. ^{Asención-Delaney (2019)}, after analyzing the annotations for heritage speakers in introductory Spanish textbooks and observing that certain roles (i.e., linguistic, pragmatic and cultural informants, linguistic model, and input provider) were suggested in these annotations, calls for the need to validate these suggestions by collecting data from teachers and classroom observation. Responding to this need, and in order to contribute to a better understanding of the role of HLs in college beginner mixed Spanish courses, this study seeks to validate the information about the roles assigned to HLs inferred from previous research by collecting direct data from teachers to answer the following research questions: a) what roles do instructors assign to HLs in their beginner classes? and b) how do the observed roles support the HL language instruction goals and meet their needs for cultural connection?

Methods

This section presents information about the participants of the study, the variables (i.e., the roles analyzed), and the procedures used for data collection and analysis.

Results

This section presents the findings for the different roles assigned to the HLs by their Spanish instructors in beginner Spanish mixed classes. The findings present quantitative and qualitative evidence collected from the three sources of information used: the questionnaire, interviews, and class observations. First, the results from the questionnaire are presented in Table 2 and then compared with the qualitative evidence collected from the interviews and class observations for each role.

As observed in Table 2, the questionnaire data pointed to the assignment of the following roles (i.e., Average at 2.5 or beyond): Linguistic model; linguistic, cultural and pragmatic informant; and tutor/helper. In order to have a clear picture of the roles assigned to HLs by their instructors, the questionnaire information was further analyzed by role and compared with qualitative evidence from the interviews and class observations.

Five instructors' comments revealed they assigned this role, especially in regards to vocabulary. For instance, Instructor 19 said: "I may ask them if they say something differently." This role was also identified twice during the class observations. In the first instance, Instructor 20 asked a HL about a grammatical form: "¿cuándo usas ir a y cuándo usas elfuturo? - tr. When do you use going to vs. future?" In the second instance, Instructor 13 asked a HL how a Spanish word was said in his heritage variety: "¿cómo se dice helado en México? - tr. How do you say ice cream in Mexico?"

As to how instructors perceived whether these roles support or inhibit the HL language instruction goals and meet their needs for cultural connection, comments from the interviews offered some insights into the matter. For example, instructors 5 and 20 commented that, depending on their proficiency level, HLs liked helping others students that are not HLs. Other instructors (6 and 9) believed that HLs could reinforce or improve their Spanish by participating and helping other students in the class. Instructor 8 highlighted the fact that HLs liked to contribute to the class because they felt proud about their heritage and they wanted to share it with others.

Although some instructors commented on the benefits of assigning HLs these roles in class, some others recognize that HLs could feel uncomfortable or ashamed to say things in Spanish because they did not feel prepared (Instructors 10 and 11). To avoid these situations, some instructors (e.g., Instructor 8) clarified that they never asked HLs to contribute with specific information to the class, but they let the HLs volunteer linguistic and cultural information they wanted to share with their classmates. Instructor 16 explained that, in some cases, HLs could not be asked about grammar because they did not have the necessary metalinguistic knowledge about Spanish to explain any rules.

Discussion

Findings from interviews, class observations and questionnaires indicate that teachers assign multiple roles to the Spanish HLs in beginner mixed classes at the college level. The three sources of information show that teachers mostly ask their HLs to be linguistic and cultural informants in class. Teachers usually ask HLs to provide vocabulary, expressions or some linguistic forms related to the Spanish variety that they speak at home or in their community. The assignment of this role could have a positive impact on HLs' motivation towards their use of the heritage language because their language variety is being validated in the class and used as a springboard to learn more formal or academic language (^{Potowski & Lynch, 2014}).

Being a cultural informant could also contribute to having more active HLs in a mixed class while they are fulfilling their needs to connect with their heritage culture. Although mixed classes are not usually recommended for HLs, ^{Carreira (2014)} suggests the implementation of an approach to mixed classes in which topics related to HLs' bilingual and bicultural identity are discussed by incorporating the cultural and linguistic expertise that they already have.

It is important to mention that some teachers may hesitate to ask HLs about grammar rules given their lack of metalinguistic knowledge, linguistic confidence or low Spanish proficiency. In some cases, these students have never taken a formal Spanish class, but rather relied only on the Spanish exposure they have had at home. As a consequence, they are not able to talk about the language or explain how different grammar aspects work. From the affective point of view, HLs could feel that being assigned the role of linguistic informant can create false expectations about their Spanish proficiency. ^{Potowski (2002)}, for example, pointed out that in her study, some HLs felt uncomfortable when they were asked to provide linguistic or cultural information in class and that they could not perform this role. Similarly, in ^{Felix's study (2009)}, the HLs felt that they did not belong in the mixed class. Specially, after the teacher created false expectations about them being competent speakers of Spanish. In consequence, their L2 classmates might have thought that they should not be in the class. Their classmates could have also thought that the HLs were taking the class to get a high grade without much effort.

Other roles that are assigned to HLs are those of being a linguistic model and a pragmatic informant. For these two roles, only two sources of information provided evidence of HLs actually being asked to model language, especially pronunciation, for the class, or provide information about the appropriate use of certain linguistic forms in their Spanish context of use. Both roles may assume that the HLs have advanced productive skills in the language when it may not be the case for these learners. ^{Valdés (2001)} explains that the bilingual continuum among the HLs is diverse, so it is common that in the same class we could find HLs that are only able to understand the oral and written heritage language in familiar contexts (receptive bilinguals) and other HLs with an advanced language competence who need to develop their language skills in more formal or academic contexts. In beginner Spanish mixed classes, we may find a large number of receptive bilinguals, so to assume that they have advanced Spanish proficiency could result in negative consequences for the HLs' linguistic confidence. ^{Potowski (2002)} argues that posing HLs as linguistic models could cause a lowering of their self-esteem as speakers of the heritage language and decrease their motivation to study the language.

The presence of the HLs is also valued by teachers in the mixed classroom because they can play roles such as translators/interpreters, and tutors for other students in class, as found in ^{Felix's study (2009)}. On some occasions, teachers, especially those whose Spanish was their L2, could ask HLs to give the Spanish translation of what they said in English. Other instructors whose L2 is not that strong could request the HL to translate their instructions from Spanish into English to benefit the L2 beginner students. Even though HLs may be used to this role in familiar and personal contexts and have positive attitudes towards translation (^{Gasca, 2017}), it is important to consider that, as with the other roles, the effects on HLs' confidence and motivation towards translation would depend on the level of the HLs' proficiency in the heritage language. It could help HLs increase their linguistic confidence if they are able to perform the role without problems, or it could reinforce the perception of having limitations with their Spanish. This practice can also be misleading in making students, HLs and L2s, believe that anybody speaking two languages can serve as translator when translation involves having not only knowledge of the target language, but also knowledge about the text-type, the source language, the real-world, and the linguistic differences between the two languages (^{Bell, 1991}).

As it has been observed in previous studies (^{Belpoliti, 2017}; ^{Bowels & Montrul, 2014}; ^{Campanaro, 2013}; ^{Dones-Herrera, 2015}; ^{Felix, 2009}; ^{Russel & Kuriscak, 2015}), some evidence showed that HLs in mixed classes are assigned the role of tutor/helper. ^{Potowski (2002)} suggests that HLs may feel at a disadvantage in comparison with the L2 students because they do not have the same knowledge of the grammar rules that the L2 learners have. Taking this situation into account, HLs may feel that they are unprepared to perform the tutor role in classes that are heavily focused on grammar rules or in which language use is explained with linguistic grammar terminology. On the other hand, HLs may feel empowered when helping other students with certain words or completing activities in the classroom (^{Felix, 2009}). Also, it may be the case that tutoring could provide opportunities to encourage interactions between HLs and L2 learners that support the growth of not only L2 learners, but also the HLs, as observed in advanced mixed language classes (^{Edstrom, 2007}).

In this study, no evidence was found for the role of input provider. It does not seem that teachers at beginning levels ask the HLs to produce oral or written Spanish so that L2 students can analyze the language or answer questions about the content of the HLs' messages. Although this role was suggested by the annotations of HLs in introductory Spanish textbooks in ^{Asención-Delaney's study (2019)}, the actual Spanish instructors in this study did not consider the HLs as a source of Spanish language in the mixed classes. It seems that the instructors in this study do not believe that the comprehension skills of the L2 learners could directly benefit as a result of their interactions with the HLs, as suggested in other studies in advanced mixed classes, such as the one by ^{Edstrom (2007)}.

Conclusions

Spanish instructors seem to assign multiple roles to HLs in beginning mixed classes. HLs could be asked to provide information about the language variety they speak, for example, alternative ways of saying a word or expression, and the cultural practices of their heritage groups. These roles could help validate the HLs' language varieties in the class and pose them as experts in the classroom. Other roles that may be assigned include a tutor, translator, pragmatic informant, and linguistic model. No evidence was found to support the belief that teachers assign HLs the role of input provider in the mixed classroom. It is worth mentioning that depending on the HLs' proficiency in the heritage language, these roles could support or inhibit students' goals at pursuing further education in their heritage language.

These findings have particular implications for language teacher training and professional development. In preparation for dealing with the complexities involved in teaching mixed language classes, some researchers and practitioners suggest the implementation of a new kind of language teaching preparation program that targets HL education either in mixed classes or in HL classes. This training can take the form of another teaching methods course, modules, or workshops for in-service teachers addressing the needs of this learner population in Spanish L2 classes (^{Belpoliti, 2017}; ^{George & Salgado-Robles, 2018}). ^{Lacorte (2016)} suggests training opportunities in which teachers could develop effective skills, dispositions and attitudes such as the understanding of the historical, cultural, sociolinguistic, and academic backgrounds of HLs, awareness about the distribution of language proficiency across modalities and skills among HLs, pedagogical strategies to encourage collaboration among HL and L2 students, classroom management strategies in mixed classes, and awareness of beliefs and attitudes regarding HLs and their varieties.

Finally, although this study provides initial insights into the roles of HLs in beginner mixed classes, it did not seek to judge teachers' motivation when assigning these roles to their HLs in their classes. Factors affecting teaching practices could be rooted in teachers' personal characteristics such as HL proficiency, teaching experience, and professional training. Therefore, more research is needed to understand teachers' cognition and motivations in relation to the roles they assign to HL students in beginner mixed classes. In addition, due to limitations on the number of instructors surveyed/observed and the education context researched, it is not possible to generalize these findings to all instructors in beginner Spanish mixed classes at the elementary, secondary, and tertiary levels. It is also important to mention that the impact that these role assignments may have on this student population should be further investigated from the students' point of view in order to make teachers aware of the complexities reflected in HLs' diverse linguistic and cultural needs, goals, and affective responses to heritage language instruction in mixed language classes.